ORGANIZATIONAL PSYCHOLOGY - UPB

ORGANIZATIONALPSYCHOLOGYA SCIENTIST-PRACTITIONER APPROACH

Second Edition

Steve M. Jex and Thomas W. Britt

JOHN WILEY & SONS, INC.

ORGANIZATIONALPSYCHOLOGY

ORGANIZATIONALPSYCHOLOGYA SCIENTIST-PRACTITIONER APPROACH

Second Edition

Steve M. Jex and Thomas W. Britt

JOHN WILEY & SONS, INC.

Copyright # 2008 by John Wiley & Sons, Inc. All rights reserved.

Published by John Wiley & Sons, Inc., Hoboken, New Jersey.

Published simultaneously in Canada.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic,

mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United

States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the

appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400,

fax (978) 646-8600, or on the Web at www.copyright.com. Requests to the Publisher for permission should be addressed to

the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008,

or online at http://www.wiley.com/go/permissions.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book,

they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and

specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or

extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable

for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable

for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or

other damages.

This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold

with the understanding that the publisher is not engaged in rendering professional services. If legal, accounting, medical,

psychological or any other expert assistance is required, the services of a competent professional person should be sought.

Designations used by companies to distinguish their products are often claimed as trademarks. In all instances where John

Wiley& Sons, Inc. is aware of a claim, the product names appear in initial capital or all capital letters. Readers, however,

should contact the appropriate companies for more complete information regarding trademarks and registration.

For general information on our other products and services please contact our Customer Care Department within the United

States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002.

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in

electronic books. For more information about Wiley products, visit our Web site at www.wiley.com.

Library of Congress Cataloging-in-Publication Data

Jex, Steve M.

Organizational psychology: a scientist-practitioner approach.—2nd ed. / by Steve M. Jex,

Thomas W. Britt.

p. cm.

Includes bibliographical references and index.

ISBN 978-0-470-10976-2 (cloth)

1. Psychology, Industrial. I. Britt, Thomas W., 1966- II. Title. III. Title: Psychology : a scientist

practitioner approach.

HF5548.8.J49 2008

158.7—dc22

2007021315

Printed in the United States of America.

10 9 8 7 6 5 4 3 2 1

www.wiley.com

Contents

Preface xi

CHAPTER 1

Introduction to Organizational Psychology 1

What Is Organizational Psychology? 2

Organizational Psychology in Context 3

The Scientist-Practitioner Approach 5

Historical Influences in Organizational Psychology 9

The Chapter Sequence 17

Chapter Summary 18

Suggested Additional Readings 20

CHAPTER 2

Research Methods and Statistics 21

Methods of Data Collection 22

Special Issues in Data Collection 34

Statistical Methods in Organizational Psychology 41

Special Issues in Statistical Analysis 48

Chapter Summary 55


CHAPTER 3

Attraction and Socialization 59

The Recruitment Process: An Organizational Perspective 60

The Recruitment Process: The Applicant’s Perspective 63

Organizational Socialization 67

v

The Impact of Diversity on Organizational Socialization 90

Chapter Summary 92


CHAPTER 4

Productive Behavior in Organizations 95

Defining Productive Behavior 96

Job Performance 96

Measurement of Job Performance 103

Determinants of Job Performance 109

Organizational Citizenship Behavior 116

Innovation in Organizations 122

Chapter Summary 128


CHAPTER 5

Job Satisfaction and Organizational Commitment 131

Job Satisfaction 131

Organizational Commitment 152

Chapter Summary 163


CHAPTER 6

Counterproductive Behavior in Organizations 167

Defining Counterproductive Behavior 167

Ineffective Job Performance 169

Employee Absenteeism 175

Employee Turnover 179

Less Common Forms of Counterproductive Behavior 186

Chapter Summary 195


vi Contents

CHAPTER 7

Occupational Stress and Employee Health 199

A Brief History 200

Approaches and Terminology 201

Occupational Stress Terminology 202

Occupational Stress Models 205

Workplace Stressors 210

Reducing the Impact of Workplace Stressors 221

Cross-Cultural Occupational Stress Research 226

Generalizability 226

Stressors Experienced 227

Coping with Stress 228

Chapter Summary 229


CHAPTER 8

Theories of Motivation 233

Defining Motivation and Theoretical Approaches 234

Need-Based Theories of Motivation 235

Cognitive Process Theories of Motivation 239

The Behavioral Approach to Motivation 250

Self-Determination Theory 254

Job-Based Theories of Motivation 256

The Practical Value of Motivation Theories 264

Chapter Summary 265


CHAPTER 9

Organizational Applications of Motivation Theory 269

Some Basic Assumptions 269

Behaviors Organizations Attempt to Influence 270

Organizational Reward Systems 272

Contents vii

Motivation through the Design of Work 287

Organizational Disciplinary Procedures 298

Chapter Summary 300


CHAPTER 10

Leadership and Influence Processes 303

Defining Leadership 303

General Approaches to Leadership 306

Modern Theories of Leadership 311

Power and Influence in Organizations 326

Chapter Summary 335


CHAPTER 11

Group Dynamics and Behavior 339

Why Do People Join Groups? 340

Defining Characteristics of Groups 341

Group Structure 343

Stages of Group Development 349

Intergroup Dynamics and Behavior 355

Types of Intergroup Interactions 355

Predictors of Intergroup Interaction Patterns 358

Intergroup Conflict 362

Improving the Quality of Intergroup Relations 366

Chapter Summary 371


CHAPTER 12

Team Effectiveness 375

Defining Team Effectiveness 376

Models of Team Effectiveness 378

Determinants of Team Effectiveness 389

viii Contents

Behavioral Processes 396

Enhancing the Effectiveness of Teams 402

The Future of Teams in Organizations 406

Chapter Summary 407


CHAPTER 13

Organizational Theory and Design 411

What Is an Organizational Theory? 412

Major Organizational Theories 414

Determinants of Organizational Design 421

Recent Innovations in Organizational Design 432

Research on Organizational Design 436

The Future of Organizational Design 437

Chapter Summary 438


CHAPTER 14

Organizational Culture 441

Defining Organizational Culture 442

Models of Organizational Culture 446

Manifestations of Organizational Culture 447

The Development of Organizational Culture 454

Measuring Organizational Culture 456

Changing Organizational Culture 458

The Impact of Organizational Culture 464

Chapter Summary 468


CHAPTER 15

Organizational Change and Development 473

What Is Organizational Development and Why Is It Used? 473

A Brief History of Organizational Development 475

Contents ix

The Theory Base of Organizational Development 477

Organizational-Change Interventions 486

Conditions Necessary for Successful Organizational Change 497

Evaluation of Organizational-Development Programs 502

Special Issues in Client–Consultant Relationships 508

Chapter Summary 511


References 515

Author Index 587

Subject Index 611

x Contents

Preface

Organizations are complex socialsystems that sometimes performwell and sometimes fail miserably.Organizational psychology is asubfield within the larger domain

of industrial/organizational psychology thatseeks to facilitate a greater understanding ofsocial processes in organizations. Organiza-tional psychologists also seek to use theseinsights to enhance the effectiveness oforganizations—a goal that is potentially ben-eficial to all.

This book is designed to provide stu-dents with a thorough overview of both thescience and practice of organizational psy-chology. It was originally written to serve asthe primary text for a course in organiza-tional psychology (graduate, or upper-levelundergraduate), but could also meet theneeds of an organizational behavior courseas well. Because this book will be used atthe graduate level, we’ve invested consider-able effort to provide a solid research base inall of the chapters. At the same time, we havealso tried to write the book in a style thatstudents will find enjoyable, accessible, andperhaps, at times, even entertaining.

NEW TO THE SECOND EDITION

The most visible change to the second editionof this book is the addition of Dr. ThomasBritt from Clemson University as a coauthor.I met Tom several years ago during a summer

while working at Walter Reed Army Insti-tute in Washington, DC. Because Tom’s aca-demic training is in social psychology withresearch interests in organizational psychol-ogy, he was really able to improve many ofthe chapters, particularly those dealing withjob attitudes, teams, and leadership. We alsofound that we had very similar writing styles,and perhaps the best part is that we’veworked very well together.

In terms of content, we did not makedrastic changes because the feedback I’vereceived during the past 5 years about thefirst edition has been very positive. As withthe first edition, Chapters 1 through 4 pro-vide an introduction to the field of organi-zational psychology, an examination of themost common research methods used tostudy behavior in organizations, and theprocesses by which employees are socializedinto organizations and finally become pro-ductive members.

Chapters 5 through 8 offer an examina-tion of the processes by which employeesdevelop feelings of satisfaction and com-mitment toward the organization, an explo-ration of counterproductive behaviors thatthey may engage in, how they might cometo view the workplace as stressful, andsome theories of motivation.

Chapters 9 through 12 include an exami-nation of the various methods organizationsuse to influence the behavior of employees,leadership and influence processes, and

xi

group behavior. Readers will note that twochapters are devoted to groups. One isdesigned to provide an overview of basicsocial psychological processes in groups,and the second is focused more specificallyon the factors that influence team effective-ness. Some readers may recall that in the firstedition there was also a chapter dealing withprocesses governing interactions betweenteams in organizations. In the second editionthe material from this chapter is integratedinto the two chapters dealing with groupsand team effectiveness.

Chapters 13 through 15 are focused on‘‘macro’’ or organization-level processes.These include the design of organizations,organizational culture, and organizationalchange and development.

In terms of content, the major changesinvolved updating the substantive materialin the chapters. Since the first edition manynew developments have occurred in the fieldof organizational psychology, and a greatdeal of new research has been conducted.We have worked hard to convey these newdevelopments and summarize importantnew research findings. Another change inthe second edition is a section in eachchapter where we’ve asked a prominentresearcher to describe how he/she becameinterested in a particular topic. We felt thatthese would be interesting to students, andmore importantly, put a ‘‘human face’’ onsome of the names that are cited in the text.

UNIQUE FEATURESOF THE BOOK

In this second edition we’ve been careful toretain the features of the first edition of thebook that we felt made it unique. For exam-ple, we still have a full chapter on researchmethodology and statistics. Furthermore,since the first edition came out, there have

been even more methodological develop-ments in the field—this further validatesthe decision to include such a chapter inthe first edition.

The second edition also covers manytopics that are not traditionally part of orga-nizational psychology such as recruitment,job performance, and compensation. As withthe first edition, this was done largelybecause of the belief that there is consider-able interrelationship between the ‘‘I’’ andthe ‘‘O’’ sides of the broader field of indus-trial/organizational psychology.

A third and final unique feature that hasnot changed is the use of ‘‘Comment’’ boxesthroughout the book. While the content ofmany of these boxes has changed, the moti-vation behind them has stayed the same—toencourage students to think about and discussthe chapter material. We both strongly believethat students learn much more when they arehighly engaged in the material, and do notfeel that reading is a chore. Some of thecomments relate current events, some pro-vide extended commentary on chapter mate-rial, and others are simply designed to helpthe reader get to know the authors a littlebetter.

ACKNOWLEDGMENTS

The authors would first like to thank the manycolleagues and students who have providedfeedback on the first edition since it came outin 2002. Many of you have stopped by to chatabout the book at conferences, and this feed-back has been extremely valuable. While wecannot guarantee that we’ve addressed everycriticism of the first edition, we’ve certainlylistened and made an effort to do so.

We would also like to thank Tisha Rossi,our editor at Wiley, for her help during therevision process. Tisha provided us withsome very valuable editorial feedback, yet

xii Preface

also allowed us a considerable amount ofdiscretion on the content of the book. Wealso greatly appreciate Tisha’s flexibility ondeadlines.

The authors would also especially likethe thank Jason Kain from Bowling GreenState University, and Adam Beeco, LaurenHaack, Genevieve Mendes, Brittany Moree,and Kristen Purvis from Clemson Universityfor all of their help during the revision proc-ess. Jason read through every chapter andmade some very valuable suggestions. Healso did a tremendous job organizing themany new citations that were added to thisedition. The Clemson students helped con-duct literature searches necessary for up-dating some of the chapters and madesuggestions for applied examples.

A NOTE FROM TOM BRITT

I would first like to thank Steve for givingme the opportunity to work with him inthe revision of his textbook. I have learneda lot from the process, and find myselfbecoming even more committed to thefield of organizational psychology. I also

have been fortunate to supervise a numberof graduate students within Clemson’sIndustrial/Organizational psychology doc-toral program, including Heather Odle-Dusseau, Tiffany Greene-Shortridge, Hai-ley Herleman, and Eric McKibben. Thesestudents continue to provide me with con-stant sources of energy and support. Mostimportantly, I would like to thank my twinsons, Noah and Jordan, who continue toremind me of what life is all about.

A NOTE FROM STEVE JEX

I would first like to thank my former col-league Gary Adams for his unwaveringfriendship and support through the years. Iwould like to provide a heartfelt acknowl-edgment to my family. My wife Robin hasbeen a constant source of love and supportfor over 20 years, and there is certainly noway this book could have been completedwithout her. I would also like to thank mytwo sons, Garrett and Travis, for their loveand support. They’re still my two bestfriends, and I greatly appreciate the sacrificesthey made while I was working on the book.

Preface xiii

ORGANIZATIONALPSYCHOLOGY

Chapter One

Introduction toOrganizationalPsychologyT

he behavior of individuals actingas members of formal organiza-tions has a tremendous impacton many aspects of our lives.Everything—the food we eat, the

cars we drive, the houses we live in—depends on the coordinated effort of indi-viduals in organizational settings. Thisimpact, in fact, is so great that we typicallytake it for granted. In most cases, we onlytake notice when the results are either verygood or very bad. For example, we marvel atthe coordinated effort of a professionalsports franchise that is highly successfuland express disdain when corruption occursin a government agency. Most of the time,however, the impact of behavior in formalorganizations goes relatively unnoticed.

Organizational psychology is a field thatutilizes scientific methodology to betterunderstand the behavior of individualsworking in organizational settings. Thisknowledge is also used, in a variety of ways,to help make organizations more effective.Effective organizations are typically moreproductive, often provide higher-qualityservices to customers, and are usually morefinancially successful than less effective orga-nizations. For private organizations, finan-cial success often results in greater jobsecurity for employees, and increased share-holder wealth for investors. For public orga-nizations, such as police departments,municipal governments, and public univer-sities, success means higher-quality servicesand cost savings to taxpayers.

Enhanced organizational effectiveness,and the success that often comes with it,often provides more indirect benefits as well.

Successful organizations provide employ-ment opportunities, which helps to fosterthe economic well-being of society as awhole. Also, in many instances, employeesin successful organizations are more satis-fied and fulfilled in their work than employ-ees in less successful organizations. Thesepositive attitudes may carry over to non-work-related roles such as parent and com-munity member. Consumers also benefitfrom enhanced organizational effectivenessbecause well-managed, efficient organiza-tions are often able to produce productsand provide services at a much lower cost

1

than their less successful competitors. Suchcost savings are often passed on to consum-ers in the form of lower prices. In sum,everyone is a potential winner when organi-zations function effectively. Organizationalpsychology seeks to enhance the effective-ness of organizations through scientificresearch and the application of researchfindings.

WHAT IS ORGANIZATIONALPSYCHOLOGY?

This book is designed to provide studentswith a comprehensive treatment of thescience and practice of organizational psy-chology. In the most general sense, organi-zational psychology is the scientific study ofindividual and group behavior in formalorganizational settings. Katz and Kahn, intheir classic work, The Social Psychology ofOrganizations (1978), stated that the essenceof an organization is patterned humanbehavior. When behavior is patterned, thisimplies that some structure is imposed onthe behavior of individuals. In organizationsthis structure typically comes from thingssuch as job descriptions and organizationalpolicies. Many organizations also have amore general set of values that they wantemployees to abide by. Thus, an organizationcannot exist when people just ‘‘do their ownthing’’ without any awareness of the behaviorof others.

Given Katz and Kahn’s defining charac-teristic of organizations (e.g., patternedbehavior), it is easy to see that there are manyorganizations in this world. A group of fivepeople who regularly play poker on Fridaynights would fit this definition, as would amajor multinational corporation. Therefore,to further define the field of organizationalpsychology, it is important to distinguishbetween formal and informal organizations.

A formal organization is one that exists tofulfill some explicitly stated purpose, andthat purpose is often stated in writing. For-mal organizations also typically exhibit somedegree of continuity over time; that is, theyoften survive far longer than the foundingmembers do. Business organizations obvi-ously exhibit these defining characteristicsof a formal organization, as do many othernonprofit organizations and governmentagencies.

In contrast, an informal organization isone in which the purpose is typically lessexplicit than for a formal organization. Goingback to the previous example of five pokerplayers, these individuals are obviouslyspending time together because they enjoyplaying poker and, in all likelihood, eachother’s company. It is doubtful, though, thatthese reasons for playing poker are formallystated in writing, or even explicitly stated.It is also doubtful (though obviously notimpossible) whether this small group wouldcontinue to exist if three of the five membersmoved to another city or simply lost interestin poker.

The field of organizational psychology isconcerned with the study of formal organi-zations. That is not to say that the formalorganizations of interest to organizationalpsychologists are always businesses orprofit-making organizations (a commonmisconception that we have noticed amongmany of our colleagues trained in otherareas of psychology). Throughout the chap-ters in this book, many studies will bedescribed that have been conducted not onlyin businesses but also in government agen-cies, universities, and nonprofit social serv-ice agencies.

Another point worth noting is that thefocus on formal organizations does not pre-clude the study of informal organizationalprocesses, or even occasionally informal

2 Introduction to Organizational Psychology

groups and organizations. It has been shown,for example, that informal friendship tiesexist in formal organizations, and they haveimportant implications for employees (Rior-dan & Griffeth 1995). In this same vein,processes that occur in informal groupsand organizations may provide researcherswith valuable insights into processes thatoccur in formal organizations. For example,the manner in which a status hierarchydevelops in an informal group such as asports team may help researchers betterunderstand the emergence of leadership informal organizations.

Another point of clarification in the pre-vious definition provided has to do withthe term psychology itself, since organiza-tional psychology is part of this larger field.Psychology is the scientific study of individ-ual human behavior and mental processes.Two things are important to note aboutthis definition. First, organizational psychol-ogists use methods of scientific inquiry toboth study and intervene in organizations.This simply means that organizationalpsychologists use a systematic, data-basedapproach to studying organizational proc-esses and solving organizational problems.The ‘‘data’’ used by organizational psycholo-gists may come in a variety of forms,including survey responses, interviews,observations, and, in some cases, organiza-tional records.

The other important part of this defini-tion is that psychology focuses on individualbehavior. This may seem a bit odd to somereaders, given that a substantial portion ofthis text is devoted to both group and orga-nizational-level processes. What it means isthat regardless of the level at which somephenomenon occurs, psychologists viewindividual behavior as the most importantmediating factor (cf. Porras & Robertson,1992). Thus, to understand the impact of

group and organizational-level variables,we must focus on how they influence, andare influenced by, individual behavior.Groups and organizations don’t behave; peo-ple do. This strong focus on individualbehavior also serves to distinguish organiza-tional psychology from other social sciencedisciplines (e.g., sociology, economics, polit-ical science) that attempt to explain organi-zational processes but are less focused onindividual behavior. It is also one way inwhich organizational psychology differs fromthe closely related field of organizationalbehavior (see Comment 1.1).

ORGANIZATIONALPSYCHOLOGY IN CONTEXT

While organizational psychology representsa legitimate field of study in its own right, it isalso part of the broader field of industrial/organizational (I/O) psychology. I/O psy-chology is defined as the application of themethods and principles of psychology to theworkplace (Spector, 2006). Figure 1.1 pro-vides a comparison of the topics that aretypically of interest to those in the industrialand organizational portions of the field.Notice that the topics listed on the industrialside are those that are typically associatedwith the management of human resourcesin organizations. Contrast these with thetopics on the organizational side, which areassociated with the aim of understandingand predicting behavior within organiza-tional settings.

Given this distinction between the in-dustrial and organizational sides of the field,it is very tempting to polarize into different‘‘camps’’ based on one’s professional inter-ests. Unfortunately, this ‘‘I’’ and ‘‘O’’ dis-tinction underestimates the considerableinterdependence among the topics that con-stitute each of these subfields.

Organizational Psychology in Context 3

To illustrate this point, let’s say a lifeinsurance company decides to develop atest to select people to sell insurance policies.To do so, this organization would likely

conduct some form of job analysis to findout what exactly is involved in selling lifeinsurance policies, develop performance cri-terion measures based on this job analysis,

COMMENT 1.1

ORGANIZATIONAL PSYCHOLOGY VERSUS ORGANIZATIONAL BEHAVIOR:

WHAT’S THE DIFFERENCE?

MANY READERS, PARTICULARLY those who have

received at least a portion of their training in

a university business school, have heard of thefield of organizational behavior. What is the

difference between organizational psychology

and organizational behavior? In all honesty,

these two fields are quite similar—so much so,

in fact, that many faculty who teach organiza-

tional behavior in business schools received

their training in departments of psychology.

Though less common, there have been someinstances where faculty who teach organiza-

tional psychology received their training in

business schools.

Despite the outward similarities, there are

actually subtle differences between organiza-

tional psychology and organizational behavior.

Moorhead and Griffin (1995) define organiza-

tional behavior as ‘‘the study of human behaviorin organizational settings, the interface

between human behavior and the organiza-

tion, and the organization itself’’ (p. 4). If we

focus only on the first part of this definition,

there is no difference between organizational

psychology and organizational behavior. How-

ever, the differences lie in the portion of the

definition stating that organizational behavioris concerned with ‘‘the organization itself.’’

Specifically, the field of organizational behav-

ior is concerned not only with individual

behavior in organizations, but macro-level

processes and variables such as organizational

structure and strategy are viewed as interesting

and worthy of study in their own right.

Organizational psychology is also con-cerned with the impact of macro-level variables

and processes, but only to the extent that such

variables and processes have an impact on

individual behavior. Much of the reason for thisdifference is that organizational behavior

draws from a greater variety of disciplines

than does organizational psychology. While

organizational psychology draws primarily

from various subfields within psychology,

organizational behavior draws from a variety

of discplines including psychology, sociology,

anthropology, economics, and labor relations,to name a few. This greater variety provides

organizational behavior with a somewhat more

eclectic theoritical base than organizational

psychology, although both fields largely study

the same phenomena.

Perhaps the most tangible difference be-

tween organizational psychology behavior

andorganizationalpsychologyisinsalarylevels.Faculty inbusiness schoolswhoteachorganiza-

tional behavior are typically paid significantly

more than faculty who teach organizational

psychology within psychology departments.

This explains why many who are trained in

psychology want to teach organizational behav-

ior in business schools; in fact, a perusal of the

background of faculty at business schools willshow that many have been trained in psychol-

ogy. In recent years, however, the hiring of

psychologists has waned a bit. This is due to

the job market in general, and the fact that

business schools now produce more Ph.D.s

than they did 25 to 30 years ago.

Source: G. Moorhead and R. W. Griffin. (1995). Organi-

zational behavior: Managing people and organizations (4th

ed.). Boston: Houghton Mifflin Company.


develop a selection test to measure thingsthat are thought to be predictive of perfor-mance, and ultimately conduct a study toinvestigate whether performance on theselection test is correlated with the per-formance criterion measure (Cascio, 1998).Because all of these are ‘‘I’’ activities, whatrelevance does the ‘‘O’’ side of the field havefor the life insurance company in this exam-ple? On first glance, it would appear to bevery little. However, if you think about it,organizational topics are highly relevant. Forexample, after these life insurance agents areselected, they must be socialized into theculture of the specific agency in which theywill be working, as well as the broader com-pany culture (Bowen, Ledford, & Nathan,1991; Kristof, 1996). Also, demands of lifeinsurance sales may necessitate the hiring ofindividuals who will cope well with thesedemands (Jex, 1998). Thus, the organizationneeds to understand the unique stressorsthat are associated with this job, as well asthe attributes that facilitate coping. As we

will see, socialization and occupational stressare important topics within organizationalpsychology.

This point can also be illustrated bytaking an ‘‘O’’ topic and describing the rele-vance of the ‘‘I’’ side of the field. Let’s saythe U.S. Army is interested in improvingdecision-making and communication proc-esses among the small groups that comprisespecial-forces units. Fortunately, in organiza-tional psychology, there is considerable liter-ature on group effectiveness and processes,and the Army could draw on these sources tohelp guide its efforts (e.g., Guzzo & Shea,1992). Can issues that are relevant to the ‘‘I’’side of the field be ignored? Absolutely not.To be effective, a group must have a certainmix of skills, abilities, and personality traits.Thus, regardless of the team processes thatare taught to these units, care must be takento select the right mix of individuals in thefirst place. It is also unlikely that decision-making processes would improve unlessthese teams receive accurate and timely per-formance feedback. Selection and perform-ance appraisal, of course, are two of the majortopics on the ‘‘I’’ side of the field.

THE SCIENTIST-PRACTITIONERAPPROACH

Organizational psychology should be viewedas a science. In fact, much of the content ofthis book is based on scientific studies ofbehavior in both organizational and labora-tory settings. Organizational psychology,however, is also concerned with the applica-tion of scientific knowledge to enhance theeffectiveness of organizations. The scientist-practitioner model captures this dynamicinteraction between generating scientificknowledge and the application of thatknowledge for some practical purpose. At avery general level, the scientist-practitioner

FIGURE 1.1A Breakdown of Topics Associated with theIndustrial and Organizational Sides of the Field ofI/O Psychology

Industrial Side

Recruitment

Selection

Classification

Compensation

PerformanceAppraisal

Training

Organizational Side

Socialization

Motivation

Occupational Stress

Leadership

Group Performance

OrganizationalDevelopment

Industrial/Organizational Psychology


model states that science and practice are notindependent and, in fact, often feed off eachother.

To illustrate how the scientist-practitionermodel works, let’s say the branch manager ofa bank is frustrated by high turnover amongtellers. Fortunately, this individual may drawon the findings of many scientific investiga-tions of turnover to guide his or her efforts toreduce it (e.g., Griffeth, Hom, & Gaertner,2000). Conversely, scientific investigationsof organizational phenomena are often moti-vated by the practical concerns of organiza-tions. For example, the past decade hasindicated a considerable rise in research onhow organizations can assist employees inbalancing the demands of both work andfamily domains (e.g., Adams, King, & King,1996; Kossek & Ozeki, 1998). Althoughcertainly useful from a purely scientificstandpoint, another important factor moti-vating this research is that organizationswant to avoid losing valuable employeesbecause of difficulties in balancing workand family demands.

Within the general field of I/O psychol-ogy, the scientist-practitioner model hasbecome so important that it serves as theunderlying philosophy for many if not mostgraduate training programs in the field.Graduate training guided by the scientist-practitioner model suggests that, first andforemost, students need the skills necessaryto conduct scientific research. This explainswhy virtually all graduate programs in I/Opsychology require training in statistics,research methodology, and psychologicalmeasurement. The other important implica-tion of the scientist-practitioner model ingraduate training is that students are typi-cally provided with some opportunity,through internships or other field experi-ence, to apply what they have learned in realworld settings (see Comment 1.2).

The scientist-practitioner model is alsoquite relevant to the field of organizationalpsychology, and thus was chosen as theguiding theme for this book. As will becomeevident as readers proceed through thechapters, research by organizational psy-chologists has greatly enhanced our under-standing of behavior in organizations. Forexample, research by organizational psy-chologists has provided valuable insightsinto things such as group effectiveness,socialization of new employees, and goal-setting processes. At the same time, findingsgenerated from scientific research in theseareas have been used to guide interventionsdesigned to help organizations become moreeffective.

The impact of the scientist-practitionermodel also can be seen in the work settingsand activities of those trained in organiza-tional psychology. Many hold academicpositions—typically, in departments of psy-chology or management. The primary jobduties of most academicians are teaching,scientific research, and service to one’s aca-demic department and university. However,many in academia also use their researchskills to help organizations solve a varietyof practical problems. The careers of bothauthors of this text have certainly containedthis blend of science and practice (see Com-ment 1.3).

The training of organizational psycholo-gists who pursue academic careers is notdrastically different from the training of orga-nizational psychologists who pursue nona-cademic careers. Consistent with thescientist-practitioner model, students ingraduate programs in I/O psychology andrelated fields typically receive courseworkin research methodology, statistics, andmeasurement, as well as in specific contentareas (e.g., motivation, leadership). It is alsocommon for all students, regardless of their


career plans, to conduct research and obtainpractical experience.

There are, however, some importantcomponents that future academicians typ-ically need to incorporate into their gradu-ate training. For example, it is importantfor those planning an academic career tobecome involved in research early intheir graduate training. This increases thechances of gaining authorship on journalarticles and conference presentations—something that definitely helps in a com-petitive job market. Research involvementalso facilitates the development of closeworking relationships with faculty. Theserelationships are crucial in learning how to

do research. Another essential componentof the training of future academicians isteaching experience. Regardless of the typeof institution in which one is employed,teaching is a major component of any aca-demic position and all universities arelooking for good teachers. Thus, graduatestudents who obtain significant teachingexperience are much better prepared foracademic positions than those with littleor no experience.

Typical nonacademic employment set-tings for organizational psychologists in-clude business organizations, consultingfirms, nonprofit research institutes, govern-ment agencies and research institutes, and

COMMENT 1.2

TRAINING SCIENTIST-PRACTITIONERS: THE ROLE OF PRACTICAL

EXPERIENCE

MOST GRADUATE PROGRAMS in I/O psychology, as

well as other fields, incorporate some form of

practical experience into their curriculum.This can be accomplished in a variety of ways.

Most programs, for example, encourage stu-

dents to participate in formal internship pro-

grams in corporations and consulting firms.

Typically, internships span between 6 months

and 1 year, and require that students work

under the supervision of an experienced I/O

psychologist. Other less formal ways that stu-dents obtain practical experience include class

projects, working with faculty on research and

consulting projects, and field-based practi-

cum courses.

The major benefit of students participating

in field experiences is that they gain a chance

to put what they’ve learned in their courses

into practice in a real organization. Studentsalso benefit in a more subtle way: They develop

a greater understanding of how the real world

actually works. For example, students working

on field projects are often surprised at how

quickly organizations want things done, as

well as the importance of building positiveinterpersonal relationships with ‘‘clients’’ in

organizations. Many students are also sur-

prised that their methodological and statistical

training comes in quite handy as they work on

these field projects.

Despite the many advantages of practical

experience, there can be some disadvantages of

incorporating it into graduate programs. Theprimary single experience by many doctoral

programs is that, in some cases, students who

take internships never finish their degree.

Other problems that can occur are lack of com-

petent supervision and, in some cases, the pro-

jects organizations assign to students are not

meaningful. Despite these potential disadvan-

tages, carefully monitored practical experienceis usually a valuable component of graduate

training. It is also an excellent way to teach

the scientist-practitioner model to students.


COMMENT 1.3

SCIENCE AND PRACTICE IN OUR OWN CAREERS

STEVE JEX-WHEN I reflect on my own career, the

science-practice theme is very evident. Since

receiving my Ph.D. in industrial/organiza-

tional psychology in 1988, I have carried onan active program of research in the area of

occupational stress. Thus, a good deal of what

I do centers around the science. However, in

addition to scientific activity, I have con-

ducted a number of projects in organizations

that have been designed to solve practical

problems. For example, not long after starting

my first job out of graduate school, I was theassistant investigator on a project conducted

for the U.S. Army Research Institute. This

project involved conducting an organizational

assessment of the recruiting operations

branch of the U.S Army. The Army was inter-

ested in ways that the recruiting branch could

facilitate the training of field recruiters. Since

that first project, I have worked with a numberof organizations conducting applied research

projects and developing training programs.

What have I learned from working with

organizations? Probably most important, I

have developed a great deal of respect for those

who do applied work on a full-time basis.

Applying research findings in organizational

settings is tough work that requires consider-able skill. Another thing I have learned is that

good science has practical value; that is, when

projects in organizations are conducted in a

scientifically rigorous manner, organizations

typically obtain much more useful information

than when they are not. Finally, working in

organizations has really convinced me of the

viability of the scientist-practitioner model.The opportunity to do scientifically meaning-

ful work that has practical value makes the field

of I/O psychology very unique and exciting.

Thomas Britt- THE FURTHER INTO my career I

get, the more I realize the importance of the

scientist-practitioner model. I received my

Ph.D. in social psychology in 1994, and then

immediately started active duty in the U.S.

Army as a research psychologist. I quickly

realized that the Army was not necessarily

interested in the identity regulation of roman-tic partners (the topic of my doctoral disserta-

tion), but was interested in how soldiers could

be motivated to perform well during stressful

military operations. Therefore, I tried to con-

duct applied research ‘‘in the field’’ that met my

own (and journal reviewer’s) standards for

scientific rigor. I ended up having a lot of

fun in the Army conducting research on howthe identity images of soldiers as ‘‘warriors’’ and

‘‘peacekeepers’’ influenced motivation and

health in different types of operations, how

being personally engaged in work could serve

as a buffer against many deployment stressors,

and how soldiers could possibly derive bene-

fits such as increased self-confidence and

appreciation for life as a result of successfullyhandling the rigors of military operations.

Somewhat to my surprise, I also enjoyed

communicating the importance of research

findings to military leaders, and thinking about

the applied relevance of the research I con-

ducted. I found that leaders were much more

likely to take recommendations to heart when

they were backed by data collected using asound research design. I also found that leaders

in applied settings appreciated the utility of a

well-supported theory in making sense of the

findings. Like Steve, I was impressed with how

leaders were really willing to devote the time

and attention necessary to understand the

implications of scientific research for the

well-being and performance of their person-nel. I find myself being guided by the scientist-

practioner even more as I have begun new

programs of research on understanding stres-

sors faced by foreign language analysts who

work for the U.S. government and how positive

motivational states at work can be used to

identify employees who thrive under stress.


even market research firms. While actual jobduties vary widely by setting, many organi-zational psychologists employed in nonaca-demic settings are involved in organizationalchange and development activities. Thismight involve assisting an organization inthe development and implementation of anemployee opinion survey program, design-ing and facilitating the implementation ofteam development activities, or perhapsassisting top management with the strategicplanning process. The other major activity ofthose employed in nonacademic settings isresearch. This is particularly true of thoseemployed in nonprofit research institutes,government research institutes, and marketresearch firms. Given the diversity of thesesettings, it is difficult to pin down the exactnature of the research that is conducted.However, in the most general sense, theseindividuals conduct scientific research thatis designed to have some practical benefitto the organization or even to society ingeneral.

To prepare for a nonacademic career,graduate students need training in many ofthe same areas as those pursuing academiccareers. These include courses in researchmethodology, statistics, measurement, andseveral substantive topical areas. There isone important difference, however: Com-pared to those seeking academic employ-ment, it is more essential for studentsplanning nonacademic careers to obtainpractical experience during their graduatetraining. This experience can often be gainedby assisting faculty with consulting projects,or, in some cases, through formal internshipprograms (see Comment 1.4). Obtainingpractical experience is crucial not onlybecause it enhances a student’s credentials,but because it provides valuable opportuni-ties to apply what has been learned in grad-uate courses.

HISTORICAL INFLUENCESIN ORGANIZATIONALPSYCHOLOGY

The year 1992 marked the hundredth anni-versary of the field of psychology. To markthis centennial, much was written about thehistory of industrial/organizational psychol-ogy. This section, therefore, will not providea detailed, comprehensive history of the fieldof organizational psychology. Rather, theintent is to provide a relatively concise sum-mary of some of the people and historicalevents that have shaped the field.

Historical Beginnings

As Katzell and Austin (1992) point out,interest in the behavior of individuals inorganizational settings undoubtedly datesback to ancient times: ‘‘In the organizationalfield, perhaps the earliest recorded consul-tant was the Midianite priest, Jethro, whoadvised his son-in-law, Moses, on how tostaff and organize the ancient Israelites(Exod. 18)’’ (p. 803). Formalized attemptsto study and influence such behavior, how-ever, have a much more recent history.

In order to understand the more recenthistorical roots of organizational psychology,we must first examine the beginnings of thebroader field of industrial/organizationalpsychology. Based on most historical ac-counts of the development of the field of I/O psychology, the industrial side of the fieldwas much quicker to develop than the orga-nizational side. Chronologically, the begin-nings of the field of I/O psychology can betraced to work, during the early part of thetwentieth century, by pioneers such asHugo Munsterberg, Walter Dill Scott, andWalter Bingham (Katzell & Austin, 1992).Most of the work at that time dealt withtopics such as skill acquisition and personnel


selection. Very little work dealing with theorganizational side of the field was con-ducted. Table 1.1 provides a chronologicalsummary of some of the major events thatshaped the development of the field of orga-nizational psychology in the twentieth cen-tury.

Ironically, the beginnings of the orga-nizational side of the field were heavilyinfluenced by the work of several nonpsy-chologists. Perhaps the best known ofthese was Frederick Winslow Taylor,who developed the principles of scientificmanagement (Taylor, 1911). Although formany the term scientific management typi-cally conjures up images of time-and-motion

study, as well as piece-rate compensation, itwas actually much more than that. Scientificmanagement was, to a large extent, a philos-ophy of management, and efficiency andpiece-rate compensation were the most visi-ble manifestations of that philosophy. Whenone looks past these more visible aspectsof scientific management, three underlyingprinciples emerge: (1) those who performwork tasks should be separate from thosewho design work tasks; (2) workers arerational beings, and they will work harderif provided with favorable economic incen-tives; and (3) problems in the workplacecan and should be subjected to empiricalstudy.

COMMENT 1.4

THE INSTITUTE FOR PSYCHOLOGICAL RESEARCH AND APPLICATION (IPRA)

ONE OF THE most important features of the

graduate program in I/O psychology at the

Bowling Green State University is the experi-

ence students receive working on projectsthrough the Institute for Psychological

Research and Application (IPRA). IPRA was

created by the I/O faculty at Bowling Green in

the late 1980s in order to provide graduate

students with the opportunity to apply, in

actual organizational settings and under the

supervision of faculty, what they learn in the

I/O program. A secondary pupose of IPRA is toprovide graduate students with funding to

attend professional conferences.

Typically, local organizations approach

the IPRA director (or some other I/O faculty

member) with some proposed organizational

need that might match the expertise of the I/O

faculty at Bowling Green. Examples of projects

that have been done through IPRA includeemployee opinion surveys, training needs

assessment, customer service satisfaction sur-

veys, and performance appraisal system devel-

opment. After an organization has expressed a

need, a faculty member is sought to serve as a

supervisor on the project. Once a faculty mem-

ber agrees to supervise a project, a meeting is

typically set up with a representative from thatorganization to obtain more concrete informa-

tion about the projects. This is typically fol-

lowed by the submission, to that organization,

of a formal proposal that includes the nature of

the work to be done, the time frame under

which the work will be done, the deliverables

that the organization will receive at the con-

clusion of the project, and an itemized budget.The vast majority of students who graduate

from the I/O program at Bowling Green State

University feel that their work on IPRA projects

was one of the most valuable components of

their education; this is particularly true for

students who end up working for corporations

and consulting firms. Students feel that work

on these projects helps them to sharpen theirtechnical skills, provides valuable opportuni-

ties to apply what they learn in their classes,

and provides a realistic preview of the world of

consulting.


In considering the underlying princi-ples of scientific management describedpreviously, the first principle is certainlycontrary to much of the thinking in the fieldof organizational psychology today. Manyorganizational psychologists, in fact, haverecommended that employees be involvedin decisions impacting the design of theirwork (e.g., Hackman & Oldham, 1980).The second principle, namely that employ-ees will respond to financial incentives, hasactually received considerable support overthe years (Jenkins, Mitra, Gupta, & Shaw,1998; Locke, 1982). Most organizationalpsychologists, however, do not believe thatfinancial incentives will completely com-pensate for extremely dull and repetitivework—something that seems to be anassumption of scientific management. Thethird principle, empirical study, has beenfully embraced by the field of organizationalpsychology and is clearly the one that estab-lishes the link between the two fields. It isalso worth noting that by employing scien-tific methodology to study production-related processes, Taylor was ahead of histime and is considered a pioneer by some.(Most of his studies dealt with cutting sheet

metal.) Unfortunately, despite the impact ofscientific management, many of Taylor’sideas met with a great deal of controversy(see Comment 1.5).

Another early nonpsychologist who con-tributed greatly to the development of or-ganizational psychology was Max Weber.Weber’s academic training was in law andhistory, but his legacy is largely in the field oforganizational design. Weber is best knownfor his development of the notion of bureau-cracy as an organizing principle. The basicidea of a bureaucratic organization is thatemployees know exactly what they are sup-posed to be doing, and the lines of authorityare clearly stated. Another major principleof bureaucracy is that advancement andrewards should be based on merit and noton things such as nepotism or social class.Many principles of bureaucracy are taken forgranted today and are even looked at with abit of disdain, but these ideas were quiteinnovative at the time they were proposedby Weber.

Weber was also a pioneer because hewent beyond merely giving advice aboutorganization and management issues, andsubjected many of his ideas to empirical

TABLE 1.1A Chronological Summary of the Major Historical Influences on the Field of Organizational Psychologyduring the Twentieth Century

Early 1900s Development and growth of Scientific Management (Taylor); beginning of the scientific

study of organizational structure (Weber)

1920s–1930s Hawthorne Studies; growth of unionization; immigration of Kurt Lewin to the United States

1940s–1950s WWII; publication of Vitele’s book Motivation and Morale in Industry; development of the

‘‘Human Relations’’ perspective; Lewin conducts ‘‘action research’’ projects for the Comission

on Community Relations and establishes the Research Center for Group Dynamics at M.I.T.

1960s–1970s U.S. involvement in Vietnam; Division 14 of the APA is changed to ‘‘Industrial/Organizational

Psychology’’; ‘‘multi-level’’ perspective in organizational psychology; increasing attention to

nontraditional topics such as stress, work-family conflict, and retirement.

1980s—1990s Increasing globalization of the economy; changing workforce demographics; increasing reliance

on temporary or contingent employees; redefining the concept of a ‘‘job.’’

2000-Present Advances in communication technology, continued increases in globalization, greater

flexibility in work arrangements, boundaries between ‘‘work’’ and ‘‘non work’’ less clear


investigation. In addition to studying orga-nizational design, Weber wrote extensivelyon organizational topics such as leadership,power, and norms at a time when thesetopics were largely ignored by psychologists.This willingness to study organizationalissues empirically is one of the major defin-ing characteristics of the field of organiza-tional psychology, and thus represents animportant aspect of Weber’s legacy.

The Field Takes Shape

Despite the early work of Taylor and Weber,and others, the vast majority of effort in

industrial psychology in the early twentiethcentury was focused on what were describedearlier as industrial topics. The event thatchanged that—an event many see as thebeginning of organizational psychology—was the Hawthorne studies. The Hawthornestudies, a collaborative effort between theWestern Electric Company and a group ofresearchers from Harvard University, tookplace between 1927 and 1932 (Mayo, 1933;Whitehead, 1935, 1938). The original pur-pose of the Hawthorne studies was to inves-tigate the impact of environmental factors—such as illumination, wage incentives, andrest pauses—on employee productivity.

COMMENT 1.5

ARTHUR KORNHAUSER: ADVOCATE FOR ORGANIZED LABOR

WHILE THE STUDY of labor unions is generally

not considered one of the core areas of indus-

trial/organizational psychology, it has cer-

tainly increased dramatically in the past 25years. Much of the credit for this interest in

labor unions can be traced back to the pio-

neering work of Arthur Kornhauser. Korn-

hauser received his Ph.D. in psychology

from the University of Chicago in 1926, and

remained there as a faculty member until

1943. After a brief period in a research posi-

tion at Columbia University, Kornhauseraccepted a full professorship at Wayne State

University in 1947 and remained there until

his retirement in 1962. Kornhauser died in

1990 at the age of 94.

In a 2003 article published in the Journal of

Applied Psychology, Mike Zickar provides a

fascinating and quite detailed account of the

life and work of Kornhauser, which was basedon a number of interviews with Kornhauser’s

colleagues, family, and former students.

According to Zickar, Kornhauser was unique

in that he was one of the few early industrial/

organizational psychologists who conducted

research on behalf of labor unions and was an

advocate for their interests. Most industrial/

organizational psychologists, both early in the

development of the field and in thepresentday,have been interested primarily in issues that are

important to the management of organiza-

tions—turnover, productivity, and selection,

to name few issues.

Unfortunately, Kornhauser’s work has not

been given the attention of other early figures in

the field of industrial/organizational psychol-

ogy; this is one of the reasons Zickar wrote thearticle. It is probably a reasonable assumption

that Kornhauser’s connection to unions, not a

mainstream topic in the field, was one reason

for this neglect. Other reasons for this neglect,

according to Zickar, were that Kornhauser’s

work was interdisciplinary (he held a joint

appointment in the Institute for Labor and

Industrial Relations at Wayne State) and thathe had only three doctoral students through-

out his career.

Note: From Zickar, M. J. (2003) Remembering Arthur

Kornhauser: Industrial psychology’s advocate for worker

well-being. Journal of Applied Psychology, 88, 363–369.


When one considers the time period inwhich the Hawthorne studies were initiated(early 1920s), it is not surprising that thesetopics were investigated because scientificmanagement was the dominant school ofmanagerial thought at the time.

What made the Hawthorne studies soimportant to the field of organizational psy-chology were the unexpected findings thatcame out of this series of investigations. Per-haps the best known were the findings thatcame from the illumination experiments.Specifically, the Hawthorne researchersfound that productivity increased regard-less of the changes in level of illumination.This became the basis for what is termed theHawthorne effect, or the idea that people willrespond positively to any novel change in thework environment. In modern organiza-tions, a Hawthorne effect might occur whena relatively trivial change is made in a per-son’s job, and that person initially respondsto this change very positively, but the effectdoes not last long.

The significance of the Hawthorne stud-ies, however, goes well beyond simply dem-onstrating a methodological artifact. Forexample, in subsequent studies, Hawthorneresearchers discovered that work groupsestablished and strongly enforced produc-tion norms. The Hawthorne researchers alsofound that employees responded differentlyto different styles of leadership. The overallimplication of the Hawthorne studies, whichlater formed the impetus for organizationalpsychology, was that social factors impactbehavior in organizational settings. This mayseem a rather obvious conclusion today, butwhen considered in the historical context, itwas a very novel and important finding.Focusing only on the specific conclusionspublished by the Hawthorne researchers, aswell as the methodological shortcomings ofthis research (e.g., Bramel & Friend, 1981;

Carey, 1967), misses the much larger impli-cations of this research effort.

During roughly the same time period inwhich the Hawthorne studies took place,another important historical influence onorganizational psychology occurred: union-ization. This is somewhat ironic, consideringthat I/O psychology is often viewed warily byunions (Zickar, 2001), despite the fact thatthere has been cooperation between the two.However, the union movement in the UnitedStates during the 1930s was importantbecause it forced organizations to consider,for the first time, a number of issues that arelargely taken for granted today. For example,organizational topics such as participativedecision making, workplace democracy,quality of work life, and the psychologicalcontract between employees and organiza-tions are rooted, at least to some degree, inthe union movement. Many of these issueswere addressed in collective bargainingagreements in unionized organizations.Many nonunionized organizations wereforced to address these issues due to thethreat of unionization (see Comment 1.5).

During the period of union growth inthe 1930s, another event occurred thatwould prove to be very significant for thedevelopment of the field of organizationalpsychology: Kurt Lewin fled Nazi Germanyand ultimately took a post at the Universityof Iowa Child Welfare Research Station. Bythe time he immigrated to the United States,Lewin was already a prominent social psy-chologist who had a variety of research inter-ests, many of which were relevant to theemerging field of organizational psychology.Lewin’s ideas, for example, have had a majorimpact in the areas of group dynamics, moti-vation, and leadership. Perhaps Lewin’sgreatest contribution was his willingness touse research to solve practical problems inboth organizational and community settings.


The term action research, which is typicallyassociated with Lewin, refers to the idea thatresearchers and organizations can collabo-rate on research and use those findings tosolve problems. The scientist-practitionermodel can be traced to the action researchmodel and thus stands as one of Lewin’smost important contributions to the field(see Comment 1.6).

A Period of Growth

World War II had a tremendous impact onthe growth of organizational psychology. Forexample, one of the results of World War IIwas that women were needed to fill many ofthe positions in factories that were vacatedby the men called into military service.Also, shortly after World War II in 1948,

COMMENT 1.6

KURT LEWIN: THE PRACTICAL THEORIST

KURT LEWIN WAS born in 1890 in the village

of Mogilno, which was then part of the Prus-

sian province of Posen (now part of Poland).

Lewin’s father owned a general store, as wellas a small farm, so the family was prosperous

although not wealthy. In 1905, Lewin’s family

moved to Berlin, largely to gain better educa-

tional opportunities than were available in

Mogilno. Lewin entered the University of

Frieberg in 1909, initially with the goal of

studying medicine. His distaste for anatomy

courses contributed to Lewin’s abandoningthe goal of becoming a physician. He switched

his interest to biology. This led to a transfer

first to the University of Munich and ulti-

mately to the University of Berlin, where he

eventually earned his doctorate in 1916. After

returning from military service during World

War I, he began his academic career.

The years at Berlin were very productive,and Lewin’s work became quite influential.

At this time, Lewin began to develop an interest

in the application of psychology to applied

problems such as agricultural labor, produc-

tion efficiency, and the design of jobs. Lewin

became quite interested in scientific manage-

ment, particularly the impact of this system on

workers. Lewin and his family left Germany in1933 due to the rise of the Nazi party. He

initially received a temporary appointment at

Cornell University, and ultimately moved to

the University of Iowa Child Welfare Research

Station. While at Iowa, Lewin conducted influ-

ential studies on a variety of topics, includingchild development, the impact of social cli-

mates, and leadership. Following his years at

Iowa, Lewin became deeply involved in the

Commission on Community Relations, which

was established by the American Jewish Con-

gress. During his involvement, Lewin initiated

a number of ‘‘action research’’ projects aimed

at enhancing understanding of communityproblems such as racial prejudice, gang vio-

lence, and integrated housing. Remarkably,

during this same time, Lewin also founded

the Research Center for Group Dynamics at

MIT. Lewin’s work at the Center continued

until his death in 1947, at the age of 56.

In retrospect, it is hard to imagine anyone

having a greater impact on the field of organi-zational psychology than Kurt Lewin. His ideas

continue to influence the study of a number of

areas such as employee motivation, leadership,

group dynamics, and organizational develop-

ment. However, perhaps Lewin’s most endur-

ing legacy was his innovative blending of

science and practice.

Source: A. J. Marrow. (1969). The practical theorist: The life

and work of Kurt Lewin. New York: Basic Books.


President Harry S. Truman made the deci-sion to pursue racial integration of themilitary. Both events were extremely impor-tant because they represented initialattempts to understand the impact of diver-sity on the workplace, a topic that hasbecome quite pertinent in recent years.

World War II also served as the impetusfor major studies of morale and leadershipstyles. Although Hollywood has managed toportray a somewhat idealized version ofWWII, the U.S. military experienced prob-lems with low morale and even desertion.Thus, troop morale and the influence ofleadership were issues of great practicalimportance during this time.

Another very important event in thedevelopment of organizational psychologywas the publication of Morris Viteles’ bookMotivation and Morale in Industry (1953).This was significant because Viteles’ 1932book, Industrial Psychology, had containedvery little on the organizational side of thefield, largely because there simply wasn’tmuch subject matter at that time. Thus,the 1953 book signified that the organiza-tional side of the field had finally arrivedand had a significant role to play in thebroader field of industrial psychology. Itwas also during the post-WWII period thatthe human relations perspective emergedwithin the field. Those who advocated thisperspective (e.g., McGregor, 1960) arguedthat the way organizations had traditionallybeen managed kept employees from beingcreative and fulfilled on the job. Duringthis time, for example, Herzberg conductedhis studies of job design and job enrich-ment, and major research programs inves-tigating both leadership and job satisfactionwere conducted. By the early 1960s, orga-nizational psychology was clearly an equalpartner with the industrial side of the field(Jeanneret, 1991).

Another broader social factor impactedthe development of organizational psychol-ogy during the 1960s and early 1970s: theUnited States’ involvement in the VietnamWar, which led to many cultural changes inAmerica and in other countries. During thisperiod, for example, many young peoplebegan to question conventional societalnorms and the wisdom of traditional societalinstitutions such as education, government,and the legal system. Many, in fact, suspectedthat the federal government was not truthfulabout many important details of the war.Furthermore, subsequent accounts of thewar by historians have proven that many ofthese suspicions were justified (e.g., Small,1999). People at that time also began to feelas though they should have much more free-dom to express themselves in a variety of ways(e.g., hairstyles, dress, speech).

For organizations, the cultural changesthat arose out of the 1960s had major impli-cations. In essence, it was becoming less andless common for people to blindly followauthority. Therefore, organizations had tofind methods of motivating employees, otherthan simply offering financial incentivesor threatening punishment. It was alsobecoming more and more common foremployees to seek fulfillment in areas of theirlives other than work. Thus, it was becomingincreasingly difficult to find employees whowere willing to focus exclusively on work.

Maturity and Expansion

From the early 1970s into the 1980s, orga-nizational psychology began to mature as afield of study. For example, during the early1970s, the name of Division 14 of the Amer-ican Psychological Association (APA) wasformally changed from Industrial Psychol-ogy to Industrial/Organizational Psychology.Also during this period, organizational


psychologists began to break significant newground in both theory and research. As just afew examples, Salancik and Pfeffer (1978)proposed Social Information ProcessingTheory (SIP) as an alternative to more tradi-tional need-based theories of job satisfactionand job design. Also, roughly during thisperiod, organizational psychology began torediscover the impact of personality and dis-positions on things such as job attitudes(Staw & Ross, 1985) and perceptions ofjob-related stress (Watson & Clark, 1984).

Another noteworthy development thattook hold during this period, and continuestoday, was the recognition that behavior inorganizations is impacted by forces at boththe group and organizational levels (e.g.,James & Jones, 1974; Rousseau, 1985). Thismultilevel perspective has had major impli-cations for the field in guiding theory devel-opment as well as statistical methodology(e.g., Dansereau, Alutto, & Yammarino,1984; James, Demaree, & Wolf, 1984). Dur-ing this same period, organizational psychol-ogists began to devote increasing attention towhat could be called nontraditional topics.For example, more literature began to appearon work/family issues (e.g., Greenhaus &Buetell, 1985), job-related stress and health(Beehr & Newman, 1978), and retirement(Beehr, 1986). This willingness to explorenontraditional topics was significant becauseit served as evidence that the interests oforganizational psychologists had broadenedbeyond purely management concerns.

From roughly the late 1980s to the year2000, a number of trends have impacted thefield of organizational psychology. If onetakes a global perspective, perhaps the mostsignificant event of this period was thebreakup of the Soviet Union and the eventualfall of many Communist regimes. Theseextraordinary events have implications fororganizational psychology because a number

of the nations that embraced democracy dur-ing this period have also attempted to estab-lish free-market economies. As many of thesenew democracies found out, managing andmotivating employees in state-owned busi-nesses is quite different from doing so in afree-market economy (Frese, Kring, Soose,& Zempel, 1996; Puffer, 1999; Stroh &Dennis, 1994). The science and the practiceof organizational psychology can potentiallyhelp these nations make this difficult eco-nomic transition.

Another important trend that is takingplace, both in the United States and world-wide, is the change in the demographic com-position of the workforce. The world popula-tion is aging rapidly and becoming moreethnically diverse. One of the implicationsof these demographic shifts is that organiza-tional psychologists will likely devote muchmore time and attention to understandingthe process of retirement (e.g., Adams &Beehr, 1998). Organizational psychologistswill likely help organizations as they assistemployees in making the retirement transi-tion. The increasing level of cultural diversitywill also have wide-ranging implications.Organizational psychologists will increas-ingly be called upon to investigate the impactof cultural differences on organizationalprocesses such as socialization, communica-tion, and motivation.

A third trend that has become evidentduring this period is the move away fromhighly specific jobs, and toward more tem-porary, project-based work. Some havelabeled this dejobbing (Bridges, 1994). Thistrend has a number of implications for orga-nizational psychology. At the most funda-mental level, this trend has impactedand will continue to impact the psycholog-ical contract between organizations andemployees. What does an organization oweits employees? What do employees owe the


organization they work for? In the past,the answers to these questions were ratherstraightforward; now, they have becomeincreasingly complex.

Another implication of this trend is thatmany individuals are not employees in theway this word has typically been used in thepast. Rather, it has become increasingly com-mon for individuals to hire themselves outon a project or per diem basis. This trendsuggests a number of interesting and chal-lenging issues for organizational psycholo-gists. How does an organization maintain aconsistent culture and philosophy with arelatively transient workforce? Is it possibleto motivate temporary employees to performbeyond an average level of performance?Although some research has been done ontemporary, project-based work (Gallagher,2005), more research clearly needs to bedone before these questions can be answeredwith any degree of certainty.

Recent Past and Beyond

On the morning of September 11, 2001,hijacked commercial aircraft crashed intothe World Trade Center in New York Cityand The Pentagon outside of Washington,DC. In terms of casualties, 9/11 representsone of the worst terrorist attacks in history,and certainly the worst on U.S. soil. Further-more, for many readers of this text, 9/11represents the defining moment of theirgeneration, much the same way that theKennedy assassination was for previousgenerations.

What are the implications of 9/11 fororganizational psychology? This is a difficultquestion to answer with a high degree ofcertainty because of the magnitude of theseevents. Probably the most direct way thatmany organizations were impacted was inthe area of emergency preparedness. That

is, 9/11 made many organizations aware ofthe need for having plans in place in case ofemergencies. Had it not been for the emer-gency plans of many of the organizationswith offices in the World Trade Center, thedeath toll of 9/11 would likely have beenmuch higher.

Other than 9/11, other recent trends thathave had the greatest impact on organiza-tions have been technological change,increasing use of telecommuting and otherflexible work arrangements, and increasedglobalization, to name a few. Advances incommunication technology, for example,allow things to be done much quicker inorganizations than was possible even 20years ago. This technology has also allowedemployees much greater flexibility in work-ing arrangements, yet at the same time, madeit more difficult for them to separate theirwork and nonwork lives. Considering all ofthese trends, it is clear that the work world ofthe recent past and not-too-distant futurewill be highly complex and fast paced. Thismay seem rather intimidating, but it is also avery exciting prospect for the field of orga-nizational psychology because it will allowfor truly groundbreaking research and prac-tical applications. In fact, this is one of themost exciting times in history to be involvedin the science and practice of organizationalpsychology.

THE CHAPTER SEQUENCE

A textbook should function as a tour guidefor the student. In our experience, both asstudents and course instructors, the best wayto guide is in a logical sequential fashion. Thesequence of chapters in this book was devel-oped with this consideration in mind. Thechapters in Part I provide introductory mate-rial on the field of organizational psychologyas well as its methodological foundations.

The Chapter Sequence 17

Some students (and maybe even someinstructors) may find it unusual to have achapter on research methodology. I’veincluded it for three primary reasons. First,having at least a rudimentary understandingof research methodology is fundamental tounderstanding many of the concepts andresearch findings discussed throughout thetext. Second, research methodology is a legit-imate area of inquiry within organizationalpsychology. In fact, a great deal of importantresearch within organizational psychology inrecent years has been methodologically ori-ented. Finally, as a course instructor andsupervisor of student research, I have foundthat students often forget (or perhapsrepress) what they learn in research methodscourses. Covering research methods in con-tent courses often compensates for this for-getting.

The first seven chapters focus on thebehavior of individuals in organizational set-tings. A close examination of these chaptersreveals a sequential ordering. It is assumedthat individuals are initially socialized intoan organization (Chapter 3), become pro-ductive members of that organization (Chap-ter 4), and derive some level of satisfactionand commitment (Chapter 5). It is also rec-ognized that individuals may engage inbehaviors that are counterproductive to theiremployers (Chapter 6), and that work mayhave a negative effect on the health and well-being of employees (Chapter 7).

The next three chapters focus on themechanisms that organizations use to influ-ence employees’ behaviors. To this end,Chapter 8 covers the major motivation theo-ries in organizational psychology. In Chap-ter 9, we examine the various ways in whichorganizations utilize theories of motivation toactually influence employees’ behaviors.Chapter 10 examines the other primary mech-anism that organizations use to influence

behavior, namely leadership. Chapter 10 alsoexamines power and influence processes thatare at the core of leadership and yet influencemany other behaviors in organizations.

In the next two chapters, the focus of thebook shifts from the individual to the grouplevel. This is very important, given theincreased reliance on teams in many organi-zations. Chapter 11 introduces the basicconcepts underlying group behavior. Chap-ter 12 describes the factors that have thegreatest impact on group effectiveness, aswell as the dynamics underlying intergroupbehavior.

In the final three chapters, the focusshifts from the group to the organization—the macro level. Chapter 13 reviews severaltheoretical approaches used to define anorganization and examines approaches toorganizational design. Chapter 14 probesthe concepts of organizational culture andclimate. Chapter 15 describes the variety ofways in which organizations engage inplanned change with the assistance of behav-ioral science knowledge.

One topic that readers will notice is notthe focus of any one chapter is internationalor cross-cultural issues. This book examinescross-cultural issues in the context of thevarious topics covered in the chapters. Thiswas done intentionally because we believecross-cultural findings are best understoodand assimilated in the context of specifictopics.

CHAPTER SUMMARY

Organizational psychology is the scientificstudy of individual and group behavior informal organizational settings. While it is alegitimate field of study in its own right,organizational psychology is actually part ofthe broader field of Industrial/Organiza-tional (I/O) psychology. Organizational


psychologists use scientific methods to studybehavior in organizations. They also usethis knowledge to solve practical problemsin organizations; this is the essence of thescientist-practitioner model, the model onwhich most graduate training in I/O psychol-ogy is based. Thus, those with training inorganizational psychology are employed in

both academic and nonacademic settings.Historically, organizational psychology wasslower to develop than the industrial sideof the field. The event that is usually consid-ered the historical beginning of organiza-tional psychology was the Hawthornestudies, although many other events andindividuals have helped to shape the field

PEOPLE BEHIND THE RESEARCH

MIKE J. ZICKAR AND THE HISTORY OF I-O PSYCHOLOGY

I am a firm believer in attacking life with a

balance between goal-directed behavior as

well as an openness to the moment. I received

my Ph.D. from University of Illinois in 1997

and have been at Bowling Green State Univer-

sity since then. My training and primary

research has been in psychometric work

related to employment testing, a very quanti-tative area that is about as left-brain domi-

nated as any research area in I-O psychology.

I got into historical research about 10 years

ago when I started asking questions about the

current state of I-O psychology; these ques-

tions were stimulated by a frustration with the

way that I-O psychologists ignored certain

topics that I viewed important. What functiondo labor unions have in the day-to-day lives of

employees? Why do I-O psychologists, for

the most part, focus on improving efficiency

more so than improving worker well-being?

These questions prompted me to look back inthe history of our field.

Historical research relies much more

on the right side of the brain compared to

quantitative research. I find that balance very

stimulating and useful in maintaining my

intellectual curiosity. One of my favorite his-

torical pieces is an article about my hero

Arthur Kornhauser who was an early progres-sive I-O psychologist that worked actively

with labor unions in the auto industry. When

I contacted one of Kornhauser’s former stu-

dents, he seemed unfriendly and impatient

until I told him that I wanted to discuss his

former advisor. Instantly on hearing his for-

mer advisor’s name, he warmed up and said

in a mournful but pleasant tone, ‘‘I alwaystold my wife that if we had had a son, I would

have wanted to name him Arthur!’’

I did not start off my career interested in

historical research, though I now find it an

important part of my intellectual identity.

Regardless of what path your career takes

you, I hope that you will remain open to change

and that you stumble occasionally onto newdirections and ideas.

Mike J. Zickar

Department of Psychology

Bowling Green State University

Chapter Summary 19

over the years and will continue to do so. Aconstant thread through the history of thefield is the dynamic interaction between sci-ence and practice—in most cases for the bet-terment of organizations and their employees.

SUGGESTED ADDITIONALREADINGS

Dunnett, M. D. (1998). Emerging trendsand vexing issues in industrial and organiza-tional psychology. Applied Psychology: AnInternational Review, 47, 129--153.

Koppes, L. L. (1997). American femalepioneers of industrial and organizationalpsychology during the early years. Journalof Applied Psychology, 82, 500--515.

Landy, F. J. (1997). Early influences onthe field of industrial and organizationalpsychology. Journal of Applied Psychology,82, 467--477.

Hyatt, D., Cropanzano, R., Finfer, L. A., Levy,P., Ruddy, T.M., Vandaveer, V., & Walker,S. (1997). Bridging the gap between academ-ics and practice: Suggestions from the field.Industrial-Organizational Psychologist, 35, 29--32.

Kanigel, R. (1997). The one best way: Fred-erick Winslow Taylor and the enigma of effi-ciency. New York: Viking.

Zickar, M. J. (2003). Remembering ArthurKornhauser: Industrial psychology’s advo-cate for worker well-being. Journal of AppliedPsychology, 88, 363--369.


Chapter Two

Research Methodsand StatisticsO

rganizational psychologists oftendesign scientific investigationsto answer a variety of researchquestions about behavior in orga-nizational settings; in some cases

research is designed to test theories. In orderto conduct research, one must make use ofresearch designs, as well as a variety of sta-tistical analyses. As will be shown in thischapter, research methods may range fromsimple observation of behavior to more elab-orate designs. Likewise, statistical methodsmay range from very simple descriptivemeasures, to very elaborate model testing.

Research methodology and statisticalanalysis are also crucial to the practice oforganizational psychology. For example,organizational psychologists often use sys-tematic research methods to provide organi-zational decision makers with informationregarding employees’ attitudes. In othercases, research methodology and statisticalanalysis may be used to evaluate some inter-vention designed to enhance organizationaleffectiveness. An organization may want toknow, for example, whether a team develop-ment intervention will enhance the function-ing of work groups. This question, andothers like it, can also be answered with theaid of typical research methods and statisticalanalyses used in organizational psychology.

In addition to facilitating the scienceand practice of organizational psychology,research methodology and statistical analysishave both emerged as legitimate fields ofstudy within organizational psychology.Some organizational psychologists studytopics such as job satisfaction, motivation,and organizational change; others have

devoted their attention to methodologicaland statistical issues. For example, thereare organizational psychologists who inves-tigate the validity of self-report measures(e.g., Spector, 1994), as well as the analysisof data from multiple organizational levels(Bliese & Jex, 2002). Both topics will bediscussed later in the chapter.

This chapter is designed to provide anintroduction to the methods organizationalpsychologists use to collect data, as well asthe statistical techniques used to analyzethat data. From the student’s perspective,research methodology and statistics are oftenviewed with some degree of apprehension.Even at the graduate level, courses inresearch methodology and statistics are often

21

the most feared. Despite these negative per-ceptions, research methodology and statis-tics courses are probably the most valuablepart of graduate training. Students who arewell grounded in research methodology andstatistics are in the best position to read andcritically evaluate the research literature.They also possess a set of skills that are quitevaluable, regardless of the setting in whichthey choose to work.

METHODS OF DATACOLLECTION

There are literally thousands of researchquestions that have been, and continue tobe, explored by organizational psychologists.Are employees who perceive a high levelof autonomy in their work likely to behighly satisfied with their jobs? Does ahigh level of conflict between work andfamily responsibilities lead to poor health?Does job performance remain consistentover time? Regardless of the research ques-tion being asked, there is a need forrelevant data to be collected if the questionis ever to be answered. In this section, fourdata-collection methods will be discussed.These include observational methods, sur-vey research, experimentation, and quasi-experimentation.

Observational Methods

Observational methods actually encompass avariety of strategies that may be used to studybehavior in organizations (Bouchard, 1976).Simple observation, the most basic of thesestrategies, involves observing and systemati-cally recording behavior. If one wishes, forexample, to investigate decision-makingprocesses used by corporate boards of direc-tors, one might observe these individualsduring quarterly meetings and record rele-

vant observations. These observations mayreveal that the chairperson has more inputinto decisions than other board members, orperhaps that younger board members haveless input into decisions than their moreexperienced counterparts.

The primary advantage of simple obser-vation is that it allows behavior to be cap-tured in its natural context. This allows theresearcher to avoid the problem of reactivity,or changing the phenomenon of interestin the process of measuring it. This is onlya potential advantage, however, becausethe presence of an observer could causeresearch participants to act differently thanthey normally would. One way to addressthis issue is to establish rapport withresearch participants to the point where theyare comfortable enough with the researcherto act naturally. Another option would beto observe the behavior of interest withoutbeing detected. For example, if one wereinterested in the emotions displayed by serv-ice employees toward customers, one mightsit in a coffee shop and observe as customers’orders are taken. This technique is also usedby many retail stores; they send mysteryshoppers to stores in order to measure thequality of customer service. Observingbehavior in this way raises ethical concerns,however, because when it is used, researchparticipants typically are not able to makean informed choice as to whether they wishto participate in the research.

Despite potential advantages, a primarydisadvantage of simple observation is that itis a very labor-intensive activity. Observingand recording behavior takes a great deal oftime and effort. Also, once observations arerecorded, making sense out this informationcan be very time consuming as well. Anotherpotential disadvantage is that observationsare often subjective and may be impactedby the observer’s biases. Nevertheless,

22 Research Methods and Statistics

simple observation can often be quite useful,particularly in the very early stages of aresearch program. Also, from a practical per-spective, managers may find the informationgenerated from observational studies easierto understand, and therefore more useful,than numerical data.

A form of simple observation that maybe useful in some cases is participant obser-vation. Participant observation is essentiallythe same as simple observation, except thatthe observer is also a participant in the eventhe or she is studying. In the previous exam-ple of a researcher studying corporate boardsof directors, this could be participant obser-vation if the researcher were also a memberof the board. Participant observation canbe highly useful, particularly when beinga participant in an event provides the re-searcher with information that may not beobtained otherwise. This point is illustratedwell by Van Maanen’s (1975) investigation ofpolice recruits as they made the transitionfrom the training academy to regular policework. In conducting this study, Van Maanenparticipated in the police academy trainingas a recruit, and thus became a participantin the event he was studying. By doingthis, he undoubtedly was able to gatherinformation that would have been unavail-able through the use of other methods (seeComment 2.1).

Despite the potential advantages of par-ticipant observation, this method also carriessome risks. The biggest of these is that bytaking on the role of participant, a researchermay change the phenomenon under investi-gation. This is somewhat ironic, consideringthat the general advantage of observationalmethods is that they reduce the risk of reac-tivity. Being a participant may also lead theresearcher to lose his or her objectivity. Aspreviously stated, all observations are subjectto distortion, but assuming the role of a

participant may compound this problem. InVan Maanen’s (1975) study, this problemwas dealt with by supplementing his obser-vations with survey data from other policerecruits.

Archival Data

A second method for studying behavior inorganizations is through the use of archivaldata sources. Archival data represent anyform of data or other records that are com-piled for purposes that are independent ofthe research being conducted (Webb, Camp-bell, Schwartz, Sechrest, & Grove, 1981).Compared to other observational methods,the use of archival data is more prevalent inorganizational psychology, largely becauseof the sheer abundance of archival datasources. Within organizations, records aretypically kept on many employee behaviorssuch as job performance, absenteeism, turn-over, and safety, to name a few. In addition,the governments of many countries maintaindatabases that may be relevant to the study ofbehavior in organizations. In the UnitedStates, for example, the Department of Laborproduces the Dictionary of OccupationalTitles (DOT), which contains informationon the working conditions of a vast numberof occupations. This database has beenused in several investigations of behavior inorganizations (e.g., Schaubroeck, Ganster, &Kemmerer, 1994; Spector & Jex, 1991).Recently, the DOT has been supplementedby a more extensive database in the form ofthe Occupational Information Network(O�NET). This represents an improvementover the DOT because the occupations thatcomprise the O�NET are more up to date,and the dimensions on which these occupa-tions are described are more extensive. Todate, only a few studies have used O�NET asan archival data source in the same manner


as the DOT (e.g., Liu, Spector, & Jex, 2004),but it is likely that more will follow.

In addition to these common archivaldata sources, organizational psychologistshave also made use of other less commonsources to study organizational processes.Sports statistics, in particular, are widelyavailable and can be used to examine (albeitindirectly) organizational processes. Organi-zational psychologists, for example, haveused the performance of professional base-

ball players to study equity theory (Lord &Hohenfeld, 1979), and professional hockeyplayers to study leadership processes (Day,Sin, & Chen, 2004).

The use of archival data offers severaladvantages to researchers. First, manyarchival databases are readily available tothe public and can be accessed quiteeasily—in many cases, via the Internet. Sec-ond, archival data are nonreactive. Archivaldata typically are not collected for the

COMMENT 2.1

THE PROS AND CONS OF QUALITATIVE RESEARCH METHODS

WITHIN THE GENERAL field of psychology, and

organizational psychology in particular, qual-

itative data collection methods such as obser-

vation are not widely used. In other fields suchas sociology and anthropology, qualitative

methods are used quite frequently. In psy-

chology, we make much greater use of surveys

and, to a lesser extent, experimentation and

quasi-experimentation (Sackett & Larsen,

1990). In talks with colleagues over the years,

the typical disadvantages associated with

qualitative methods have been that they aretoo labor-intensive and too many biases are

associated with the observational process.

Unfortunately, because of these disadvan-

tages, many in the field psychology fail to see

many of the positive features of qualitative

data-collection methods. Chief among these

is that observation typically provides a much

richer description of whatever one is trying tostudy than questionnaire data do. For example,

observing a group working together for a week

is probably more meaningful than knowing

that group members rate the group’s cohesive-

ness as 4.3 on a scale of 1 to 6. Another

advantage of most qualitative data-collection

methods is that they do not require research

participants to provide assessments of either

themselves or the work environment. For

example, we may be able to determine, through

observations, that an employee has a great deal

of autonomy built into his or her job. If we wereto ask the employee several questions about job

autonomy via a questionnaire, the employee’s

responses might be biased because of a tem-

porary mood state or overall job satisfaction.

In reality, researchers do not have to make

either/or decisions in choosing between qual-

itative and quantitative research methods. For

example, in conducting employee opinion sur-veys, I typically use closed-ended question-

naire items, but I also include space at the end

of the survey for employees to write comments

that are then analyzed for content. This allows

for quantitative analysis of the closed-ended

survey items, but employees can express their

opinions in their own words. Written com-

ments may also reveal very useful suggestionsto organizational decision makers.

Source: P. R. Sackett and J. R. Larsen, Jr. (1990). Research

strategies and tactics in industrial and organizational

psychology. In M. D. Dunnette & L. M. Hough (Eds.),

Handbook of industrial and organizational psychology (2nd

ed., Vol. 1, pp. 419–490). Palo Alto, CA: Consulting

Psychologists Press.


researcher’s purpose, so there is no chancethat participants will distort responses in away that would impact the validity of theresearch. Finally, when archival data areused to measure employee behaviors, suchrecords are usually less subject to distortionthan self-reports of the same behavior.

Despite these advantages, the use ofarchival data may present several problems.One is that archival databases contain onlyindirect measures of the phenomenon ofinterest to the researcher. Use of databasessuch as the DOT or O�NET to measure char-acteristics of employees’ jobs illustrates thisproblem quite well. Information containedin both of these databases is collected at theoccupation level, so using it may maskimportant differences between individualswho may have the same occupation but per-form substantially different work, or whoperform under very different conditions.For example, a nurse employed in a ruralhealth clinic may have very different jobduties than one employed in a large urbanhospital, even though they are part of thesame occupation.

This issue becomes even more problem-atic when researchers use sports statistics tostudy organizational processes. Lord andHohenfeld (1979), for example, examinedthe performance of baseball free agents inwhat is termed the arbitration year, or the yearprior to going on the free-agent market. Basedon the players’ performance in this year, andthe first year of their new contract, theseresearchers made inferences about how theseplayers resolved their feelings of under-payment. What these researchers didn’t do,however, was ask these players directly aboutwhether they felt underpaid or how theyplanned to resolve feelings of underpayment.

Another potential problem with archivaldata is accuracy. Organizations differ widelyin the precision of their record-keeping prac-

tices. Furthermore, there may be instanceswhen an organization has some incentive todistort records. For example, organizationsmay underreport accidents or other negativeincidents in order to avoid negative publicityor increases in insurance costs. Accuracy isprobably less of an issue when archival dataare obtained from government agencies andwell-known academic research institutes.Nevertheless, when using any form ofarchival data it is always a good idea to askfor some evidence supporting the accuracyof the information.

Survey Research

By far the most widely used form of datacollection in organizational psychology issurvey research (Scandura & Williams,2000). Survey research simply involves ask-ing research participants to report abouttheir attitudes and/or behaviors, either inwriting or verbally. This form of research isextremely common in our society and is usedto gather information for a wide variety ofpurposes. Most readers have probably par-ticipated in some form of survey research.

Before describing the general stepsinvolved in conducting a survey researchproject, it is useful to consider the purposesof survey research. In some cases, surveyresearch is designed to provide purelydescriptive information. For example, thetop management team in an organizationmay wish to know the current level ofemployee job satisfaction, a governmentagency may want to assess the income levelof working adults, or a research institute maywant to know the level of drug use amongteenagers. Studies designed for this purposeare often referred to as prevalence studies.

Survey research is also conducted to testhypotheses regarding the relationshipsbetween variables. For example, a researcher


may want to assess whether employees whoperceive a great deal of autonomy in theirjobs also report a high level of job satisfac-tion. The researcher in this case is not reallyconcerned with the actual levels of autonomyor job satisfaction, but rather is interested inwhether these two variables are related.

As shown in Figure 2.1, the first step inconducting a survey research project is toidentify the variables that one will be meas-uring. For theoretically based research pro-jects, the variables will be directly linked tothe research question one is examining. Aresearcher studying the relationship betweeninterpersonal conflict on the job andemployees’ satisfaction with their jobs wouldobviously measure both of these variables.The choice of variables to be measured in

more applied research is often based on theconcerns of upper-management personnelor, in some cases, input from employeesfrom all levels of the organization. Determin-ing what to measure in surveys is oftenachieved through the use of focus groupsconsisting of either top managers or groupsof other employees. A focus group is a qual-itative data-gathering technique that is oftenused to generate ideas during the prelimi-nary stages of a research project. For exam-ple, to determine what to measure on asurvey, a researcher might conduct a focusgroup with the top management of an orga-nization. The researcher might begin thefocus group session by posing a question:‘‘What are the biggest concerns of employeesin this organization?’’ This would be fol-lowed by an open-ended discussion, duringwhich the researcher would take note ofmajor issues that come up.

Once the researcher has decided whichvariables to measure, the next step is to con-duct an extensive search of relevant literatureon these variables. This is done to determinewhether acceptable measures of the variablesexist. For many variables of interest to orga-nizational psychologists, several acceptablemeasures are available. Using previouslydeveloped measures saves a researcher con-siderable time because there is no need todevelop new measures. While this can usu-ally be done in theoretically based researchprojects, using established measures inapplied projects such as employee opinionsurveys is often more difficult to do. This isbecause many of the variables measured inemployee opinion surveys may be unique toa particular organization. In the presentauthors’ experience, organizations oftenwant survey items customized in order toenhance the relevance of the information.

Once a researcher has decided on thevariables to be measured and identified

FIGURE 2.1Steps Involved in Conducting a Survey ResearchProject

Identify Variables

Literature Search

Questionnaire Design

Sampling

Data Collection

Data Analysis and Presentation


acceptable measures, the next step is todesign the questionnaire or survey instru-ment. This step is extremely importantbecause the quality of the questionnaire willstrongly impact the integrity of the data gen-erated. Designing a high-quality surveyinstrument is a time-consuming, painstakingprocess. Fortunately, there are excellentsources of information one can refer to forassistance in the questionnaire design proc-ess (e.g., Dillman, 2000). One general ruleshould guide the development of any ques-tionnaire: It should be easy for the respondent tocomplete. That is, instructions should be easyto understand, response categories should bewell defined, and the items should be clearlywritten. It is probably for this reason thatmore and more researchers have been mak-ing use of Internet-based collection of surveydata. Respondents need only click the appro-priate response categories when completingan Internet-based survey, and simply click asend button when they’re done. A potentialdisadvantage of Internet-based data collec-tion, of course, is that the researcher haslittle control over who actually completesthe survey instrument. It is also possible thatsamples generated via the Internet may sys-tematically differ (e.g., they may be moreeducated) from samples generated throughother methods.

Another important step in the question-naire design process is to conduct some formof pilot testing, even if this involves simplyasking a colleague to read through the ques-tionnaire. Careful pilot testing may revealunclear instructions, poorly worded items,or even misspellings.

After the questionnaire is designed andpilot tested, the next step is to determinespecifically who the respondents will be.When research is conducted within organi-zations, this may simply involve asking allemployees to complete the survey. In other

cases, it may be necessary to narrow the poolof responding employees. For example, if aresearcher were studying customer servicebehavior among employees, he or she wouldhave to restrict the pool of respondents tothose employees who have at least somecontact with customers.

In some cases the number of potentialrespondents may be so large that it is imprac-tical for the researcher to include everyone(e.g., a multi-national corporation with50,000 employees). If this is the case, someform of probability sampling may be utilized.The idea behind probability sampling is thatthe researcher selects a sample from a largergroup (or population) in order to generalizethe results to that larger group, within somemargin of error (Fowler, 1984). The mostbasic form of probability sampling is simplerandom sampling. This involves selectingmembers of a population such that all havean equal and nonzero probability of beingincluded in the sample. As an example, aresearcher could randomly select 200employees from a current employee directoryto participate in an organizational survey.

Another form of probability samplingsometimes used is stratified random sampling.This essentially represents the application ofsimple random sampling within identifiablegroups or strata. Stratified random samplingis often used to increase the precision ofestimates (Fowler, 1984); the logic is that ifestimates are made within strata and pooled,these will be more precise than applyingsimple random sampling within an entirepopulation. Stratified random sampling canalso be used to increase the representative-ness of samples. If, for example, an organi-zation consists of five different employeegroups that are represented in equal propor-tions, proportional stratified random sam-pling can be used to increase the chancesthat the sample will reflect this.


A third form of probability sampling thatmay be useful in some cases is cluster sam-pling. What distinguishes this from the othertwo forms of sampling previously describedis that the unit of sampling is no longer theindividual but, instead, some larger unit orcluster. An illustration of how cluster sam-pling can be used comes from a researchproject one of the authors conducted severalyears ago for the U.S. Army Recruiting Com-mand (USAREC). This organization is verygeographically dispersed and consists ofmultiple levels (brigades, battalions, compa-nies, and stations). In the initial stages of theproject, it was decided that approximately50 face-to-face interviews needed to be con-ducted with personnel at brigade, battalion,and company levels. Rather than randomlyselecting individuals from these three levels, itwas decided to first randomly select twobattalions within each brigade. Two indi-viduals were interviewed in each battalion,as well as in the company located closest toeach battalion.

A major advantage of cluster samplingis that it allows a researcher to cut downon travel time and expense. In the previ-ously described project, imagine if simplerandom sampling had been used insteadof cluster sampling. The 50 individualsselected to be interviewed may have beenso geographically spread out that a sepa-rate trip would have been required to con-duct each interview. Of course, the riskone runs in using cluster sampling is thatthe sample may not be as representative asit would be if simple random samplingwere used. In most cases, though, re-searchers will accept the risk of decreasedrepresentativeness in order to cut down oncosts (see Comment 2.2).

Once the researcher determines who theparticipants will be, the next step is toactually collect data. In collecting survey

data, several options are available, and eachoption has advantages and disadvantages.With written organizational surveys, theideal way to collect data is to have groupsof employees complete the questionnaire ina centralized location and return the com-pleted questionnaire to the researcher uponcompletion. This is ideal because it providesthe best chance for a favorable response rate.A very low response rate is undesirablebecause it raises concerns about whetherthe survey results truly represent the targetgroup. For example, in an organizationwhere one of the authors once worked, anemployee opinion survey was conductedand the response rate was approximately10%! This low response rate was revealingin and of itself, but it also raised questionsabout the validity of the information.

In some cases, centralized data collec-tion is not possible because of employees’schedules or concerns about confidentiality.Other options that are used in some cases aremailing questionnaires to employees’ homes,administering a questionnaire verbally bytelephone, or e-mailing questionnaires viathe Internet or Intranet (i.e., internal sys-tem). Although these methods are somewhatless desirable than centralized on-site datacollection, there are actually many ways thatresearchers can use them and achieve veryfavorable response rates (e.g., Dillman et al.,2000).

The final step in conducting a surveyresearch project is the analysis and presenta-tion of the data. The analysis of survey data isdictated by the purpose of the survey. If thepurpose is description (which is usually thecase when organizations initiate surveyresearch projects), analyses are relativelysimple and straightforward. Descriptive sta-tistics (e.g., means, ranges, percentages) willusually suffice in such situations. In caseswhere survey data are used for theoretically


based hypothesis testing, analyses are con-ducted to test hypothesized relationsbetween variables. More detailed informa-tion on statistical analyses used in hypothesistesting will be discussed later in the chapter.However, it is worth noting here that surveydata are typically best for assessing covaria-tion among variables; that is, a change in onevariable is associated with a change inanother. Assessing whether one variablecauses a change in another variable is difficultto do with survey data because such data areusually collected at one point in time.

One way to make survey data more ame-nable to the assessment of causality is to use alongitudinal design. An example of this mightbe measuring employees’ job perceptions atone point in time, and then measuring jobattitudes 6 months later. Compared to cross-sectional designs, longitudinal research isbetter because it at least gives the researchera temporal basis on which to make causal

statements. In the previous example, sincejob perceptions were measured prior to jobattitudes it is certainly plausible that jobperceptions might have a causal impact onjob attitudes. As another example, Britt,Castro, and Adler (2005) examined whetherbeing personally engaged in one’s job couldbuffer soldiers from the adverse effects ofworking long hours and being away a largenumber of days for training. These authorsfound support for this hypothesis by show-ing that work hours and days trainingassessed at one point in time predictedhealth symptoms 3 months later only forsoldiers who reported low levels of jobengagement. However, an obvious downsideto longitudinal research is that it is oftenimpractical because researchers have tokeep track of respondents. In recent years,many researchers have begun using veryshort-term intensive longitudinal designs inwhich participants provide a large number of

COMMENT 2.2

THE COST OF SURVEY RESEARCH

SURVEY RESEARCH IS by far the most commonly

used data-collection method in organizational

psychology. While survey research has many

advantages, it is also true that surveys can bequite costly. Even if a researcher conducts a

relatively simple, paper-and-pencil, self-

administered survey of 200 employees in an

organization, there are monetary costs associ-

ated with photocopying, incentives for

respondents, and in many cases postage for

return envelopes. This type of research project

may also require that a researcher spend timepersonally contacting respondents and

reminding them to complete the survey.

Internet-based surveys reduce photocopying

costs, but the time required to properly design

a web-based data-collection tool is often con-

siderable.

As the size and scope of a survey research

project grows, researchers often must hiresurvey research firms to handle the data col-

lection. This increases efficiency for the re-

searchers, but unfortunately is very costly.

For example, it is not unusual for survey

research firms to charge several thousand dol-

lars to collect survey data even when sample

sizes are relatively modest. Typically when

researchers wish to sample very large numbersof respondents, they must seek financial sup-

port from government agencies or private

foundations in order to cover the costs of these

projects.


observations over a short time period (e.g.,Fuller et al., 2003).

Experimentation

Another common form of data collection inorganizational psychology is experimentation.An experiment is a highly controlled situa-tion that provides a researcher the bestopportunity to assess cause-and-effect rela-tionships. This is important because the hall-mark of any science is to detect and explaincausal relationships.

Because the term experiment is verycommonly used, students are often unclearabout what constitutes a true experiment.According to Cook and Campbell (1979),three characteristics distinguish a true exper-iment from other methods. These are (1)manipulation of an independent variableand measurement of a dependent variable;(2) random assignment to experimental treat-ment conditions; and (3) maximum controlby the experimenter. Let’s examine each ofthese characteristics.

The term independent variable is used todesignate the variable that is proposed tohave some effect on other variables, andhence is typically of primary interest to theresearcher. When the independent variableis manipulated, this means that research par-ticipants experience different levels of thisvariable. If a researcher were interested inthe impact of feedback on performance, forexample, the independent variable would befeedback. This variable could be manipu-lated by providing one group of researchparticipants with feedback after performinga task, while providing no feedback to asecond group.

The measurement of the dependent varia-ble simply involves some systematic record ofthe research participants’ behaviors or atti-tudes that may be impacted by the independ-

ent variable. Choice of dependent measures isoften based on prior research, or acceptedpractice. In organizational psychology, forexample, it is common practice to measureattitudinal-dependent variables with surveys.It is always important, however, to keep inmind that the dependent measure being usedis really just an operational definition of aconcept. For example, job satisfaction repre-sents whether a person has a positive or neg-ative feeling about his or her job or a jobsituation. If a five-item scale is used to assessjob satisfaction, this measure is really beingused to represent this conceptual definition.

The second defining characteristic of ex-perimentation, random assignment, meansthat research participants are assigned togroups receiving different levels of theindependent variable (also called treatmentconditions) in a random or nonsystematicfashion. Randomly assigning research partic-ipants can be done quite easily—for example,by flipping a coin. The logic behind randomassignment is very simple—if research par-ticipants are assigned in a truly random fash-ion, it is likely that the different treatmentgroups will be similar in all ways except forthe independent variable. This allows theresearcher to isolate the independent variableas the cause of any differences between treat-ment conditions on the dependent variable.

The third defining characteristic of anexperiment, maximum control, means thatmanipulation of the independent variableand the measurement of the dependent var-iable are done under controlled conditions.The researcher tries to make sure that all var-iables other than the independent variableare held constant. Like random assignment,this is done to isolate the independent vari-able as the cause of any differences amongthe treatment groups. When experiments areconducted in laboratory settings, researcherscan usually achieve a desirable level of


control. This is a much greater challengewhen experiments are conducted in field set-tings, though not impossible. Eden (1985),for example, conducted a field experiment inthe Israeli Defenses Forces to evaluate theimpact of a team development intervention.

Quasi-Experimentation

According to Cook, Campbell, and Perrachio(1990), a quasi-experiment is similar to a trueexperiment except that it lacks one or moreof the essential features previously described.In organizational settings, the independentvariable of interest often cannot be manipu-lated because it is under the control of theorganization, or may even be a naturallyoccurring event. Examples of independentvariables that are usually under organiza-tional control would include training pro-grams or the redesign of jobs. Naturallyoccurring events that could be used asindependent variables may include computershutdowns, changes in government regula-tions, or mergers. In all of these cases, theresearcher has no control over which researchparticipants receive which treatments.

Quasi-experimental designs are also usedin organizational settings because researchparticipants usually cannot be randomlyassigned to treatment conditions. Assign-ment to training programs provides a goodexample of nonrandom assignment. Employ-ees typically participate in training programs,either voluntarily or on the basis of an iden-tified training need (Goldstein, 1993). Thus,in most cases, if a researcher were to comparetraining-program participants to nonpartici-pants, these two groups could possibly differin important ways.

Given the constraints that accompanyquasi-experimentation, how do researchersset about proving that an independent vari-able has a causal impact on a dependent

measure? One way is to measure and statisti-cally control variables that may obscure therelationship between the independent anddependent variables. For example, if theaverage age of a group of employees receiv-ing one level of the independent variable ishigher than the age of groups receiving otherlevels, age can be measured and statisticallycontrolled when comparing the groups. Thiswould be using age as a covariate.

Other than statistical control, quasi-experimentation typically requires that re-searchers systematically identify and ruleout alternatives to the independent variablewhen differences between treatment groupsare found. According to Cook and Campbell(1979), there are a variety of explanations,other than the independent variable, thatmay lead to a difference between a treat-ment group(s) and a control group inquasi-experimental designs. For example,participants in different groups may beexposed to different historical events, partic-ipants may change at different rates, or par-ticipants may have differing views aboutparticipating in the research.

A researcher conducting a quasi-experiment can never know for sure whetherany number of alternative explanations areimpacting his or her findings. However, it isoften possible to assess the plausibility ofdifferent alternative explanations. For exam-ple, let’s say a researcher conducted a quasi-experiment in which the job of bank tellerwas redesigned at one branch of a bank,but remained the same at another. After 3months it is found that customer satisfactionis much higher at the branch where the jobredesign took place compared to the branchwhere the job was not changed. The jobredesign may have caused the increase incustomer satisfaction, but since this was nota true experiment, there may be explanationsother than the job redesign. To rule out these


alternative explanations, the researcher couldbegin by comparing these two branches tosee whether any preexisting differencesbetween employees in the two branchescould have caused the difference in customersatisfaction. If the employees at these twobranches were similar in terms of tenureand overall job performance, these could beruled out as alternative explanations for thefindings. The researcher could also gatherinformation on the nature of the customerswho frequent each of the two branches. Ifcustomers at the two branches tend to bedemographically similar, and have similarincome levels, this could also be ruled outas an alternative explanation of the differencein customer satisfaction. The researcher, ineffect, plays detective in order to identify andrule out alternative explanations for his orher findings. Note that it is never possible toidentify every possible alternative explana-tion, so researchers typically attempt to ruleout only the most plausible.

Choosing Among Data-CollectionMethods

Given the information presented about eachmethod of data collection, readers may won-der how to choose which method to use.Unfortunately, there is no concrete formulafor making this choice. Perhaps the bestapproach is to weigh the advantages anddisadvantages of each method. As is illus-trated in Figure 2.2, the primary advantageof observational methods is that they providethe researcher with an opportunity to studybehavior in its natural context. Unfortu-nately, observational techniques tend to behighly labor intensive.

Archival data may allow researchers toavoid potential problems associated withself-report measures. An additional advant-age of archival data is that they are often

widely available. The primary disadvantageof archival data is that the researcher usuallyhas little control over how such data werecollected. That is, one must take on faith thatsuch data have been properly collected andare accurate.

FIGURE 2.2Summary of the Primary Advantages andDisadvantages Associated with Each of the FourData Collection Methods

Observational MethodsAdvantages Disadvantages• Behavior is captured in its natural context.• Avoids the problem of “reactivity.”• Some forms of observational data are readily available.

• May be highly labor intensive.• Observations may be subject to bias.• Some forms of observational data only measure behavior indirectly.

Archival DataAdvantages Disadvantages• Easy to obtain• Non-reactive

• Measures behavior indirectly.• Not always accurate

Survey ResearchAdvantages Disadvantages• Allows the collection of data from large numbers of participants at low cost.• Survey data can typically be analyzed with very powerful statistical methods.

•Difficult to draw causal interferences from survey data.• Response rates for some forms of survey data are low.• Survey design is a difficult, time-consuming process.

ExperimentationAdvantages Disadvantages• Best way to assess causal relationships.• Best way to isolate the impact of a specific variable.• Gaining compliance of participants is easier compared to survey research.

• Generalizability of findings may be questionable.•Examining a variable in isolation may be unrealistic.• Participants may not take the experimental situation serious.

Quasi-ExperimentationAdvantages Disadvantages• Allows the researcher a way to access causality in naturalistic settings.• An excellent way to evaluate the impact of organizational interventions.

•Organizations may be reluctant to allow these to be conducted.• Researchers have very little control.


Survey methodology allows the research-er to obtain data from a large number ofparticipants at a relatively low cost. However,it is typically difficult to draw causal infer-ences from survey data, especially when thedata are cross sectional. Experimentationprovides the researcher with the best wayto assess causal relationships. In some cases,however, the generalizability of experimentalfindings may be questionable. Finally, quasi-experimentation, in many cases, offers theresearcher a way to assess causal relation-ships in naturalistic settings. However, quasi-experiments may be difficult to conduct

because researchers typically have little con-trol in most field settings.

Given the advantages and disadvantagessummarized in Figure 2.2, the choice of adata-collection method depends largely on aresearcher’s objectives. If establishing causal-ity is of primary importance, then experimen-tation is likely to be the method of choice.On the other hand, if capturing behavior inits natural context is the primary concern,then observation or quasi-experimentationmay be preferred. Ideally, the best courseof action is to use multiple methods of datacollection (see Comment 2.3).

COMMENT 2.3

THE CASE FOR MULTIPLE DATA-COLLECTION METHODS

UNFORTUNATELY, A SIGNIFICANT portion of research

in organizational psychology suffers from

what has been termed mono-operation bias.

This means that, in many studies, all of thevariables are measured using only one form

of data collection. Often, this one form of data

collection is a self-report questionnaire,

although it does not have to be. For example,

a study would suffer just as much from this

form of bias if all variables were measured

using simple observation.

Why is it a problem to measure all variablesin a study with only one form of data collec-

tion? One obvious reason is that the relation-

ships among variables may be inflated because

they share a common method (e.g., common-

method bias). Another way to view this issue is

to think about the positive impact of using

multiple forms of data collection in a single

study. Let’s say a researcher is interested inwhether job autonomy is positively related to

job satisfaction. Further assume that, in this

study, job autonomy is measured through a

self-report measure completed by employees,

and through archival information collected

during a job analysis. Job satisfaction couldbe measured through a self-report measure and

thorough observation of employees through

their workday.

After these data are collected, we would

likely find that the self-report autonomy meas-

ure would be positively related to the self-

report job-satisfaction measure. However,

what if the archival measure of job autonomyis also related to the self-report job-satisfaction

measure? What if the self-report job autonomy

is positively related to the observational meas-

ure of job satisfaction? If both of these results

occur, this would most certainly strengthen the

conclusion that job autonomy really does pos-

itively relate to job satisfaction. Thus, the real

benefit of using multiple data-collection meth-ods is that it allows us to show relationships

between variables in multiple ways.


SPECIAL ISSUES IN DATACOLLECTION

Now that the most common methods of datacollection have been described, we willexplore, in this section, some important con-temporary issues related to these methods.Contemporary issues include validity of self-report measures, generalizing laboratoryfindings to field settings, gaining access toorganizations for data collection, and con-ducting research in different cultures.

Validity of Self-Reports

Self-report measures are used very frequentlyin organizational psychology. For example,employees are asked to report how muchthey like their jobs, how much variety theyperceive in their work, how committed theyare to their employing organization, and howanxious they feel about their jobs—just tocite a few examples. Because self-reports areused so frequently, we often don’t give muchthought to the assumptions we are makingwhen we use such measures, or whether ornot they are valid. Both issues are examinedin this section.

Self-report measurement is really basedon two implicit assumptions. First, weassume that respondents know the informa-tion we are asking for in self-report meas-ures. Many of the questions asked inorganizational surveys are subjective (i.e.,there is no right or wrong answer), so it ispretty reasonable to assume that respondentsknow this information. Most people knowwhether they like their job, for example. Inother cases, lack of knowledge may compro-mise the validity of self-report measures. Forexample, one of the authors once worked ina university system that conducts an annualsurvey of the job-related activities of faculty.Faculty were asked on this survey to indicate

the number of hours in a typical week theydevote to course preparation, teaching,research, and university service. While someuniversity faculty may keep detailed logs ofwhat they do each day, most probably haveonly a very vague idea of the number ofhours spent on each of the activities on thesurvey.

A second assumption underlying self-report measurement is that respondents willbe truthful in their responses. Compared toresearchers interested in some forms ofbehavior (e.g., drug use, criminal activity),organizational psychologists are relativelyfortunate in this regard. Because most ofthe items on organizational surveys are nothighly sensitive or invasive, employees prob-ably respond truthfully to such items, pro-vided they believe their responses will beheld in confidence. In reality, however,employees’ comfort levels with surveys varygreatly. For example, when organizationalresearchers use self-reports to measurethings such as absenteeism, turnover inten-tions, or various forms of counterproductivebehavior (e.g., theft, sabotage), employeesmay not answer truthfully. In such cases,all a researcher can really do is take greatcare to reassure employees and conduct thesurvey in such a way that supports the prom-ise of confidentiality. This might includeproviding employees with stamped enve-lopes to mail completed surveys to theresearcher offsite, or perhaps making surethere is no identifying information containedon the survey instrument itself.

The situation that has generated the mostcontroversy surrounding the use of self-reports is when such measures are used torate job and organizational conditions. Aresearcher, for example, may ask respond-ents about the level of time pressure in theirjobs. According to Spector (1994), self-reports often do not correlate well with more


objective measures of the work environment,such as ratings by job analysts or by othersfamiliar with the same job (Liu, Spector, &Jex, 2004; Spector, Dwyer, & Jex, 1988;Spector & Jex, 1991). Use of self-reportmeasures is also controversial when suchmeasures are correlated with other self-report variables. When this is the case, thecorrelations between such variables maypossibly be inflated due to common methodvariance—a term that is used quite fre-quently but is rarely explicitly defined. Com-mon method variance represents sharedsources of measurement bias between twovariables that can be directly tied to themethod of measurement being used (Spec-tor, 1987b). As an example, let’s say that aresearcher is measuring two variables viaself-report. Further assume that both of thesemeasures, for some reason, are impacted bysocial desirability responding (Crowne &Marlowe, 1964)—that is, responses to itemsin both measures differ in their levels ofsocial desirability. This shared source ofmeasurement bias may lead these two varia-bles to be correlated, even if there is littleor no underlying conceptual relationshipbetween the two variables. In cases in whichthese measures are conceptually related, thepresence of common method variance mayinflate the magnitude of the relationshipbetween the two variables.

Should researchers be concerned aboutcommon method variance? The consensus inthe literature seems to be ‘‘Yes’’ (e.g., Podsak-off, MacKenzie, Lee, & Podsakoff, 2003).However, empirical efforts to actually dem-onstrate the effects of common method biason relationships between variables have pro-vided only mixed results. Spector (1987b),for example, empirically investigated theprevalence of common method variance inthe measurement of job characteristics andjob satisfaction. Based on an analysis of sev-

eral data sets, he concluded that there was nostrong evidence that correlations wereinflated due to common method variance.

Spector’s (1987) investigation promptedseveral attempts to replicate his findings;most of these attempts utilized more com-plex statistical techniques (e.g., Bagozzi & Yi,1990; Williams & Anderson, 1994; Wil-liams, Cote, & Buckley, 1989). A completediscussion of the findings of these studies isbeyond the scope of this chapter, but thegeneral conclusion of these studies was thatthe impact of common method variance isgreater than Spector had estimated. How-ever, as Brannick and Spector (1990) point-ed out, there are problems in the use ofcomplex statistical methodology to test forthe effects of common method variance.

Perhaps the best way to empiricallyassess the impact of common method var-iance is to compare correlations that containa shared method with those that do not sharea method. Crampton and Wagner (1994)conducted a meta-analysis in which theysummarized 42,934 correlations from stud-ies using single and multiple methods. Over-all, they found that correlations betweenvariables that were both measured via self-report were not appreciably larger than othercorrelations. In the measurement of somevariables, however, correlations based on asingle source were larger than others. Thissuggests that the impact of common methodvariance is real; however, the magnitude ofthis effect varies widely, depending on thenature of the variables being measured.

The best conclusion one can draw aboutthe validity of self-report measures is that itdepends primarily on the variable beingmeasured, and the research question beingasked. For example, if one were interested inmeasuring employees’ feelings about theirjobs, then a self-report measure would bequite appropriate. On the other hand, if


one were interested in measuring employees’levels of job autonomy, levels of discretion indecision making, or (perhaps) workload,then measuring these variables only withself-report measures is not really appropri-ate. This is because in all of these examplesthe researcher is interested in characteristicsof the environment, not those of the individ-ual employee. When researchers wish tomeasure characteristics of the work environ-ment, the best course of action is to usemultiple measurement methods (e.g., Glick,Jenkins, & Gupta, 1986). Given the reliance

of much organizational research on self-report measurement, it is likely that the prosand cons of self-report measurement arelikely to be debated for quite some time(see Comment 2.4).

Generalizing Laboratory Findings

A common criticism of psychology is that itis a science based largely on laboratory stud-ies that investigate the behavior of white ratsand college students. Research in organiza-tional psychology tends not to be conducted

COMMENT 2.4

THE SELF-REPORT CONTROVERSY

SELF-REPORT MEASUREMENT is undoubtedly the

most common form of data collection in orga-

nizational psychology. It is also a form of data

collection that has evoked a great deal of con-

troversy, particularly when self-reports are

used to measure all of the variables in a study.

Dr. Steve Jex has followed this issue for over a

decade, primarily because it has a great deal ofrelevance to his own research program in

occupational stress, since self-report measures

tend to predominate.

On the positive side, self-reports allow us

to measure something that is important in

determining human behavior—namely, indi-

viduals’ perceptions of their environments,

their emotional states, and, in some cases, theirviews of other people. Self-report measure-

ment is also very economical. In the time it

might take to collect meaningful observations

of 20 people, a self-report measure could be

distributed to 100 times that many people.

The primary drawback to self-report meas-

urement is that humans are not analytical

instruments; thus, self-reports may not alwaysproduce accurate information. For example,

when we ask employees to provide self-reports

of characteristics of their jobs, these ratings

may be biased by internal mood states, social

influences of coworkers, or stable internal dis-

positions (Spector, 1994). These same biases

may also influence self-reports of emotional

and affective states.

What is the most reasonable conclusion

one can draw about self-report measures? Inmy opinion, it is that self-report measurement,

like any other data collection method, has both

advantages and disadvantages. Whether one

uses self-report measurement should be dic-

tated primarily by the variables one is trying to

measure, which are ultimately dictated by the

research question one is trying to answer. As a

general rule, if one is primarily interested inperceptions, then self-report measurement is a

logical choice. However, if one is interested in

actual environmental conditions, then self-

reports should be supplemented with other

forms of data collection.

Source: P. E. Spector. (1994). Using self-report question-

naires in OB research: A comment on the use of a con-

troversial method. Journal of Organizational Behavior, 15,

385–392.


as much in laboratory settings in compar-ison to other areas of psychology (e.g.,physiological, cognitive). Laboratory studiesdo, however, still account for a substantialportion of the research in both organiza-tional psychology and I/O psychology ingeneral (Locke, 1986; Sackett & Larsen,1990; Scandura & Williams, 2000). Thepurpose of this section is to examine theissue of whether findings from laboratoryinvestigations can be generalized to realorganizational settings.

The strongest argument made againstlaboratory findings’ generalizing to field set-tings is that laboratory situations lack real-ism. University laboratories are not realorganizations; thus, laboratory settings lackwhat is referred to as mundane realism. Real-ism, however, must also be considered fromthe perspective of the research participant.It is certainly possible to place a researchparticipant in a situation that lacks mun-dane realism, yet manipulate variables insuch as way that participants react genu-inely to the situation. For example, onecan be in a completely contrived situationyet still feel pressure to perform well orconform to group norms. When this is thecase, it can be said that there is a high degreeof experimental realism for research partici-pants. Many classic laboratory studies con-ducted over the years, particularly in socialpsychology (e.g., Asch, 1957; Milgram,1974), have lacked mundane realism yethave retained a very high degree of experi-mental realism.

Another reason laboratory investigationsmay generalize has to do with the researchparticipants. At the beginning of this section,it was remarked, somewhat facetiously, thatlaboratory investigations often utilize collegestudents as participants. This often leads tothe criticism that, because college studentsare different from the general population,

research findings cannot be generalized. Thiscriticism certainly does have some merit—college students are more intelligent andtypically come from higher income levelsthan the general population (Sears, 1986).However, for the study of many organiza-tional issues, use of college students asresearch participants probably does notcompromise generalizability a great deal.College students, for the most part, representthe cadre of individuals who will hold manyof the white-collar jobs in the future. Thus,they may be quite similar to such employees,both in terms of attitudes and abilities, eventhough they are obviously lacking in relevantorganizational experience. By contrast,college students are probably a poor researchsample if the aim is to generalize to employ-ees holding blue-collar and manual laborjobs.

Despite these arguments for the general-izability of laboratory experiments, there areclearly important differences between labo-ratory and field settings. In particular, thehigh level of experimental control in labora-tory settings allows the researcher to isolatethe impact of a variable in a way that isimpossible in field settings because so manythings are occurring that the impact of anysingle variable may be greatly diluted. Also,when variables are investigated in laboratorysettings, they are taken out of their naturalcontext. By taking a variable out of context,the researcher runs the risk of changing thesubstantive nature of that variable. A goodillustration of this point is laboratoryresearch on the effects of ambient temper-ature on aggression (e.g., Baron & Bell,1976). In a laboratory setting, it is possibleto completely isolate the impact of temper-ature. In natural settings, however, temper-ature increases often occur in conjunctionwith other variables such as loud noise andcrowding.


Another important difference to consideris that laboratory settings are typically shortterm (Runkel & McGrath, 1972). As a result,participants in laboratory investigations havevery little time invested and have no reasonto form any social ties with others. In con-trast, employees in organizations invest aconsiderable amount of time in their jobs,and often develop important social ties withfellow employees. These differences betweenlaboratory research participants and actualemployees may lead to very different reac-tions to the same situations.

A final important difference between lab-oratory and field settings is the nature of thetasks performed by research participants.Since laboratory investigations are shortterm, it is very difficult to match the com-plexity of the tasks performed by employeesin real organizations. Thus, many laboratorystudies ask participants to perform relativelysimple tasks such as assembling tinker toys,solving anagrams, and putting together puz-zles. In contrast, employees in organizationalsettings perform much more complex tasks.

After examining the pros and cons oflaboratory investigations, we are still leftwith the question of whether laboratory find-ings generalize to field settings such as orga-nizations. Unfortunately, there is nodefinitive answer to this question, althoughit has been examined extensively (e.g., Ber-kowitz & Donnerstein, 1982; Dipboye &Flanagan, 1979). The most comprehensiveanalysis of this issue, relevant to organiza-tional psychology, is contained in Generaliz-ing from Laboratory to Field Settings, a bookedited by Edwin Locke in 1986. The generalconclusion one can draw from this book isthat well-designed laboratory investigationsoften do generalize to field settings. A well-designed laboratory investigation is one inwhich participants are highly engaged in thetask being performed and variable(s) of

interest are well simulated. The readershould be cautioned, however, againstconcluding that all findings do or do notgeneralize. In the end, generalizability is anempirical question, and the best course ofaction is to replicate laboratory findings infield settings whenever possible.

Gaining Access to Organizations

One of the biggest challenges from fieldresearch is simply gaining access to an orga-nization for data collection. The authors haveknown, over the years, many colleagues andstudents who have come up with interest-ing research questions, but could find noorganization in which to collect data. Un-fortunately, there is very little in the organi-zational literature to help guide researchersin their efforts to gain access to organiza-tions. Thus, most of this section is basedon both the authors’ experience as research-ers, and the experiences of fellow organiza-tional researchers.

Before exploring ways to gain access toorganizations, consider reasons why organi-zations would not let a researcher gatherdata. Based on past experience, there aretwo primary reasons: (1) data collection usu-ally requires employees’ time, and (2) orga-nizations are concerned that employees maydivulge sensitive or proprietary informationabout the organization. Organizations thatoperate in very competitive industries (e.g.,consumer products, high technology) areoften very concerned with divulging anyinformation that might put them at a com-petitive disadvantage. In such organizations,the secrecy surrounding activities such asproduct development often carries over toother activities, regardless of whether theconcerns are warranted.

Given these potential objections tothe collection of research data, how can


organizational researchers still gain access?Perhaps the most fundamental suggestionthat can be made in this regard is: Ask. Manyresearchers who complain about lack ofaccess have actually asked relatively feworganizations for their cooperation. Theysimply assume that they will be unable tocollect data. One way to enlist an organiza-tion is to contact several organizations bytelephone and try to make contact withsomeone in the human resources depart-ment. Another approach is to mass mail toorganizations, asking for cooperation. T. E.Becker (1992), for example, mailed lettersto the presidents of 30 organizations askingfor permission to collect data and eventuallycollected data in one of these.

General appeals or cold calling may resultin a data-collection opportunity, but it isoften much more efficient to use establishedconnections in organizations. Most peoplehave family and/or friends who work inorganizations, and such people may be in aposition to either authorize the collectionof data or put the researcher into contactwith someone who has the authority to doso. This suggests that researchers shouldnot be afraid to use established connectionsin organizations. Researchers should alsoinvest time and energy to develop connec-tions with people who can help with datacollection in their organizations in thefuture. This often takes time and energybut, in the long run, the contact may resultin excellent data-collection opportunities(see Comment 2.5).

Let’s now assume that a researcher haspersuaded an organization to at least con-sider the possibility of data collection. Howcan a researcher convince an organization toactually go ahead with data collection? Themost useful suggestion that can be made inthis regard is: The researcher should offer theorganization something in return for its

cooperation. For example, researchers oftenprovide a summary of the research findingsto the organization, in return for its cooper-ation. Other researchers may offer to per-form some consulting service at no cost toan organization. Organizations typically donot provide a researcher with access to theiremployees unless the access will providesome tangible benefit in return.

After an organization gives permission tocollect data, there is often some negotiationbetween the researcher and the organization,regarding issues such as research design andmeasures. At this stage, researchers and orga-nizations often clash, because of theirdiffering goals and objectives. Researcherstypically desire a high level of methodolog-ical rigor in their investigations becausetheir ultimate goal is to publish their findingsin peer-reviewed journals. Unfortunately,methodological rigor may be perceived bythe organization as costly in a number ofways. For example, supplementing self-report measures with organizational recordsmay be time consuming and require thatemployees reveal identifying information. Itmay also be impossible for an organizationto allow a researcher the control neededfor experimental or even quasi-experimentalinvestigations. This is a tricky issue for re-searchers to navigate because just gainingaccess to organizations is such a challenge.The key is this: The researcher must be will-ing to accommodate the organization, butnot to such an extent that it completely com-promises the scientific integrity of the inves-tigation. Unfortunately, researchers oftenseverely compromise the methodologicalrigor of studies without attempting to per-suade organizations of their value. In mostcases, a well-designed, methodological, rig-orous study will not only help the researcherbut will also be more informative to theorganization (Campion, 1996).


COMMENT 2.5

GAINING ACCESS TO ORGANIZATIONS: SOME EXAMPLES

STEVE JEX: AS I wrote the section on gaining

access to organizations, I thought of the vari-

ous ways I have gained access to organizations

in order to collect data. Like many researchers,I have used family connections. For example,

I was able to gain access to an insurance

company in Tampa, Florida, to conduct my

Master’s thesis research while in graduate

school. My wife was employed there at the

time. To this day, I can’t figure out whether my

wife was trying to advance science, or just

wanted me to get out of graduate school!Another study I conducted, which was ulti-

mately published in the Journal of Applied

Psychology (Jex, Beehr, & Roberts, 1992),

was actually made possible through the efforts

of my mother. This study was conducted at

a hospital in Saginaw, Michigan (my home-

town), where my mother was employed as a

nurse. She introduced me to a person in thehuman resources department who was ulti-

mately able to grant me access to all hospital

employees. In this case, I think my mother’s

help was driven primarily by a desire to see

her son get tenure. In addition to using family

connections, I have gained access in many

other ways. In some cases, current and former

students have helped facilitate data-collectionefforts. I have also, on occasion, relied on

former graduate school classmates, or other

colleagues, to provide either data-collection

sites or useful contacts.

Is there any underlying theme when I think

about the various ways in which I have gained

access to organizations? The most obvious

theme is that developing and maintaining rela-tionships with people is important. This

includes family, students, and professional

colleagues. I’m not suggesting that relation-

ships should be initiated only on the basis of

what people can do for you. However, the fact

is, it is much easier to ask someone for assis-

tance if you’ve taken the time to maintain an

ongoing relationship with him or her. The

other important lesson I’ve learned over the

years is simply to ask. We often assume incor-rectly that family, friends, and colleagues do

not want to be bothered helping with data

collection. However, my experience has been

that people often are very willing (and even

flattered) to help if they’re asked.

Thomas Britt: Getting access to samples is a

critical issue when trying to do quality orga-

nizational research. When I was conductingprimarily social-psychological research I took

for granted how easy it was to obtain samples

by using the subject pool of students taking

‘‘Introduction to Psychology.’’ The more I got

into organizational psychology, the more I

realized the difficulty and tenacity required

to gain access to samples. When I was in the

U.S. Army as an organizational psychologist, Ihad a somewhat captive audience of soldiers to

participate in research projects (although it

should be noted that all soldiers provided

informed consent to participate). However,

even when conducting research with soldiers,

it was necessary to convince unit commanders

of the importance of the research and why it

was worth the time of their soldiers when theycould be spending more time training. You also

had to provide the commanders with summa-

ries of the results that told them something

important, and provide recommendations for

what they should do given the results.

Since arriving at Clemson I have continued

my ties with the military but have also started

collecting data from different applied samples.For example, I have recently begun an assess-

ment of staff at Clemson University examining

the influence of positive motivational states at

work on well-being and performance. I have

worked closely with the administration at all

levels to convey the importance of the research


Conducting Research in DifferentCultures

Given increasing globalization, it is more andmore common for organizational psycholo-gists to examine cross-cultural issues.Despite the value of cross-cultural research,data collection in such studies is often chal-lenging for a number of reasons. For exam-ple, when self-report measures are used,these often must be translated from one lan-guage to another. This may seem rather sim-ple; often, it is not. The typical procedureused to translate self-report measures intodifferent languages is called back translation.This involves translating the items on ameasure from one language to another(e.g., from English to Chinese), and thenback to the original language. The researchermust assess whether the items have retainedtheir meaning to respondents after beingtranslated from a different language.

Another issue researchers must considerin conducting cross-cultural research is sam-pling. Researchers conducting cross-culturalresearch often want to compare employees inone culture to employees in another, so it isimportant to utilize samples that are similarin all aspects except culture (Arvey, Bhagat,& Salas, 1991). The ideal way to accomplishthis would be to utilize employees from dif-ferent cultures who work for the same orga-

nization (e.g., De La Rosa, 2006). If this couldnot be done, a researcher would typicallywant to select samples from different culturesthat work in the same industry and perhapshave similar levels of work experience.

Researchers conducting cross-culturalresearch must be on the lookout for thingsthat are specific to a given culture and mayadversely affect data collection. For example,a researcher utilizing self-ratings of perform-ance must be aware of the fact that, in Asiancultures, it is considered improper to rateoneself high in performance (Fahr, Dobbins,& Cheng, 1991). There may also be vastcultural differences in research participants’degree of comfort when they are asked toprovide ratings of persons in positions ofauthority (Hofstede, 1980).

STATISTICAL METHODSIN ORGANIZATIONALPSYCHOLOGY

Regardless of the data-collection methodused, once data are collected, researchersmust analyze those data to assess whetheror not their hypotheses are supported. For-tunately for organizational researchers, manystatistical methods are available to help makesense out of data. Because a comprehensivereview of statistical methodology is beyond

(and why we need supervisor ratings of per-

formance in addition to employee reports), and

my graduate students and I have spent a great

deal of time coordinating with individual units

in the university to ensure a high response rate

of participation. I have also been struck by the

diversity of jobs people have in a university

setting (e.g., firefighters, library personnel,campus recreation, facilities). When orienting

new graduate students into our Industrial/

Organizational Psychology program, I empha-

size that getting access to a quality sample of

employees who will be willing to work with

you on a research question is often more

difficult than coming up with the hypothesis

you want to test!

Source: S. M. Jex, T. A. Beehr, and C. K. Roberts. (1992).

The meaning of occupational ‘‘stress’’ items to survey

respondents. Journal of Applied Psychology, 77, 623–628.


the scope of this chapter, we will review, inthis section, the statistical methods that areused most frequently in analyzing researchdata.

Descriptive Statistics

The first thing a researcher needs to do afterobtaining a set of data is to get a feel forgeneral trends. For example, if we were tocollect data on job satisfaction within anorganization, two relevant questions mightbe (1) what is the overall level of job sat-isfaction in the organization, and (2) areemployees very similar in their levels of jobsatisfaction, or do they vary widely? Toanswer the first question, some descriptivemeasure of central tendency would be used.The most commonly used measure of centraltendency is the mean (also called the aver-age), which is calculated by simply addingup all of the scores on a variable and dividingby the total number of scores. Other com-mon measures of central tendency includethe median and mode. The median is thescore on a variable that splits the distributioninto two equal halves. Unlike the mean, themedian is unaffected by the presence ofextremely high or extremely low values.Because of this, the median is useful as asupplement to the mean, in cases in whicha distribution contains extreme scores. Themode is simply the most frequently occurringscore and is typically not very informativeunless there is a very dramatic preference forone response over others.

Measures of central tendency are usefulbecause they provide information about themanner in which variables are distributed.This is important because most statisticalmethods are based on assumptions aboutthe manner in which variables are distrib-uted. Measures of central tendency are alsovaluable when organizational policy makers

are assessing survey results. Figure 2.3, forexample, contains a graphical representationof employee opinion survey data collected byone of the authors. This figure graphicallyrepresents the mean values of four dimen-sions contained on the survey. A quickperusal of this figure indicates relativelylow satisfaction with the levels of communi-cation and fairness in this organization. Onthe other hand, employees in this organiza-tion appear to be committed to the organi-zation and are reasonably satisfied with theirfringe benefits package. While this is cer-tainly not complicated information, it couldbe important to an organization. In this case,the organization used the information as thebasis for interventions to enhance commu-nication and fairness.

In addition to measures of central ten-dency, researchers often want to knowwhether responses are uniform or whetherthere is a great deal of dispersion. The most

FIGURE 2.3Graphical Representation of Mean Levels of FourDimensions Measured in an Employee OpinionSurvey.

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

Communication

Level of SatisfactionLow High

Fairness

Benefits

Commitment

Notes: Communication = Satisfaction with amount of

communication in the organization;

Fairness = Satisfaction with level of fairness in the

organization; Benefits = Satisfaction with current

fringe benefit package; Commitment =

Organizational commitment. Mean values may range

from 1 to 4.


basic measure of dispersion is the range,which is the difference between the highestand lowest value for a particular variable. It isoften useful to compare the observed rangefor a given variable to the possible range. Forexample, if a variable is scaled such that itmay range from 10 to 50 and the observedrange is 30 to 50, this indicates potentialproblems with range restriction.

While the range may be useful in identi-fying problems with range restriction, it isstill a very crude measure of dispersion.More precise and more commonly usedmeasures of dispersion include the varianceand standard deviation. The variance repre-sents the variability of scores around themean. To calculate the variance, you simplysubtract the mean from each score in a dis-tribution, square each value, add up thesesquared values, and divide by the total num-ber of scores. The standard deviation is sim-ply the square root of the variance.

Given the way in which the variance andstandard deviation are calculated, higher val-ues indicate greater dispersion about themean. The standard deviation is also usefulbecause it can be used in converting rawscores to standard scores. A standard scoreis simply the score on a given variable,expressed (in terms of its distance from themean) in standard deviation units. The sim-plest form of standard score is a z-score,which is calculated by subtracting the meanfrom a raw score and dividing the result bythe standard deviation. Standard scores canbe useful in cases in which the researcherwishes to compare a respondent’s scores ondifferent variables that may utilize differentscales of measurement (see Comment 2.6).

A final type of descriptive measure that isused in the analysis of research data is reli-ability. Reliability is defined as the extent towhich a variable is being measured withouterror (Nunnally & Bernstein, 1994). What is

COMMENT 2.6

CONFESSIONS OF A STATISTICAL MINIMALIST

IN HIS INITIAL statement as editor of Journal of

Applied Psychology in 1995, Philip Bobko

referred to himself as a ‘‘statistical minimalist’’

(Bobko, 1995, p. 4) in describing his views on

statistical analysis. What is a statistical mini-

malist? Perhaps the best way to understandthis is to consider more of Bobko’s editorial

statement. Specifically, he advised potential

authors: ‘‘Please look at ‘simple’ statistics, such

as means, standard deviations, correlations,

effect sizes, and so forth. And do not just look

at them; consider them when attempting to

understand and explain what’s going on. I

believe that one can often (usually?) learnmore by looking at these simple statistics with

a critical and understanding eye than one can

learn by computing the newest fashion in

statistics with an amazed eye’’ (p. 4).

The important point that Bobko was trying

to make in this editorial is that even relatively

simple descriptive statistics are important if

one’s goal is to understand their data. A moresubtle message here is that the choice of sta-

tistical methods to use should be driven by the

question being asked, not by the latest fad.

Although not always the case, it is often pos-

sible to answer important research questions

without resorting to overly complex statistical

analyses.

Source: P. Bobko. (1995). Editorial. Journal of Applied

Psychology, 80, 3–5.


considered error, however, depends on theparticular context in which a measure isbeing used. When multi-item measures areused, which is typically the case in organiza-tional research, it is necessary to assess theinternal consistency reliability. A measure ofinternal consistency reliability provides anestimation of the extent to which all itemson a scale are measuring the same attribute.Suppose, for example, we constructed a five-item measure of job satisfaction. If internalconsistency reliability were estimated to bevery high, this would suggest that all fiveitems were measuring the same thing.

In other cases, researchers must provideother reliability estimates. For example, if avariable is going to be assessed at multiplepoints in time, it is important for the re-searcher to show that the measure of thevariable is not strongly impacted by ran-dom fluctuations over time. In this case, anappropriate form of reliability assessmentwould be test-retest reliability, which simplyinvolves administering a measure at two dif-ferent points in time and calculating thecorrelation between these scores. If this cor-relation is high, it suggests that the measureis not strongly impacted by random temporalfluctuations.

Another form of reliability assessment,interrater reliability, may be necessary incases in which multiple raters are utilizedto assess some attribute of a person (e.g.,performance) or the environment (e.g., jobcharacteristics). There are many ways toassess interrater reliability, but they all basi-cally allow the researcher to assess whetherthe ratings provided by different raters aresimilarly ordered. The researcher can alsoassess whether raters agree on the absolutevalue of the ratings. This issue will be dis-cussed in greater detail in the final section ofthe chapter, which deals with aggregationand levels of analysis issues.

Why do researchers need to be con-cerned about reliability? The answer to thatquestion has to do with the nature of meas-urement error. Measurement error, by defi-nition, represents sources of influence onthe measure other than the hypothesizedconstruct. A constant or systematic errorwould be the tendency of a respondent toanswer all items in a way that he or she feelsis socially desirable. A random error wouldbe a momentary distraction causing as re-spondent to respond to an item ‘‘StronglyAgree’’ when he or she really meant to re-spond ‘‘Strongly Disagree.’’ When a meas-ure is reliable it is relatively free of randomerror, though it may still contain consider-able constant error. When a measure isunreliable, however, it contains a great dealof measurement error. This is problematicbecause random error, by definition, is un-related to other variables; thus, reliabilitysets an upper bound on the magnitude ofrelationships between a measure and othervariables.

Tests of Mean Differences

After assessing descriptive measures, re-searchers should hopefully be able to con-clude that there are no major distributionalproblems, and that all variables are measuredwith a minimal amount of error. If this isindeed the case, the next step is to performsome analysis to test whatever hypothesesare being proposed. There are many differenttypes of hypotheses; a common type ofhypothesis involves testing differences inthe mean level of a given variable. For exam-ple, a researcher may hypothesize thatemployees in white-collar jobs have higherorganizational commitment than blue-collaremployees, or that the performance ofgroups that participate in team-buildingactivities is higher than that of groups not


participating. In this section, we cover thetwo most common statistical tests of meandifferences.

Before describing these statistical tests, itis useful to provide a brief overview of thelogic behind tests of statistical significance.Regardless of the statistical test being used, atest of statistical significance essentiallyinvolves establishing a rule for distinguish-ing chance from nonchance outcomes. Allstatistical significance tests begin with theassumption of what is termed the null hypoth-esis, which is another way of saying there isno effect or no relationship between varia-bles. Assuming that the null hypothesis istrue, it is possible for a variety of researchoutcomes to occur simply on the basis ofchance. Thus, the researcher needs somedecisive rule for determining whether a givenresult represents a chance occurrence or alegitimate scientific finding. The standardused most often for distinguishing chancefrom nonchance—the one that has come tobe adopted in the behavioral sciences overthe years—is 5%. Assuming that the nullhypothesis is true, if the probability of aresearch outcome occurring by chance is5% or less, scientists typically conclude thatit is a legitimate scientific finding (e.g., theyreject the null hypothesis). Thus, when thestatement is made that a finding is ‘‘signifi-cant beyond the .05 level,’’ the researcher issaying that it is very unlikely that the findingobserved is a chance occurrence.

When testing mean differences, the sim-plest scenario is testing the differencebetween two groups. For example, a re-searcher may wish to test whether the aver-age age of those who participate in trainingand development activities differs fromthose who choose not to participate. Thestatistic most commonly used in this situa-tion would be a t-test. The magnitude of the tstatistic depends on the absolute difference

between means relative to the level of varia-tion within the groups being compared.Thus, even if the absolute difference betweenthe means is substantial, a high degree ofvariation within the different groups willkeep the t value at a relatively low level,and lead the researcher to conclude thatthere is no meaningful difference betweenthe groups.

There are other instances in organiza-tional research in which the means of morethan two groups must be compared. Forexample, a researcher might want to com-pare the mean level of job satisfaction inseveral different work groups that have andhave not participated in team developmentactivities. In this case, the statistical proce-dure used would be analysis of variance.The general purpose of analysis of varianceis to assess the variation between differentgroups, relative to the variation withingroups. To perform an analysis of variance,it is necessary to calculate several differentvariance estimates or mean-squares. These areused to estimate the variance between groupsand the variance within groups. The actualtest of statistical significance employed inanalysis of variance is the F-test, which issimply a ratio of the variance between groupsto the variance within groups. When an F isstatistically significant, this indicates that theratio of variance between groups to the var-iation within groups is very unlikely to haveoccurred by chance, given the null hypoth-esis. Recall that the same basic logic isemployed with the t-test. If a statisticallysignificant F is found in analysis of variance,this indicates that there is some differenceamong the means in the groups of interest,although it does not tell the researcherwhich means are different. To figure thisout, follow-up tests would be used to assessthe difference within each possible pair ofgroup means.


Given the basic logic behind analysis ofvariance, this statistical procedure can beused a variety of ways. For example, differentforms of analysis of variance can be used toassess (1) the impact of multiple indepen-dent variables, (2) repeated measures ofdependent variables, and (3) the impact ofmultiple dependent variables. Readers inter-ested in more detailed information on anal-ysis of variance procedures should consultKeppel and Zedeck (1989).

Correlation and Regression Analysis

Given the prevalence of cross-sectional fieldsurveys in organizational research, hypoth-eses are often tested by assessing the co-variation among the variables of interest.The most commonly used statistical indexof covariation is the Pearson product-momentcorrelation coefficient. The correlation coeffi-cient can range from þ1.00 to �1.00, buttypically falls in between these values. Thelarger the absolute value of a correlationcoefficient, the greater the degree of co-variation. This degree is often expressed bysquaring the correlation coefficient to obtainthe amount of shared variation between twovariables. For example, if the correlationbetween two variables is .30, they share 9%of their variance in common (e.g., [.30]2).When the sign of a correlation is positive,this simply means that two variables covaryin the same direction. A negative sign, bycontrast, indicates that two variables covaryin opposite directions.

The correlation coefficient is useful intesting many hypotheses in organizationalresearch, but it provides very limited in-formation about causal relationships. Forexample, if job satisfaction were found tobe correlated with job autonomy, this couldmean that high job autonomy causesemployees to be more satisfied with their

jobs. On the other hand, it could also meanthat a high level of job satisfaction causesemployees to see greater levels of autonomyin their jobs. It is also possible that twovariables may be correlated primarilybecause of the influence of a third variable;for example, employees who have highersalaries may be more satisfied and tend tohold jobs with high autonomy. If this is thecase, it is said that the relationship is spurious.

Correlational analysis is also limited bythe fact that only two variables may be exam-ined at a time. In many instances, a research-er may be interested in the extent to whichseveral variables are related to some othervariable of interest. For example, a researchermay be interested in the degree to which pay,length of service, level of performance, age,and job type all contribute to employees’overall satisfaction with their employingorganization. One way to address this ques-tion would be to examine the correlationbetween job satisfaction and each of thesevariables individually. Unfortunately, suchan analysis does not provide the researcherwith information about the extent to whichthis entire set of variables is related.

The statistical procedure that is used toassess the relation of a set of variables (calledpredictors) to another variable (called thecriterion) is multiple linear regression or, sim-ply, multiple regression. Multiple regressionis useful because it provides a quantitativeestimate of the amount of covariationbetween a set of predictors and a criterionvariable. This is assessed by the multiple Rstatistic, which is analogous to the correla-tion coefficient. In most instances, however,researchers report the squared value of mul-tiple R, which serves as a measure of theamount of variance in the criterion variablethat is explained by a set of predictors.

Multiple regression is also useful becauseit allows the researcher to assess the relative


impact of each predictor in explaining thecriterion variable. When a set of predictors isused to estimate a criterion variable, thecriterion is estimated to be a linear functionof the predictor set. The general form of thisequation is:

Y ¼ Aþ B1X1 þ B2X2 þ . . . BkXk

where Y is the criterion variable that is beingpredicted, the Xs represent the predictorvariables, A is a constant, and each B-valuerepresents the weighting of a given predictoror the extent to which it contributes to theprediction of the criterion. The advantage ofusing these statistical weights, as opposed tocorrelations, is that they are calculated in away that takes into account the intercorrela-tions among the other predictor variables inthe set. Thus, B-values in multiple regressionrepresent the unique contribution of a givenvariable to the prediction of some criterionmeasure.

Beyond correlation and regression anal-ysis, many other related methods can beemployed for data analysis. Most of these fallunder the general category of multivariatemethods (e.g., Tabachnick & Fidell, 1996)and, due to their complexity, are not coveredin this chapter. These methods are quiteuseful to the researcher, particularly in fieldinvestigations. Like all statistical methods,they should be used only if necessary to testa given hypothesis.

Meta-Analysis

A final form of statistical analysis that is beingused increasingly in organizational researchis meta-analysis. Meta-analysis involves thequantitative summary of research findingsand is typically used in research domainswhere a considerable number of studies havebeen conducted (Rosenthal, 1991). Forexample, meta-analyses have been con-

ducted on the relation between job satisfac-tion and job performance (Podsakoff &Williams, 1986), the impact of unemploy-ment on well-being (McKee-Ryan, Song,Wanberg, & Kinicki, 2005), and the impactof different leadership styles (Judge, Piccolo,& Illies, 2004). In all three cases, so manystudies have been conducted that it would bedifficult to provide an accurate qualitativesummary of the findings.

Statistically, meta-analysis essentially in-volves averaging effect sizes (e.g., from cor-relation coefficients or differences betweentwo means). Before these effects sizes areaveraged, however, researchers typicallycontrol for a number of statistical artifacts—factors that may lead to differences in thefindings between studies. The most basicstatistical artifact is sample size. Studies withlarger sample sizes need to be weightedmore heavily when averaging correlationscompared to those with smaller samplesizes. Another common statistical artifactcontrolled in meta-analyses is measurementunreliability. Earlier in the chapter, reliabilitywas defined as the degree to which a variableis measured without error. When measure-ment procedures are unreliable, this meansthat they contain considerable error. As wasstated earlier, this is important because it setsa lower boundary on the degree to which avariable can be correlated with other varia-bles. Controlling for unreliability puts allvariables on a level playing field in terms ofmeasurement error.

The other common statistical artifactcontrolled in meta-analyses is range restric-tion. In some studies, correlations betweenvariables may be reduced because the valuesdo not cover the entire possible range. Thismay occur because of a variety of factors(e.g., Johns, 1991), but it always serves tolimit the magnitude of correlations. Whenresearchers control for range restriction, they


are estimating what the correlations wouldbe if the variables of interest were measuredwithout any range-restriction problems.

Once all relevant statistical artifactsare controlled, two important statistics aretypically calculated in meta-analysis. Mostresearchers calculate some overall estimateof effect size between two variables. Thisestimate represents the effect size after con-trolling for the impact of important statisticalartifacts, and it provides a good estimate ofthe true relationship between variables. Theother statistic typically calculated in meta-analysis is the amount of variation in effectsizes that remains after important statisticalartifacts are controlled. Usually, after impor-tant statistical artifacts are controlled, thereis a relatively small amount of variationbetween studies’ findings. However, if thereis still a substantial amount of variation,factors other than statistical artifacts may becontributing to the differences in findingsbetween the studies. Such factors are calledmoderator variables. Some of the more typicalmoderator variables examined in meta-analyses include aspects of the study design(laboratory experiment versus field study),characteristics of the research samples(employees versus college students), andspecific measures used to assess key variables(well-established measures versus measuresdeveloped for one study).

SPECIAL ISSUES INSTATISTICAL ANALYSIS

At this point, readers should have a basicunderstanding of the more typical statisticalmethods used in organizational psychology.The purpose of statistical methodology is tohelp researchers answer questions (i.e., it is ameans to an end), but it has also become avibrant field of inquiry in and of itself. Infact, within organizational psychology, many

researchers focus on statistical and method-ology issues. Because of this focus, severalissues in statistical methodology have sur-faced over the years and have been the sub-ject of inquiry and debate. In this section, webriefly review four important contemporaryissues in the use of statistical methodology inorganizational research.

Statistical Power in OrganizationalResearch

Statistical power refers to the sensitivity ofstatistical tests to detect meaningful treat-ment effects. To use an analogy, one mightthink of the statistical power of different testsin the same way as differences between typesof microscopes. An inexpensive microscopepurchased from a toy store provides somemagnification, but extremely small objects(e.g., viruses) cannot be detected. In con-trast, an expensive electron microscope pro-vides a much higher level of magnificationthat allows for the detection of evenextremely small particles.

Several factors contribute to statisticalpower (Cohen, 1992). One is sample size.All things being equal, larger sample sizesprovide higher levels of statistical power.This is one reason why survey researchersare concerned about nonresponse, and lab-oratory researchers are concerned aboutparticipants not showing up. A second fac-tor impacting power is effect size, or therelative strength of the effect a researcheris trying to detect. There are actually severalways to express effect size, but the easiestway to explain it is based on the size ofcorrelations. Generally speaking, if the truecorrelation between two variables is verysmall, this effect is much harder to detectthan a much larger effect. Smaller-effectsizes require a more powerful ‘‘microscope’’for detection.


A third factor that impacts statisticalpower is the alpha level chosen in statisticalsignificance testing. The alpha level repre-sents the cutoff for distinguishing chancefrom nonchance findings. Recall, from theprevious discussion of statistical significancetesting, that 5% has become the conventionalrule in the behavioral sciences. The reasonthat the alpha level is set so low is to reducethe probability of committing a Type I error,or falsely concluding that one has uncovereda legitimate scientific finding. In an organi-zational setting, an example of committing aType I error would be falsely concluding thata training program had a positive effect onemployee performance. In contrast, a Type IIerror is committed when a researcher fails todetect a legitimate effect when it is present.In the previous example, this would involve

conducting a statistical test and falsely con-cluding that a useful training program hadno impact on employee performance (seeComment 2.7).

As the alpha level becomes more strin-gent (e.g., smaller than 5%), this reduces thechance of committing a Type I error, but alsotends to reduce power and hence increasesthe chances of committing a Type II error. Incontrast, a more liberal alpha level (e.g.,10%) tends to increase power, although thiscomes at the cost of an increase in the prob-ability of committing a Type I error.

A final factor impacting power is meas-urement error. Specifically, higher levels ofmeasurement error are associated with lowlevels of power. This is simply due to theunsystematic nature of measurement error,which was discussed earlier.

COMMENT 2.7

TYPE I VERSUS TYPE II ERROR: WHICH IS THE GREATEST SIN?

GIVEN THE FACT that the alpha level is typically

set at .05 or, in some cases, even .01, one

would assume that committing a Type I error

is a bad thing. Recall that when a Type I error is

made, a researcher concludes that a finding is

scientifically meaningful when it really is not.Why is this bad? From a scientific point of

view, Type I errors are bad because they lead

us down blind alleys, and ultimately may lead

to faulty theories. From a practical point of

view, a Type I error may lead an organization

to spend a considerable amount of money on a

training program that ultimately is not effec-

tive. Given these negative effects of a Type Ierror, we want to minimize the chance that one

will occur, so we set alpha at a very low level.

Unfortunately, in minimizing the chances

of Type I error, we increase the chances of Type

II error. As you recall, Type II error is commit-

ted when a researcher fails to uncover a legit-

imate scientific effect. Is it better to make a Type

II than a Type I error? It really depends on the

situation. Let’s say, for example, that a re-

searcher is testing a drug that could potentially

neutralize the HIV virus. It would obviously be

bad if this researcher were to falsely concludethat this drug was effective (e.g., commit a Type

I error). However, consider the implications of

committing a Type II error in this case. If this

drug is effective, and research does not show it,

a great chance to reduce human suffering has

been missed.

Ultimately, research should be designed to

balance the risks of both Type I and Type IIerrors. To minimize the risk of Type I error,

alpha levels should be set sufficiently low, and

proper statistical procedures should be used.

On the other hand, Type II error can be mini-

mized primarily by employing adequate sam-

ple sizes and reliable measures.


Given the previously described determi-nants of statistical power, let us now considerthe level of statistical power in organization-al research. Mone, Mueller, and Mauland(1996) examined this issue in a meta-analysisof the level of power in 26,471 statisticaltests from 210 research studies conductedbetween 1992 and 1994. These authors alsoexplored common practices with respect tothe assessment of power prior to conductingresearch.

The results of the meta-analysis wererevealing—and, in fact, somewhat troubling.Given that an acceptable level of statisticalpower is considered to be 80% (e.g., there isan 80% chance of detecting a true effect;Cohen, 1992), the authors found that acrossall effect sizes, an acceptable level of powerwas achieved only 50% of the time. Whatthis means is that across all studies in thismeta-analysis, researchers assume a 50%chance of failing to detect a true effect whenit is present. This suggests that many studiesconducted in organizational research areunderpowered.

Low statistical power is especially prob-lematic when researchers are attempting todetect small effect sizes. When Mone et al.(1996) calculated the level of statisticalpower for small effect sizes, it was found thatthe percentage of studies achieving anacceptable level of power was only 10%!That is, the vast majority of studies attempt-ing to detect small effects are grossly under-powered. This is unfortunate because smalleffects are very common in organizationalresearch, due to the vast number of variablesimpacting employees in organizations.

The results of the survey of authors werealso revealing. Perhaps the most importantfinding was that 64% of the authors surveyedreported that they do not perform any type ofpower analysis prior to conducting a study.One reason frequently cited for this was that,

in many cases, researchers have little or nocontrol over sample sizes in field research.Thus, even if a power analysis indicated thata larger sample size would be desirable, itwould not be possible to increase. Manyauthors in this survey also noted that schol-arly journals do not insist on power analysisduring the review process, although there aresome exceptions (e.g., Campion, 1993). Thisis unfortunate because scholarly journalsserve an important gate-keeping function,and insistence on power analysis wouldserve to heighten awareness of this issue.As it stands right now, there are probablymany meaningful effects in organizationalpsychology that go undetected due to lowstatistical power.

Detection of Moderator Variables

Recall from the section on meta-analysis thata moderator variable changes the relation-ship between two other variables (James &Brett, 1984). More specifically, the relation-ship between two variables differs at differ-ent levels of the moderator variable. Inorganizational psychology, many theoreticalmodels contain moderator variables; thus, itis important to understand the statisticalprocedures used for assessing whether ornot moderated relationships exist.

There are actually several ways to testmoderator effects (e.g., see James & Brett,1984), but the most commonly used proce-dure is through the use of multiple regres-sion analysis (Cohen & Cohen, 1983). Inthis procedure, which is known as cross-product regression, the independent variableis first entered into the regression equation.In the next step, the moderator variable isentered. In the final step, the cross-productof the independent variable and moderator isentered. The cross-product term is createdby multiplying the independent variable by


the moderator for each respondent. If thevariation explained by the cross-productterm is statistically significant, a moderatedrelationship is present. This means that therelationship between the independent varia-ble and the dependent variable differs as afunction of the moderator. This is usuallyshown visually by plotting the relationshipat high (one standard deviation above themean) and low (one standard deviationbelow the mean) levels of the moderator.Figure 2.4 illustrates how this is done. Inthis case, self-efficacy moderates the relation-ship between work hours and psychologicalstrain. Notice that, when self-efficacy is low,there is a positive relationship between workhours and psychological strain. In contrast,when self-efficacy is high, there is essentiallyno relationship between these two variables.

The procedure for detecting moderatorvariables is rather straightforward, but, inpractice, the actual detection of moderatorsis difficult, primarily because of low statistical

power. This is because moderator effects aretypically small, since the variance explainedby a moderator effect is that which is left overafter the effects of the independent variableand moderator are taken into account. Poweris also reduced when the independent varia-ble and the moderator are strongly correlatedand, in the case of dichotomous variables(e.g., race, gender), when the proportion dif-fers greatly from 50/50 (Aguinis & Stone-Romero, 1997).

What can be done to increase the powerof moderator tests? Given the previous gen-eral discussion of statistical power, research-ers testing moderator effects should try toemploy large samples and highly reliablemeasures. A somewhat more controversialway to increase power is to increase the alphalevel beyond the conventional.05. Recall thatthe alpha level represents the researcher’sdecision rule for distinguishing chance fromnonchance findings. If a less stringent alphalevel of .10 is adopted, for example, thismeans that results with a 10% or lowerprobability of occurring by chance are con-sidered legitimate treatment effects.

Given the low power associated withmoderator tests, the decision to adopt a lessstringent alpha level would appear to belogical. It is not extremely unusual to findresearchers using alpha levels of .10 in mod-erator tests (e.g., Jex & Elacqua, 1999), butthe practice is not widespread. This is likelydue to the fact that the .05 level is highlyingrained in our thinking. Most studentsare taught that an alpha level beyond .05 is‘‘cheating,’’ and they are extremely reluctantto raise it.

Beyond statistical considerations in thedetection of moderator effects, it is alwaysgood practice to have a solid theoreticalrationale before searching for moderators.Often, moderator variables that are veryintuitively appealing may not be theoretically

FIGURE 2.4Graphical Representation of a ModeratedRelationship

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Low Work Hours High Work Hours

Psy

chol

ogic

al S

trai

n

Low Self-Efficacy

High Self-Efficacy

Source: S. M. Jex and P. D. Bliese. (1999). Efficacy

beliefs as a moderator of the impact of work-related

stressors: A multilevel study. Journal of Applied

Psychology, 84, 349–361. Copyright # 1999 by the

American Psychological Association. Reprinted with

permission.


justified. Statistical methodology will nevercompensate for poor theory development(see Comment 2.8).

Use of Causal Modeling

Over the past 20 years, a statistical techniquethat has become increasingly popular inorganizational psychology—and many otherfields—is causal modeling (James, Mulaik, &Brett, 1982). The logic behind causal model-ing is that the researcher derives a set ofpredictions about how a set of variablesrelate to one another, and tests all of theserelations simultaneously. In practice, this istypically done through the use of either pathanalysis or structural equation modeling. Withpath analysis, the variables that constitute acausal model are the actual variables that aremeasured. This is illustrated in the simplepath model depicted in Figure 2.5. Thismodel proposes that high levels of cognitiveability and work experience lead to highlevels of job knowledge, which in turn leads

to high levels of job performance. Structuralequation modeling is similar to path analysisexcept that the variables comprising thecausal model are latent rather than measuredvariables. A latent variable is a hypotheticalvariable that is purported to cause the inter-relationships among measured variables. Asan example, cognitive ability is a latent var-iable that might lead to a high intercorrela-tion among scores on cognitive ability testsand a structured interview. An example of astructural equation model is contained in

COMMENT 2.8

THE ELUSIVE MODERATOR EFFECT

AS WILL BECOME evident as readers make their

way through this book, many theories in orga-

nizational psychology propose moderator hypo-

theses; that is, certain relationships may holdunder certain conditions, but not under others.

Moderator variables are important in theory de-

velopment because they allow us to specify the

precise conditions under which some phenom-

enon may occur. They also may have a great deal

of practical value, for example, by providing an

organization with guidance about whether there

are certain conditions under which interventionssuch as job redesign may or may not work.

Despite the theoretical and practical value

of moderator variables, they are very difficult

to demonstrate empirically. The is primarily

due to the fact that moderator variables

almost always explain a small portion of the

variance-dependent measures and, as a result,statistical power to detect these effects is often

very low. Thus, in many cases, researchers

propose theoretically sound moderator

hypotheses yet come up empty when they test

for these effects. What can researchers do to

avoid this fate? The most logical steps one can

take to increase statistical power of moderator

tests are: employ large sample sizes, utilizereliable measures, adopt a reasonable alpha

level, and try to cut down on extraneous

sources of variation.

FIGURE 2.5Simple Path Model

JobPerformance

JobKnowledge

WorkExperience

CognitiveAbility


Figure 2.6. The circles are meant to denotelatent variables, and the boxes representmeasured variables. Continuing with theexample from Figure 2.5, scores on a cogni-tive ability test and a structured interview areused as indicators of the latent variable cog-nitive ability, and so on. Notice that this isessentially the same model depicted in Fig-ure 2.5. The only difference is that the pro-posed relationships are among latent ratherthan measured variables.

Once a model is proposed, the researcherseeks to assess whether the model fits theactual data. There are actually severalindexes of model fit (Bentler, 1990), butthe logic underlying all of them is very sim-ilar. When a model is proposed, the re-searcher is placing certain restrictions onthe covariation among the variables of inter-est. Based on these restrictions, an expectedcovariance matrix of relations among varia-bles in the causal model can be calculated.This expected covariance matrix is then com-pared to the actual covariation among thevariables in the proposed model. When amodel is said to ‘‘fit the data well,’’ this meansthat the actual covariation among the varia-

bles closely matches that which would beexpected, based on the proposed relationsamong the variables.

Causal modeling is a powerful tech-nique because it allows the researcher tosimultaneously test all the relations com-prising an entire theoretical model. Withcorrelation and regression analysis, it isusually possible to test only parts or indi-vidual segments of a theoretical model. Theuse of causal modeling, however, has beensomewhat controversial. Some of thesecontroversies are technically beyond thescope of this chapter and are related tothings such as parameter estimation meth-ods and the assessing model fit. Some,however, have questioned whether thistechnique has been overused, and whethermodel tests have been too data driven andnot grounded enough in theory.

Like any statistical technique, causalmodeling is neither good nor bad. If appliedproperly, it can be a very useful part of anorganizational psychologist’s statistical toolkit. Generally speaking, causal modeling ismost powerful when the model being testedhas a strong theoretical base, and there is a

FIGURE 2.6Simple Structural Equation Model

TestScore

InterviewScore

TestScore

SelfRating

Numberof Jobs

Years inProfession

SupervisorRating

SupervisoryRating

JobPerformance

JobKnowledge

WorkExperience

CognitiveAbility


fairly large sample available. It is only at thispoint that a researcher has enough insight topropose the complex set of interrelationsamong variables that comprises most causalmodels. Thus, it is usually not appropriate touse causal modeling early in a theoreticallybased research program.

Aggregation and Levels of Analysis

A recent trend in organizational psychologyis the exploration of variables at multiplelevels of analysis; that is, researchers haveincreasingly become interested in the impactof variables that are conceptualized not onlyat the individual level but also at group andeven organizational levels. Researchers havealso become interested in how variables atdifferent levels of analysis impact each other.This latter type of investigation is known ascross-level analysis.

Exploring multiple levels of analysisobviously presents researchers with someimportant theoretical issues (e.g., Bliese &Jex, 2002; Chan, 1998; Klein, Dansereau, &Hall, 1994). However, with these theoreticalconsiderations come methodological andstatistical considerations as well. Let us firstconsider the issue of aggregation. When dataare aggregated, this simply means that onevalue is used to represent the unit of aggre-gation. An example of this would be usingthe mean level of job satisfaction within awork group to represent ‘‘group-level satis-faction.’’ Note that when a variable is aggre-gated, all individual differences within theunit of aggregation are suppressed.

When is it appropriate to aggregate indi-vidual responses? Generally speaking, re-searchers must be prepared to justifyaggregation on three different levels. First,there must be theoretical justification. Theissue here is whether the variable createdthrough aggregation is theoretically mean-

ingful. In the example provided in the pre-vious paragraph, the researcher would needto make the case that the average level of jobsatisfaction within a work group is a theoret-ically meaningful variable.

If aggregation is theoretically justified,the researcher must also provide somemethodological justification for the deci-sion to aggregate. The most basic methodo-logical question a researcher faces has to dowith choosing the unit of aggregation. Mostresearchers would probably find it accept-able to aggregate the responses of a five-person work group whose members inter-act frequently. There would likely be lessagreement on this issue if one were to aggre-gate the responses of one division of anorganization consisting of 100 members.Unfortunately, there is no hard and fast ruleregarding what is and what is not an appro-priate unit of aggregation; ultimately it comesdown to the variable one is measuring (Bliese& Jex, 2002).

A second methodological issue has to dowith the measurement of variables. In manycases, individual responses are aggregatedbecause items make reference to respondentperceptions of the unit of aggregation. Forinstance, if a researcher were to measureorganizational climate (James & Jones,1974), the items should make reference tothe organization and not the individualsresponding. This suggests that researchersshould make the decision to aggregate beforedata are collected so items can be rewordedappropriately.

Assuming that aggregating is justifiedtheoretically and methodologically, re-searchers must also be prepared to justifyaggregation statistically. In most instancesin which individual responses are aggre-gated, the researcher is doing so in order tomeasure some attribute of the unit of aggre-gation. For example, a researcher may want


to measure the level of cohesiveness in agroup, or the level of trust within an organi-zation. In such cases, it is incumbent uponthe researcher to justify aggregation byshowing some statistical evidence of agree-ment in responses within the unit of aggre-gation. If respondents within a group do notagree on the level of cohesiveness within thegroup, it usually makes little sense to averagetheir responses. There are several ways tomeasure interrater agreement, but the mostfrequently used method has become the rwg

statistic (James et al., 1984).Besides aggregation, the other major

issue confronting researchers exploringmultiple-level issues is statistical analysis.In any research investigation, the choice ofstatistical analysis is driven by the researchquestion being asked. Thus, in some cases,the analysis of multilevel data is relativelystraightforward. For example, if a researcherwere interested in the relation betweengroup cohesiveness and group performance,it would make sense to examine the correla-tion between aggregate-level measures ofboth of these variables. The only drawbackto this approach, of course, is that it greatlyreduces sample size and, hence, statisticalpower.

In other instances, the analysis of multi-level data is more complex because research-ers wish to examine the effects of multiplelevels within the same analysis. For example,a researcher may be interested in estimatingthe relative contribution of individual-levelversus group-level effects. In other cases,researchers may be interested in exploringthe impact of group- or organizational-levelvariables on the relation between individual-level variables. In still other cases, research-ers may wish to examine the behavior of asmall number of research participants overmany occasions (e.g., diary studies); suchdata are multi-level because individuals are

nested within time periods or measurementoccasions. Fortunately, statistical proceduresare available to researchers in order to allowthe analysis of data at multiple levels.

To examine cross-level relations, a statis-tical technique that has become increasinglypopular is random coefficient modeling (Bliese& Jex, 2002; Byrk & Raudenbush, 1992).Random coefficient modeling can be used,for example, to test whether the magnitudeof relations between individual-level varia-bles (represented by regression coefficients)differs as a function of some aggregate-levelvariable. While both of these techniques arevery useful, they are also very complex, andthey require the use of special computersoftware. However, if used appropriately,both can help researchers untangle the com-plexity of multilevel data.

CHAPTER SUMMARY

This chapter explored the methodologicaland statistical foundations of organizationalpsychology. As was shown, organizationalpsychologists have several options when col-lecting data about behavior in organizations.These range from simple observation meth-ods to highly complex quasi-experimentalinvestigations. The most frequently usedtechnique, however, is survey research.

In the collection of data in organizations,several important issues must be considered.For instance, researchers need to be cogni-zant of the limitations of self-report measuresand aware of limits on the generalizability ofresearch findings across research settings.When cross-cultural research is attempted,researchers must be attuned to issues of lan-guage and sampling. A more practical issue issimply gaining access to organizations tocollect research data.

A variety of statistical methods were dis-cussed that can be used to analyze data once

Chapter Summary 55

they are collected. These range from simpledescriptive statistics to more complex corre-lation and regression analysis. The choice ofany statistical technique is dictated by thenature of the question the researcher isattempting to answer.

In the statistical analysis of data, a numberof important issues must be considered. Re-searchers should be aware of the importanceof statistical power and attempt tomaximize it whenever possible. This is par-ticularly true when researchers are interested


PAUL D. BLIESE AND THE IMPORTANCE OF STATISTICS

I’ve always been attracted to the basic tenets of

statistics—the idea that probability can be

used to detect patterns in complex data. What

fascinated me about the area of multilevel

statistics was the idea that researchers and

practitioners could make accurate predictions

about central tendencies of groups even incases where the seemingly same data failed

to make accurate predictions for individuals.

In the Army, this meant that we might be able

to predict the average health and well-being of

groups of soldiers (platoons, companies) even

if we could not necessarily predict the well-

being of individual soldiers in the groups.

I became interested in this topic area earlyin my career when I observed a strong corre-

lation between the average number of hours a

group worked and the average well-being of

the group members. When I analyzed the same

data at the level of individual work hours and

individual well-being, I observed a much

weaker relationship. Over the years, I noted

that patterns such as those involving work

hours and well-being were common in orga-

nizational psychology, and I began to explorethe conditions that caused correlations to differ

across levels (as had many others).

Eventually, this work led me to become

interested in the idea that a variable like work

hours might actually change meaning when it

was averaged within groups. Up to this time,

the dominant idea had been that variables

maintained the same meaning in both theirindividual-level form and their aggregated

form. For an example of a change in meaning,

consider that individual reports of work hours

vary for numerous reasons to include work

requirements, desire to get ahead and work

ethic. When the variable is aggregated by aver-

aging across groups, however, individual dif-

ferences wash out, and one is left with a groupmean that reflects work requirements imposed

on the group. In this way, the two variables

(average group work hours and individual

work hours) have a subtle but important

change in meaning across levels.

What have I learned from all this? I guess

the answer is that taking the time to track down

answers to questions such as ‘‘why do twocorrelations differ?’’ can lead to an entire

career’s worth of work.

LTC Paul D. Bliese

Commander, U.S. Army Medical Research

Unit-Europe.


in demonstrating the effect of moderator var-iables. Complex statistical techniques, suchas causal modeling, can be useful tools toorganizational researchers, provided theyare used judiciously and are based onsound theory. The exploration of multileveldata has become increasingly popular inorganizational psychology in recent years.Researchers conducting multilevel analysismust be prepared to justify aggregation, andmust choose the analytical technique that bestrepresents the substantive issue of interest.


Bliese, P. D. (2000). Within-group agree-ment, non-independence, and reliability:Implications for data aggregatrion and anal-ysis. In K. J. Klein & S. W. Kozlowski (Eds.),Multilevel theory, research, and methods inorganizations (pp. 349–381). San Francisco:Jossey-Bass.

Dane, F. C. (1990). Research methods. PacificGrove, CA: Brooks-Cole.

De LaRosa, G. M. (2006). Towards an under-standing of individual ratings of cohesion within

work units: A multilevel study. Unpublishedmaster’s thesis. Bowling Green State Univer-sity, Bowling Green: OH.

Hays, W. L. (1988). Statistics (4th ed.) FortWorth, TX: Harcourt Brace Jovanovich.

McCall, M. W., Jr., & Bobko, P. (1990).Research methods in the service of discovery.In M. D. Dunnette & L. M. Hough (Eds.),Handbook of industrial and organizationalpsychology (2nd ed., Vol. 1 pp. 381–418).Palo Alto, CA: Consulting PsychologistsPress.

Schaffer, B. S., & Riordan, C. M. (2003). Areview of cross-cultural methodologies fororganizational research: A best practicesapproach. Organizational Research Methods,6, 169–215.

Schaubroeck, J., & Kuehn, K. (1992).Research design in industrial and organiza-tional psychology. In C. L. Cooper & I. T.Robertson (Eds.), International review ofindustrial and organizational psychology, 1992(pp. 99–121). Chichester, England: Wiley.

Tabachnick, B. G., & Fidell, L. S. (1996).Using multivariate statistics (3rd ed.) NewYork: HarperCollins.


Chapter Three

Attraction andSocializationT

he lifeblood of any social organi-zation is people. For example, auniversity must have faculty, anauto manufacturer must havedesign engineers, and a profession-

al baseball team must have players. Thus, toremain viable, all organizations must peri-odically go through the process of bringingin new employees and training them tobecome full-fledged organizational members.To begin this process, organizations must firstattract potential employees and determinewhether their qualifications match organiza-tional needs. This process, however, is notunidirectional—that is, applicants are seekingout and evaluating organizations as potentialemployers. Once employees enter an organi-zation, they must be trained not only to per-form job-specific tasks, but also to learn theculture of the organization. This process isalso bidirectional; that is, organizations dothings to socialize new employees, and newemployees seek out information to help themadjust to their new work environment. Takentogether, this entire process can be viewed asattraction and socialization.

This chapter begins with an examinationof the first stage of the attraction process,namely recruiting, from the perspective ofboth the organization and the applicant. Aswill be shown, organizations use a variety ofmethods to recruit potential employees, anda number of factors can impact the success ofrecruiting efforts. At the same time, appli-cants, or potential employees, evaluate thesuitability of potential employers. In general,potential employees attempt to make somedetermination of the extent to which they fitwith an organization.

The focus of the chapter then shiftsto employee socialization. Once a recruitaccepts employment he or she becomesan official member of an organization. Inorder to become a full-fledged organizationalmember, however, a process of socializationis needed. Organizational psychologists haveexamined the socialization process in aneffort to understand the various tactics orga-nizations use to socialize new employees,determine what employees learn as they aresocialized, and describe the tactics newemployees use to obtain information duringthe socialization process.

The chapter concludes with a discussionof the impact of diversity on employee

59

socialization. In contemporary organiza-tions, it is quite common for new employeesto enter organizations with demographiccharacteristics, experiences, and values thatare far different from those of the majority ofemployees. Because of this, it may be espe-cially difficult for such individuals to be fullysocialized into an organization. Fortunately,there are steps an organization can take todeal with the impact of diversity on thesocialization process.

THE RECRUITMENT PROCESS:AN ORGANIZATIONALPERSPECTIVE

The aim of recruiting is to generate a largepool of highly qualified applicants so that theorganization can select those who stand thebest chance of becoming successful andremaining with the organization for a longperiod of time. In college athletics, for exam-ple, coaches typically spend most of the off-season recruiting highly prized high schoolathletes. Although recruiting is usually notconsidered an organizational topic, it is cov-ered briefly in this chapter because it isstrongly related to socialization. Successfulrecruiting increases the chances that the newemployees an organization selects will fitwell into the culture of the organizationand will be socialized more successfully.

Recruitment Planning

Organizations typically do not recruit newemployees randomly. Rather, an organiza-tion’s recruiting efforts are typically basedon careful planning as to (1) the number ofemployees that will be needed in variousjobs, (2) when these new employees will beneeded, and (3) the present and future sup-ply of potential employees in the labor mar-ket. An organization that understands these

three elements of planning will be able tofocus its recruiting efforts much more effec-tively. According to Cascio (1998), this cru-cial first step in the recruitment process isknown as recruitment planning.

What type of information does an orga-nization need to develop a sound recruit-ment plan? First and foremost, recruitmentplanning should coincide with an organiza-tion’s strategic planning. A strategic plan canbe thought of as an organization’s plan for‘‘where we’re going’’ and ‘‘how we’re going toget there.’’ Strategic planning must be linkedto recruitment planning because strategicplans often have clear implications for staff-ing needs. As an example, let’s say the coachof a professional football team decides toreplace an offensive system that relies heavilyon running plays with one that relies primar-ily on passing. This change in strategy willrequire players with different skills and thuswill have implications for recruiting. Thecoach would want to focus on obtaining ahighly talented quarterback and corps ofreceivers, either through the college draftor by other means (e.g., trades, free-agentsignings).

Another factor that should be consideredin developing a recruitment plan is succes-sion planning. Succession planning involvesmaking projections as to the likelihood ofturnover within various job categories. Thisis often done on the basis of projected retire-ments, but may be based on other factors aswell (i.e., employees in limited-term jobs,employees returning to school). Based onthese projections, an organization can oftengear its recruiting efforts toward attractingindividuals who have the skills necessary toperform the work of those who may be leav-ing the organization. As with any prediction,there is some degree of uncertainty in suc-cession planning. For example, since there isno mandatory retirement age for most

60 Attraction and Socialization

occupations, organizations are often uncertainas to the retirement plans of senior employees.

A third consideration in recruitmentplanning is the skills and abilities ofcurrent employees. Many organizations askcurrent employees to periodically completewhat is known as a skills inventory. A skillsinventory may ask employees to documenttheir job experiences, continuing education(if any), and special skills and competencies.If current employees possess the skills andabilities needed by an organization, there isobviously less need to recruit from outsidesources. This is important because fillingpositions internally has certain advantages(i.e., less adjustment for the employee andless cost for the organization) and may createpositive incentives for employees.

A final piece of information that is usefulfor developing a recruitment plan is someassessment of the supply of labor for variousjob categories. This type of information canoften be obtained relatively easily from gov-ernment agencies, trade associations, and, insome cases, professional organizations. Inthe field of I/O psychology, for example,the Society for Industrial and OrganizationalPsychology (SIOP) collects informationabout the supply of labor in the profession.The basic question an organization is seekingto answer is whether the supply of employ-ees in different job categories is very plentifulor very scarce. For example, the supply ofattorneys in the United States has grown tothe point where they are quite plentiful in thelabor market. In contrast, software develop-ers and computer programmers are in rela-tively short supply.

Labor market information is usefulbecause it will influence the approach anorganization will take in its recruiting efforts,as well as the choice of specific recruitingsources. To fill jobs for which labor is inshort supply, organizations may need to be

highly aggressive in their recruiting effortsand perhaps offer other incentives (e.g., sign-on bonuses) to attract new employees. Suchrecruiting efforts may require the assistanceof executive search firms and may be interna-tional in scope. This is often the case whenorganizations need to recruit high-level exec-utives or highly skilled technical specialists.In contrast, when the supply of labor isplentiful, organizations may be able todevote fewer resources to recruiting efforts,and may adopt a much less aggressive andcostly approach. For example, if manyunskilled manual labor positions are open,organizations may rely on referrals from cur-rent employees or simply select from walk-inapplicants.

Recruiting Methods

Assuming that an organization has devel-oped a sound recruitment plan, the nextstep is to choose some methods of recruiting.A key decision for any organization thatplans to recruit new employees is whetherto invite applications from internal andexternal sources. The primary form of inter-nal recruiting is advertising to currentemployees (i.e., through job postings). Asstated earlier, recruiting internally has manyadvantages. Internal transfers and promo-tions are less expensive than bringing innew employees, may provide positive in-centives for current employees, and mayrequire less training for those employeeswho apply and are accepted. It has also beenshown that employees recruited internallymay be more likely to stay with an organiza-tion (Spector, 2006).

On the other hand, new employeesrecruited from the outside may bring a freshperspective to the organization. Also, someorganizations may be forced to hire outsidersbecause their current employees have not

The Recruitment Process: An Organizational Perspective 61

acquired the skills necessary to perform agiven job.

Compared to a current workforce, ex-ternal recruiting sources are much moreplentiful, as indicated in Table 3.1. Althougha specific recruiting source may be requiredbecause of the nature of the job, some gen-eral comments can be made about recruitingsources. For example, the most frequentlyused recruiting source is some form of adver-tising—typically, in print or electronicmedia. Use of the Internet, in particular,has become the preferred method of recruit-ing for many organizations in the past10 years (Chapman & Webster, 2003).

The recruiting sources listed in Table 3.1indicate considerable variation in cost. Theleast costly recruiting sources are typicallywalk-ins and employee referrals. In additionto their low cost, employee referrals maybe attractive because these applicants typi-cally possess greater knowledge of the orga-

nization than other applicants do. This mayexplain why employees who are referredby organizational members tend to havelower levels of turnover, compared to others(Gannon, 1971; Reid, 1972). An obviousdanger in reliance on employee referrals isthat it may perpetuate nepotism, and the resultmay be an overly homogeneous workforce.

Internet advertising is also a relatively in-expensive way for organizations to advertiseemployment opportunities. According to asurvey of human resource professionals con-ducted by Chapman and Webster (2003), aprimary advantage of the Internet is that itallows an organization to develop a verylarge applicant pool due to its wide availabil-ity. It also allows an organization to screenapplicants more efficiently through onlineapplication procedures, and provides a veryvisually appealing medium for describing theculture of an organization (Braddy, Meade,& Kroustalis, 2006).

TABLE 3.1Typical External Recruiting Sources Used by Organizations

1. Advertising: newspapers (classified and display), technical and professional journals, direct mail,

television, the Internet, and (in some cases) outdoor advertising.

2. Employment agencies: federal and state agencies, private agencies, executive search firms, management

consulting firms, and agencies specializing in temporary help.

3. Educational institutions: technical and trade schools, colleges and universities, coop work/study programs,

and alumni placement offices.

4. Professional organizations: technical societies’ meetings, conventions (regional and national), and

placement services.

5. Military: out-processing centers and placement services administered by regional and national retired

officers’ associations.

6. Labor unions.

7. Career fairs.

8. Outplacement firms.

9. Walk-ins.

10. Write-ins.

11. Company retirees.

12. Employee referrals.

Source: W. F. Cascio. (1998). Applied psychology in personnel management (5th ed.). Upper Saddle River, NJ: Prentice Hall.

Reprinted by permission of Pearson Education, Inc., Upper Saddle River, NJ.


Chapman and Webster (2003) also notesome potential disadvantages associatedwith the use of Internet advertising. Thebiggest of these is that Internet advertisingmay generate such a large applicant pool thatscreening potential applicants becomesextremely labor intensive. Respondents inthis study also noted that using the Internetresulted in somewhat of a ‘‘loss of personal’’touch in the recruitment process.

The most costly recruiting methods usedby organizations are the use of employmentagencies and, to a lesser extent, on-campusrecruiting. It should be emphasized, how-ever, that the cost of a recruiting sourceshould be weighed against other factors.For example, most organizations are willingto incur the cost of employment agencies orexecutive search firms when they must selecthigh-level senior executives. Poor hiringdecisions at this level may cost an organiza-tion millions of dollars. Campus recruiting istypically justified in cases in which organi-zations are seeking entry-level employees infields where the supply of labor is relativelylow (e.g., engineering, computer science,accounting). This method of recruitmentalso allows an organization to convey muchmore detailed information about its culture(Cable & Yu, 2006), which may help appli-cants make more informed decisions aboutwhether or not they would fit in the organi-zation. This could lead to greater retention inthe future.

Other than cost, how else can organiza-tions evaluate the potential usefulness ofdifferent recruiting sources? Two commonlyused indexes are yield ratios and time lapsedata. A yield ratio is simply the total numberof candidates generated by a given recruitingsource (Internet advertising, for example),relative to the number of qualified candi-dates. From an organization’s perspective,an ideal recruiting source is one that delivers

a large number of candidates who are wellqualified for the position the company isattempting to fill. This allows the organiza-tion to be highly selective in making its hir-ing decisions.

Time lapse data represent estimates ofthe time it takes to go from one step to thenext in the recruiting and hiring process. Forexample, organizations may estimate thetime needed for each step that takes placebetween the initial contact with an applicantand the first day of employment with theorganization. Time lapse data help an orga-nization identify bottlenecks in the recruit-ment process that may cause applicants tolose interest (see Comment 3.1). When thosebottlenecks are identified, an organizationcan sometimes take steps to speed up theprocess; however, this is not always possible.For example, recruitment for governmentjobs that require security clearance has tobe quite lengthy to allow for backgroundinvestigations.

THE RECRUITMENT PROCESS:THE APPLICANT’SPERSPECTIVE

From the organization’s perspective, therecruitment process involves trying to ‘‘putour best foot forward’’ in order to attractpotential employees. At the same time, weknow that applicants are trying to determinewhich organizations are most attractive tothem. In this section, we examine how, andon what basis, applicants make such judg-ments about organizations.

In a very general sense, when applicantsevaluate potential employers, they are mak-ing some judgment as to whether they fitwith these organizations. An applicant isreally asking himself or herself: ‘‘Can I seemyself doing this job in this organization?’’This question can obviously be answered on


many levels; thus, some explanation isneeded as to the bases on which applicants’assessments of fit are made. Perhaps theeasiest dimension on which applicants canmake an assessment of fit is whether their

skills and abilities must match the skills andabilities required to perform a given job(Kristof, 1996). For most faculty positionsin universities, for example, a Ph.D. in one’sspecialty is required, and in many cases,

COMMENT 3.1

RESEARCH ON RECRUITING

BECAUSE OF THE importance of recruiting, there

has been considerable research on it over the

years (see Rynes, 1991). One theme is very

evident in recruiting research: The recruiter isnot a key factor in whether an applicant

decides to accept employment with an orga-

nization. Rather, the nature of the job and

other conditions of employment (e.g., salary,

benefits, promotion potential) appear to be

much more important. The one thing about

recruiters that does appear to be important,

however, is their knowledge of the job that anapplicant is seeking. This may be the reason

that organizations often select technical spe-

cialists to recruit in their technical specialty.

Another very clear theme in the recruiting

literature—the way organizations treat appli-

cants during the recruiting process—is impor-

tant. For example, if an organization treats

applicants rudely, or is very lax about keepingthem informed, this approach will turn off

applicants and make them less likely to accept

an offer of employment. Why is this the case?

Most recruiting researchers contend that, dur-

ing the recruiting process, applicants form an

impression of a potential employer. Thus,

when an applicant is not treated well during

the recruiting process, negative signals tell theapplicant what the organization would be like

as an employer.

Most recently, research on recruiting has

shown that the image organizations project to

potential applicants is very important. In much

the same way that consumers form perceptions

of different brands of the same product (e.g.,

different brands of beer), job seekers form

perceptions of different employers. Research

has also shown that these perceptions are an

important determinant of whether or not job

seekers pursue employment with a given orga-nization. The practical implication of these

recent findings is that organizations need to

pay close attention to the image they portray

in recruiting materials, company websites, and

even advertising.

In summary, research has shown that, in

comparison to many other factors, recruiting

does not have a large impact on applicants’decision making. Nevertheless, the recruiting

process is important, largely because, if not

done well, it has the potential to turn off

applicants. Organizations should strive to

employ knowledgeable recruiters who treat

applicants with respect. It is also imperative

that organizations attempt to avoid lengthy

time delays, and to maintain contact withapplicants during the recruitment process.

Organizations must also be aware of the image

they project to potential applicants, as well as

the general public.

Sources: S. L. Rynes. (1991). Recruitment, job choice, and

post-hire consequences: A call for new research direc-

tions. In M. D. Dunnette and L. M. Hough (Eds.), Hand-

book of industrial and organizational psychology (2nd ed.,

Vol. 2, pp. 399–444). Palo Alto, CA: Consulting Psychol-

ogists’ Press; S. L. Rynes, R. D. Bretz, Jr., and B. Gerhart.

(1991). The importance of recruitment in job choice: A

different way of looking. Personnel Psychology, 44, 487–

522; Lievens, F., & Highhouse, S. (2003). A relation of

instrumental and symbolic attributes to a company’s

attractiveness as an employer. Personnel Psychology, 56,

75–102.


some teaching experience is desired. If aperson does not have a Ph.D., he or sheobviously would not fit in such a position.

Assuming that an applicant does pos-sess the necessary job-relevant skills andabilities, what other bases does that appli-cant use to assess his or her fit with a partic-ular organization? According to Schneider’s(1987) Attraction–Selection–Attrition frame-work, applicants are attracted to and stayin organizations with cultures that are com-patible with their personalities. Comparedto judgments about skills, abilities, and cre-dentials, making judgments about an orga-nization’s culture is far more difficult. Sinceapplicants are not organizational members,they typically must rely on secondhand in-formation sources such as company web-sites (Cober, Brown, Levy, Cober, &Keeping, 2003), company recruiting bro-chures (Cable & Yu, 2006), and experiencesthey have had with an organization as aconsumer (Lievens & Highhouse, 2003).Despite the difficulty of making this typeof assessment, research has generally sup-ported the ASA framework; that is, membersof organizations, and even work groups,tend to be rather homogeneous in terms ofpersonality (George, 1990; Jordan, Herriot,& Chalmers, 1991; Schaubroeck, Ganster,& Jones, 1998; Schneider, Smith, Taylor, &Fleenor, 1998).

To simply say that applicants areattracted to organizations with cultures theyperceive to be compatible with their person-alities is a rather imprecise statement. Such astatement begs the question: What aspects ofpersonality, and what aspects of organiza-tional culture? To address this question,Judge and Cable (1997) investigated the rela-tionship between the Big Five personalitytraits (neuroticism, extraversion, opennessto experience, agreeableness, and conscien-tiousness) and job applicants’ attraction to

organizations with different cultural profiles.Organizational culture can be thought of asthe underlying values and basic assumptionsthat guide much of the behavior of organi-zational members. (See Chapter 14 for amore extensive examination of organiza-tional culture.)

The results of this study showed thatapplicants were attracted to organizationswith cultural profiles that were congruentwith their personalities. As an example ofhow this works, consider the personalitytrait of conscientiousness. A person who ishighly conscientious is dependable andachievement oriented, and plans well. Judgeand Cable’s (1997) study showed that thosewho are highly conscientious prefer organi-zations with cultures that can be described ashighly detail oriented, and that place anemphasis on tangible outcomes. This mayvery well be due to the fact that highly con-scientious individuals are meticulous abouttheir work and are likely to produce tangibleoutcomes.

More recent research has supportedJudge and Cable’s (1997) findings, withone qualification. Specifically, there is evi-dence that applicants tend to make judg-ments about organizations in much thesame way that consumers make judgmentsabout different products (e.g., Cable & Tur-ban, 2001). Furthermore, this informationmay be more salient than judgments aboutorganizational culture when decisions aremade whether to pursue employment witha particular organization. According to Lie-vens and Highhouse (2003), this has impor-tant practical implications for recruitmentbecause organizations need to be aware ofhow they portray themselves in advertise-ments and other media such as organizationwebsites.

Another way that applicants may judgetheir fit to a particular organization is on the


basis of commonality in perceived values.Values simply represent things that areimportant to people and organizations. Sup-pose that a person places a strong emphasison the value of individual achievement. It isunlikely that this person would be attractedto an organization that places a strongemphasis on the value of teamwork andcollective achievement. Several studies havein fact demonstrated that applicants areattracted to organizations that they perceiveto have values similar to their own (Chat-man, 1991; Dawis, 1990). The major impli-cation is that an organization must be careful

to convey accurate information to applicantsregarding its values. Obviously, though,applicants base their judgments of an orga-nization’s values on more than just recruitingmaterials. For example, applicants may basesuch judgments on information from others,encounters with the organization (e.g., as acustomer), and the way the organization isportrayed in the media. A broader implica-tion is that organizations must clarify theirvalues and attempt to operate in a way that isconsistent with those values. These findingssuggest that value clarification is also a usefulexercise for applicants (see Comment 3.2).

COMMENT 3.2

VALUE CLARIFICATION: WHAT WOULD YOU WALK THE I-BEAM FOR?

VALUES REPRESENT THINGS, ideas, or goals that are

important to people. For one person, acquir-ing material wealth may be extremely impor-

tant; for another, the most important thing

might be to help other people. There is evi-

dence that when people search for jobs,

careers, and organizations, values play a very

important role. That is, people want their

work lives to be compatible with their values.

Despite the importance of values, manypeople never take the time to seriously clarify

what their values are. However, value clarifi-

cation occurs very quickly when people have

to make choices. Many readers may recall

when they were involved in the process of

selecting a college to attend. For some, being

able to maintain regular contact with family

and friends is highly valued, and thereforetheir choice may be heavily influence by geo-

graphical location. For others, the prestige

of the institution is most highly valued, and

thus they may attend a highly prestigious

college located far from home. Of course

the most difficult situation is one in which

all of one’s choices possess highly valued

attributes.

An example of a more dramatic way to look

at value clarification was encountered by one ofthe authors while attending a training seminar

several years ago. The person leading the semi-

nar described a situation in which an I-beam

approximately six inches wide was placed

between the roofs of two skyscrapers that were

about 50 feet apart. Needless to say, walking

across this I-beam would be extremely danger-

ous. She then asked one of the seminar par-ticipants whether he would walk across this

I-beam if $100,000 were waiting at the other

end. When he quickly responded ‘‘No,’’ she

then asked whether he had any children. When

he replied that he had two sons, ages 5 and 3,

she asked whether he would walk this I-beam if

his 5-year-old son were stranded on it. As you

might guess, his response was now an unequiv-ocal ‘‘Yes.’’ (This man obviously didn’t have

teenagers yet!) Few situations in life require

such dramatic value clarification. However, it

is a good way to begin thinking about what one

really values in life. So the next time you’re

unsure about your values, asking yourself

‘‘What would I walk the I-beam for?’’ might

provide some useful answers.


In addition to personality and values,applicants may make other assessments offit, based on a variety of other factors. Forexample, an applicant may have strong feel-ings about work-family issues, and thus beattracted to organizations that are very pro-gressive regarding work-family initiatives(Cunningham, 2005). Some people seekmembership in organizations for more ideo-logical reasons. As an example, enlistment inthe armed services is due at least partiallyto feelings of patriotism.

ORGANIZATIONALSOCIALIZATION

Assuming that an organization is able toattract a pool of highly qualified applicants,it will obviously utilize some selection pro-cedures, make offers to applicants, and ulti-mately end up with new employees. Whensomeone is hired, a process of socialization isrequired to transform the new outsider em-ployee into a full-fledged organizationalmember. In this section, organizational so-cialization is defined, models of the organi-zational socialization process are reviewed,and tactics used by both organizations andnewcomers during the socialization processare described. The concluding section exam-ines the impact of diversity on organizationalsocialization efforts.

Defining OrganizationalSocialization

Organizational socialization represents theprocess by which an individual makes thetransition from outsider to organizationalmember. What does a person have to learnin order to make this transition successfully?According to Van Maanen and Schein(1979), in the broadest sense, socializationinvolves learning the culture of an organiza-

tion. Thus, socialization is synonymouswith the process of acculturation of new orga-nizational members. Another important partof the socialization of new members is learn-ing the task-related and social knowledgenecessary to be successful members of anorganization (Louis, 1990). Stated differ-ently, socialization is about learning to doyour job effectively and getting along withothers in the organization.

One of the most comprehensive defini-tions of organizational socialization was pro-vided by Chao, O’Leary-Kelly, Wolf, Klein,and Gardner (1994). Their definition, whichcontains six dimensions, encompasses ele-ments of task-related learning, knowledgeof the social climate, and culture transmis-sion. These six dimensions are presented inTable 3.2.

The first dimension proposed is history.As new employees become socialized intoan organization, they gradually becomefamiliar with an organization’s history,which may include long-held customs andtraditions. Many organizations providenewcomers with this information duringtheir initial orientation. New employees atWalt Disney World, for example, learn agreat deal about the life of Walt Disneyand the traditions of the Disney Corpora-tion in their initial training, which is called‘‘Traditions 101’’ (Peters & Waterman,1982).

TABLE 3.2Six Dimensions of Organizational Socialization(Chao et al., 1994)

1. History.

2. Language.

3. Politics.

4. People.

5. Organizational goals and values.

6. Performance proficiency.


The second dimension of socialization islanguage. All organizations utilize some ter-minology and jargon that is more familiar toorganizational members than to outsiders.Some of this language may reflect the dom-inant profession within an organization (e.g.,a law firm), but in many cases it is alsoorganization specific. Perhaps the best exam-ple of organization-specific terminology is inthe military. Newcomers to military organi-zations quickly learn about the reliance onmilitary-specific terminology and acronyms.For example, presentations are referred to asbriefings, and assignments are referred to asmissions. With respect to acronyms, somereaders may recall a hilarious scene inthe movie Good Morning, Vietnam, in whichthe actor Robin Williams manages to squeezeevery possible military acronym into one

sentence. Having worked as contractors forthe U.S. Army in the past, both authors canpersonally attest to the reliance on acronymsin the military (see Comment 3.3).

A third aspect of socialization is politics.As newcomers become socialized into anorganization, they gradually begin to under-stand the politics or unwritten rules that gov-ern behavior within the organization. Forexample, this may involve learning how toget things done, how to obtain desirablework assignments, and who the most influ-ential people in the organization are. Suchthings may appear to be obvious at first, butthey may actually be more complex. In manyorganizations, newcomers often find thatpower and influence are only moderatelyrelated to hierarchical level. For example,clerical employees can sometimes become

COMMENT 3.3

ACRONYMS AND MILITARY CULTURE

ONE OF THE biggest shocks for civilians work-

ing for or with the military is the heavy reli-ance on acronyms in the military. For

example, the person you are working most

closely with is your POC (Point of Contact),

and when someone goes to another location

temporarily, he or she is TDY (Temporary

Duty). Both authors have worked extensively

with the U.S. Army, so we’re both very familiar

with military acronyms. Dr. Britt was an offi-cer in the U.S. Army, so he no doubt was

familiarized during basic training. Dr. Jex first

encountered military acronyms during work

on a year-long research contract for the United

States Recruiting Command (USAREC, of

course) in the early 1990s. Evidently, the peo-

ple we were working with on this project were

concerned about the contractors’ lack of

understanding of military acronyms; they pro-vided a booklet explaining the meaning of all

military acronyms—the booklet was about an

inch thick! However, once everyone working

on the project learned some of the more

important acronyms, they actually became

quite comfortable with this form of commu-

nication.

Why does the military make such extensiveuse of acronyms? Although there is no official

explanation for this, most likely acronyms were

adopted because they facilitate speed of com-

munication, something that might be critical

during an actual military operation. Acronyms

also serve to distinguish the military from other

types of organizations.


very influential because they control the flowof information and access to those at higherlevels of the organizations.

The fourth dimension of socialization ispeople. Most organizational newcomers typi-cally belong to some group or unit, so theymust establish and maintain good workingrelationships with others. This may involveestablishing relationships both within thework group and in the organization as awhole. Such relationships provide friend-ship to newcomers, but they also providevaluable information about the organization(Morrison, 2002). For example, they mayhelp a newcomer to understand the historyand politics of the organization, help them tomaster their job tasks, and help clarify theirroles within their work groups (Morrison,2002). In many universities, for example,this process is facilitated by pairing newfaculty with senior faculty mentors. Thesementoring relationships are important inhelping newcomers to adjust to their newsurroundings, make contacts within the uni-versity, and understand the history of theinstitution.

The fifth dimension is organizational goalsand values. Although members of organiza-tions do not become robots who blindlyfollow orders, they must learn the goalsand values of the organization and, to someextent, assimilate them as their own. Anemployee working for McDonald’s, forexample, must learn to get at least somewhatfired up about the prospect of satisfying cus-tomers. As stated earlier, some of this learn-ing is accomplished in the attraction stagebecause employees tend to be attracted toorganizations that they identify with ideolog-ically. However, applicants typically do nothave a complete grasp of the goals and valuesof an organization until they become regularemployees.

The final dimension of socialization,according to Chao et al. (1994), is perform-ance proficiency. All organizational new-comers must learn to perform their jobsproficiently or they will not be able to main-tain their membership for long. Buildingperformance proficiency is a complex proc-ess that involves developing an understand-ing of one’s job duties, as well as acquiringthe specific skills necessary to perform them.As will be shown later in this chapter, aconsistent theme in the organizational social-ization literature is that this dimension is thetop priority of new employees when theyinitially enter an organization. This is under-standable; rewards and other future oppor-tunities within the organization are oftencontingent on performance.

The Socialization Process: AnOrganizational Perspective

The process of organizational socializationcan be viewed from two distinct perspectives:(1) the organization and (2) the newcomer.When viewed from an organizational per-spective, the focus is on the stages newcomerspass through during the socialization process,and the socialization tactics used by organi-zations to get them through these stages.When viewed from the perspective of thenewcomer, the focus is on the ways in whichnewcomers learn about and make sense oftheir new organizational environment. In thissection, we examine socialization from anorganizational perspective.

Stages of Socialization

Organizational psychologists have tended toview socialization largely in terms of stagesthat new employees pass through during thesocialization process. Feldman (1976, 1981)


proposed what has become the most influen-tial stage model of organizational socializa-tion. This model is presented in Figure 3.1.

The first stage in this model is anticipatorysocialization, which refers to processes thatoccur before an individual joins an organi-zation. This form of socialization typicallyoccurs during the recruitment phase, whenapplicants gather information about theorganization and make some assessment ofwhether they would fit within it. As statedearlier, applicants rely on a number of infor-mation sources to make this assessment (e.g.,recruitment brochures, company representa-tives, company websites, past experiences asa consumer). Although all these informationsources are helpful to applicants, they areobviously not firsthand so the accuracy ofthis assessment may vary.

In some cases, however, anticipatory so-cialization may actually occur much earlierthan the recruitment phase. For example,people often have an opportunity to try outcertain occupations through internships,summer jobs, or other related experiences.According to Feldman (1981), anticipatorysocialization is most valuable when an appli-cant has a realistic picture of the organizationand the job he or she will be performing. Infact, much research has been done on thevalue of realistic job previews (RJPs), whichprepare new employees for the realities ofthe jobs they will be performing and theorganizations they will be working in (e.g.,Wanous, 1989). Related to this, it is alsodesirable if the applicant actually has the skillsand abilities that are congruent with the jobbeing sought, and has needs and values thatare congruent with the organization.

As the newcomer moves into the organi-zation and becomes an official member, theencounter stage begins. According to Feld-man (1981), this stage represents the pointat which the newcomer begins to see the joband organization as they really are. For anumber of reasons, this period may requireconsiderable adjustment. The newcomer mayhave to balance the demands placed on himor her by the organization with familydemands. A new attorney in a large law firm,for example, may find that new associatesare expected to work in excess of 80 hoursper week if they want to eventually becomepartners. This is also the time when thenew employee is learning the demands ofhis or her role within the organization. Often,this simply requires clarification of roleresponsibilities with one’s supervisor, but itmay also involve mediating conflicting roledemands.

After new employees become acclimatedto their new roles, they eventually reach thestage labeled by Feldman as change and

FIGURE 3.1Feldman’s (1981) Model of the Stages ofOrganizational Socialization

Anticipatory Socialization

Encounter

Change and Acquisition

Behavioral Outcomes

Affective Outcomes

Source: D. C. Feldman. (1981). The multiple social-

ization of organization members. Academy of Manage-

ment Review, 6, 309–318. Reprinted by permission of

the Copyright Clearance Center.


acquisition. At this point, the employee hasbecome fairly comfortable with his or hernew role both in terms of performing re-quired job tasks and, perhaps more impor-tantly, adjusting to the culture of theorganization. At this point, an employee isfiring on all cylinders, so to speak. For anattorney, this would be the point at whichhe or she is handling a number of cases and iscomfortable doing so. During the change andacquisition phase, the new employee has alsocome to some resolution regarding roledemands; that is, the employee has gained,from his or her supervisor and coworkers, agood understanding of what is and is notexpected. At this point employees are alsoable to achieve some reasonable balancebetween their work and their personal lives.

To a large extent, when the change andacquisition stage is reached, the new em-ployee has become socialized, at least accord-ing to the model. To assess the extent ofsocialization, Feldman included behavioraland affective outcomes within the model. Ata behavioral level, the extent of socializationcan be assessed by whether an employee iscapable of carrying out his or her role-relatedassignments. For example, we would hardlyconsider the socialization process successfulif an employee were unable to perform his orher job.

A second behavioral index of socializa-tion is the extent to which an employee isspontaneously innovative in carrying outrole responsibilities, and is cooperative withother employees. According to Van Maanenand Schein (1979), when an employee issocialized into a new role, this may take theform of custodianship, content innovation, orrole innovation. A custodial approach requiressimply performing a role exactly as written,with little or no deviation. Most readers haveundoubtedly heard the phrase ‘‘It’s not in myjob description.’’ Content innovation and role

innovation, on the other hand, imply that thenew role occupant may introduce changesinto the content or even into the nature ofthe role. An example of content innovationmight be a physician informing patientsdirectly about the results of lab tests ratherthan having nurses do this. An example ofrole innovation might be expanding the roleof production workers to include not onlyproduct assembly, but also quality controland perhaps even communication with prod-uct end users. Feldman’s model proposes thatputting one’s stamp on the new role beingoccupied is an aspect of socialization.

A third behavioral index of the extent ofsocialization is turnover. If an employeeleaves an organization, one could certainlymake the case that this represents a break-down in the socialization process (Feldman,1981). This is only partially true, however;turnover may occur because of plentiful jobopportunities (Carsten & Spector, 1987), orbecause an employee has exceptional skillsand thus may have opportunities in otherorganizations (Schwab, 1991). It is also pos-sible for an employee to remain in an orga-nization but resist being fully socialized (seeComment 3.4).

Affective outcomes associated with soci-alization refer to things such as attitudestoward work, level of motivation, andinvolvement in one’s job. According to Feld-man’s model, when employees are success-fully socialized, they tend to exhibit higherlevels of job satisfaction, internal work moti-vation, and job involvement. As with turn-over, these outcomes may also be impactedby many factors and are thus imperfect indi-cators of socialization.

Recently, researchers have proposed thatthe most immediate outcome of socializationis to increase an employee’s sense of embedded-ness within the organization. Embeddednessreflects the degree to which employees feel


connected to others in the organization, feel asthough they fit in, and feel as though theywould give something up by leaving (Mitchell,Holtom, Lee, Sablynski, & Erez, 2001).

Socialization Tactics

Feldman’s (1981) model, which describesthe stages employees go through during thesocialization process, has received empiricalsupport (e.g., Feldman, 1976), but it doesnot describe the specific tactics organizationsuse to socialize newcomers. For example,

what exactly does a police department doto break in new recruits after they graduatefrom the training academy? What methodsdoes a major league baseball team use tohelp talented minor league players makethe difficult transition to playing at themajor league level? How does a universityhelp a new professor make the transitionfrom graduate school to faculty status (seeComment 3.5)?

The most comprehensive description ofsocialization tactics was provided by VanMaanen and Schein (1979) in their review

COMMENT 3.4

ORGANIZATIONAL SOCIALIZATION AND CONFORMITY

AS NEWCOMERS BECOME socialized into an orga-

nization, they begin to understand the orga-

nization’s culture. Furthermore, once they

understand an organization’s culture, theybegin to assimilate that culture. Thus, it is

assumed that one of the signs that an organi-

zation is not successful in socializing new

employees is turnover. Those who do not

conform to an organization’s culture end up

leaving that organization. This may be true in

some cases but, in others, nonconformists end

up staying in an organization.Based on what we know about turnover,

there may be situations in which an individual

does not embrace the culture of an organiza-

tion, yet has few other employment options.

The nonconforming employee may simply

learn ways to cope with working in such an

organization. There may also be individuals

who do not embrace the culture of an organi-zation, yet may work there for a variety of

reasons—compensation, geographical prefer-

ences, or simplybecause it’s easier than looking

for another job. Such employees may also find

ways to cope with working for an organization

that they do not fit into.

There may be cases, however, when an

employee does not conform and the organiza-

tion must adapt. If an employee is unusually

talented, or possesses a very rare skill, anorganization may in some cases be forced to

put up with a great deal of nonconformity.

A good example of this can be found in Jeffrey

Pearlman’s 2006 book Love me, Hate me: Barry

Bonds and the Making of an Antihero. In this book

Pearlman describes how the San Francisco

Giants have essentially treated Barry Bonds

like royalty, despite the fact that the star out-fielder is often rude and condescending to fans,

members of the media, and at times even his

own teammates. If Bonds were unable to hit

home runs, it is highly unlikely that the Giants

would put up with such behavior. Employees

in most organizations have far less freedom to

deviate from conventional norms of conduct

than Barry Bonds.

Source: Pearlman, J. (2006). Love me, hate me: Barry Bonds

and the making of an antihero. New York, NY: Harper

Collins Publishers, Inc.


of the organizational socialization literature.According to these authors, socialization tac-tics can be described according to the sixdimensions that are presented in Table 3.3.Note that these are not specific tactics per se,but dimensions that form a very usefulframework for understanding specific tac-tics. As can be seen, organizations may optto socialize new organizational memberscollectively or individually. As an exampleof collective socialization, an organizationmight bring in a group of new recruits andput them through an extensive training

course together. In state police departments,for example, large groups of individuals aretypically hired at the same time, and these

COMMENT 3.5

SOCIALIZATION INTO ACADEMIA

FOR MOST PROFESSORS, socialization begins dur-

ing graduate training and continues into the

first job out of graduate school. Traditionally,

socialization into academia has been a ratherinformal process; newcomers essentially nav-

igate their own way through. In many gradu-

ate programs, advanced students are assigned

to teach courses without a great deal of guid-

ance. In some departments, graduate student

instructors are assigned a faculty mentor they

can go to if they have questions, but they are

pretty much on their own in teaching thecourses. This same basic approach is often

used when graduate students make the tran-

sition into faculty positions. Other than some

very general guidelines, and occasional advice

from a kind senior colleague, most new faculty

are left alone to navigate their way through the

first years of academia.

In recent years, there has been a trend inmany universities to institute formal mentor-

ing programs for new faculty, and for graduate

students seeking to pursue academic careers

(Perlman, McCann, & McFadden, 1999). In

the case of graduate students, formal instruc-

tion is provided in teaching and in working

with students. New faculty mentoring pro-

grams typically involve assigning new faculty

to a more senior faculty mentor. A mentor may

provide advice on things such as teaching,beginning a research program, the tenure proc-

ess, and even navigating university politics.

Do formal mentoring programs produce better

quality faculty? This is a difficult question to

answer because few, if any, programs have

been systematically evaluated. However, one

would assume that most new faculty probably

find such programs helpful. The only potentialdownside to formal mentoring is that if it is too

formal, it may decrease the creativity and indi-

viduality of new faculty. Although there is a

certain amount of comfort in having a senior

colleague there to provide advice in difficult

situations, navigating those difficult situations

alone can result in a great deal of growth and

development for new faculty.

Source: B. Perlman, L. I. McCann, and S. H. McFadden.

(1999). How to land that first teaching job. In B. Perlman,

L. I. McCann, and S. H. McFadden (Eds.), Lessons learned:

Practical advice for the teaching of psychology. Washington,

DC: The American Psychological Society.

TABLE 3.3Van Maanen and Schein’s (1979) Six Dimensionsof Organizational Socialization Tactics

Collective Individual

Formal Informal

Sequential Random

Fixed Variable

Serial Disjunctive

Investiture Divestiture


individuals subsequently attend a trainingacademy as a group or cohort. Among theclear advantages of collective socializationinclude being more economical from theorganization’s perspective and providing op-portunities for newcomers to develop a senseof cohesion and camaraderie among them-selves. Allen (2006), for example, found thatcollective socialization was associated withstronger feelings of embeddedness amongmembers of a financial services firm. Accord-ing to Van Maanen and Schein, a potentialdanger of collective socialization is that it ismost likely to produce only a custodial ori-entation among newcomers; that is, new-comers socialized in this manner may not beparticularly innovative in performing theirroles.

Examples of individual socializationwould include skilled apprenticeship pro-grams and, in a more general sense, mentor-ing. This form of socialization is typicallyused when the information a newcomermust learn is very complex, and when so-cialization takes place over a long period oftime. Compared to group socialization, indi-vidual socialization allows an organizationsomewhat more control over the informationpassed on to the newcomer, and this is morelikely to produce outcomes that are desiredby the organization. For example, Van Maa-nen and Schein (1979) point out that thosesocialized individually are more likely to beinnovative in the way they carry out theirroles, as compared to those socialized collec-tively. It has also been shown that individualsocialization may lead to greater role clarity(Jaskyte, 2005).

Despite the apparent value of individualsocialization, there are some drawbacks. Oneobvious drawback is cost. For a senior man-ager to give one-on-one mentoring to a man-agement trainee, or for a master plumber towork with a journeyman, is time consuming

and expensive. Also, in certain occupations,a custodial role orientation encouraged bycollective socialization is desirable. Forexample, if a police officer does not followproper procedures when making an arrest,the chances of obtaining a conviction may bevery slim. Also, if soldiers do not adherestrictly to agreed-on rules of engagementduring peacekeeping missions, their actionsmay result in violations of internationallaw.

The second dimension depicted in Table3.3 is formal versus informal. Policerecruits’ attendance at a residential trainingacademy is an example of formal organiza-tional socialization (e.g., Van Maanen,1975). Note from the previous discussionthat this is also collective socialization,although all forms of formal socializationneed not be collective. For example, doc-toral students are being socialized into theirchosen professions in the context of a for-mal program of study. Within doctoral pro-grams, however, much of the socializationtakes place during informal interactionsbetween students and their faculty mentors.The most common form of informal social-ization is the very familiar on-the-job train-ing. The new employee is not distinguishedfrom more experienced colleagues, but hisor her initial performance expectations areobviously lower.

According to Van Maanen and Schein(1979), formal socialization tends to beused in situations in which newcomers areexpected to assume new ranks or achieve acertain status in an organization, when thereis a large body of knowledge for newcomersto learn, or when errors on the part of a newemployee may put others (including thenewcomers themselves) at risk. This wouldcertainly apply to law enforcement jobs, aswell as many forms of professional train-ing (e.g., law, medicine, dentistry). Informal


socialization, on the other hand, is mosttypical when it is necessary for a newcomerto quickly learn new skills and work meth-ods, or to develop highly specific practicalskills. This would apply to a wide variety ofworkers, such as convenience store clerks,restaurant employees, and production em-ployees in manufacturing.

Formal socialization assures the organi-zation that all newcomers have a relativelystandardized set of experiences during thesocialization process. In professions such aslaw, medicine, and dentistry, the common-ality in educational programs ensures thatthose entering these professions have a com-mon base of knowledge. A potential draw-back of formal socialization is that it maylead to a custodial approach to one’s role.As stated earlier, this is desirable in somecases. In other cases, however, some innova-tion is desirable even if the role occupant hasto acquire a fairly standard set of facts andknowledge. Medical training, for example,involves the formal acquisition of a great dealof information. This is obviously necessary inorder to diagnose and treat many illnesses.At times, however, physicians may also needto deviate from doing things by the book inorder to provide high-quality care for theirpatients. For example, a patient may have anillness that is a very rare condition and thatrequires an unconventional form of treat-ment. Informal socialization allows employ-ees to develop their own unique perspectiveon their roles, and to introduce changes inthose roles when they are able to performindependently.

Socialization tactics can also be viewed interms of whether they are sequential versusrandom. For example, to become a physi-cian, one is required to complete a clearlydefined sequence of steps: undergraduatetraining, medical school, internship, and res-idency. In contrast, for many management

positions in organizations, socialization ismore random because there is no clearsequence of steps that one must follow.Rather, over time, one gradually acquiresthe skills and experiences necessary toassume progressively higher levels of manage-rial responsibility.

According to Van Maanen and Schein(1979), a sequential approach to socializa-tion is typically used when employees arebeing socialized to move up through a clearlydefined organizational hierarchy. In theArmy, for example, an officer cannot assumethe rank of colonel before passing throughlower-level ranks such as captain, major,and lieutenant colonel. Because of theseclearly defined steps, sequential socializationtends to produce more of a custodial than aninnovative role orientation. In many organi-zations, for example, employees must putin their time at headquarters if they hopeto obtain promotions in the future. Whensocialization is more random, however, newemployees may be exposed to a greater vari-ety of views and opinions regarding theirroles. As a result, such individuals may bemore innovative regarding their specific roleresponsibilities or perhaps even the waytheir roles fit into the organization.

Socialization efforts may also be distin-guished in terms of being fixed versus vari-able. When socialization is fixed, a newcomerknows in advance when certain transitionpoints will occur. In many entry-level man-agement training programs, for example,new employees know in advance that theywill be rotated through the organization fora specific period of time before being granteda permanent assignment. When socializationis variable, the organization does not tell thenew employee when transitions will occur.Instead, the message often given is that a newassignment will be forthcoming ‘‘when we feelyou’re ready to handle it,’’ and no specifics are


given as to how and when readiness will bedetermined.

Fixed socialization patterns are most typ-ically associated with changes in an em-ployee’s hierarchical status. In academicinstitutions, for example, faculty rank isdetermined in this fashion. Typically, a fixednumber of years must be invested before afaculty member can move from assistant toassociate, and, finally, to full professor. Anadvantage of fixed socialization is that it givesan employee greater time to understand hisor her role (Jaskyte, 2005), and tends tomake an employee feel more a part of theorganization (Allen, 2006). It is unlikely,though, that a fixed period of time can bespecified before a newcomer in an organiza-tion is fully accepted and trusted by his or hercoworkers.

According to Van Maanen and Schein(1979), fixed socialization is more likely thanvariable socialization to facilitate innovativerole responses. Variable socialization tendsto create anxiety among new employees, andsuch anxiety acts as a strong motivator towardconformity. Variable socialization also keepsnew employees off balance and at the mercy ofsocializing agents within the organization. Atfirst glance, this may appear ideal from theorganization’s perspective, but it can backfire.If an organization is very arbitrary or vagueabout the speed of a new employee’s careerprogression, highly talented employees maysimply leave for better jobs.

Socialization efforts may also be distin-guished as being serial or disjunctive. Whensocialization occurs in a serial fashion, expe-rienced members groom newcomers toassume similar types of positions in the orga-nization. In most police departments, forexample, recruits fresh from academy train-ing are paired with veteran police officerswho help them to learn the ropes. In additionto fulfilling a training function, serial social-

ization serves to pass on the culture of theorganization from one generation to the next.For example, during the socialization proc-ess, experienced employees often pass on thehistory and folklore of the organization tonewcomers. Disjunctive socialization, incontrast, occurs when new recruits do notfollow in the footsteps of their predecessors,or where no role models are available. Thiswould occur when a new employee occupiesa newly created position, or one that hasbeen vacant for some time.

According to Van Maanen and Schein(1979), serial socialization is more likelythan disjunctive socialization to facilitatesocial acceptance into an organization,which is closely related to the previouslydescribed term embeddedness (Mitchellet al., 2001). In many organizations, it isoften necessary to come up through the ranksin order to be truly accepted by others. Serialsocialization is also useful in situationswhere moving up in the organizational hier-archy requires some continuity in skills,values, and attitudes. In the military, forexample, a person coming from the civilianworld might have the necessary managerialand technical skills to assume a high-levelrank. However, such a person would likelyhave difficulty due to a lack of the under-standing of military culture and traditionsthat is needed for such a position. This isalso true in academic departments when newgraduate students enter each year (Slaughter& Zickar, 2006).

Serial and disjunctive socialization alsodiffer in that serial socialization is more likelythan disjunctive socialization to be associ-ated with a custodial role orientation. Dis-junctive socialization, on the other hand, ismore likely to facilitate innovation. Bothapproaches to socialization, however, carrycertain inherent risks. The custodial roleorientation facilitated by serial socialization


is desirable if the experienced member ofthe organization—the person doing thesocializing—does his or her job well andprojects the correct image of the organiza-tion. If this is not the case, a serial approachto socialization may perpetuate a culture ofmediocrity within the organization. This pro-cess can be seen when professional sportsteams are unsuccessful year after year. Inmany cases, veteran players on such teamspass on a culture of losing to newcomers.

An advantage of disjunctive socializationis that it may allow a newcomer to define hisor her role in a very innovative and originalmanner. This, however, requires consider-able personal initiative on the part of theemployee. An employee who is not highlymotivated, or who perhaps lacks confidence,may flounder if socialized in this manner(Gruman, Saks, and Zweig, 2006). New-comers socialized in this manner may alsobecome influenced by persons in the orga-nization who do not have particularly desir-able work habits. If disjunctive socializationis used, organizations may have to do con-siderable screening during the hiring pro-cess, and carefully monitor those whoparticipate in the socialization process.

The final dimension of organizationalsocialization tactics depicted in Table 3.3 isthe distinction between an investiture versusa divestiture approach. When investiture so-cialization is used, the organization capital-izes on the unique skills, values, andattitudes the newcomer brings to the organi-zation. The organization is telling the new-comer: ‘‘Be yourself’’ because becoming amember of the organization does not requireone to change substantially. Many organiza-tions attempt to communicate this messageduring orientation programs and in a varietyof other ways (e.g., giving employees discre-tion over how they do their jobs). Perhapsthe most powerful way to communicate this

message is simply via the way the newcomeris treated in day-to-day interactions. If a new-comer is punished for displaying his or herindividuality, this suggests that the organi-zation does not want to capitalize on thatemployee’s unique characteristics.

When divestiture socialization is used, anorganization seeks to fundamentally changethe new employee. An organization may wishto make the new employee forget old ways ofdoing things, and perhaps even old attitudesor values. Put differently, the organization isnot building on what the new employeebrings to the job; instead, it seeks a moreglobal transformation. The first year of manyforms of professional training involves a gooddeal of divestiture socialization. During thefirst year of doctoral training in many fields(including psychology), for example, stu-dents are taught to view problems from ascientific perspective and to base their judg-ments on empirical data. For many students,this is a form of divestiture socialization(albeit a mild one) because they typically havenot thought this way prior to entering gradu-ate training. More dramatic examples ofdivestiture socialization are used in organiza-tions such as religious cults, radical politicalgroups, and organized crime families. In thesecases, new members may be required to aban-don all forms of personal identity and givetheir complete loyalty to the organization.

According to Van Maanen and Schein(1979), divestiture socialization tends to beused when recruits first enter an organiza-tion, or when they are striving to gain socialacceptance. For example, a new law-schoolgraduate may dramatically change many ofhis or her attitudes and assumptions duringthe first transition to practicing law. Changesin lifestyle and spending habits may also benecessary in order to gain social acceptanceamong other attorneys in a law firm. Failureto make such changes may lead to social


isolation and perhaps to disillusionmentwith one’s chosen profession.

Perhaps not surprisingly, research hasshow that newcomers in organizations tendto react much more favorably to investituresocialization, both in terms of job attitudesand retention (Allen, 2006; Cable & Parsons,2001). According to Allen, use of investituresocialization tends to make new employeesfeel embedded within an organization, whichmakes them more likely to stay. While theremay be some extreme cases when divestituresocialization must be used (e.g., intelligenceagencies), most organizations are probablymuch better off using an investiture approachto socialization.

Van Maanen and Schein’s (1979) modelhas proven to be quite useful in facilitatingan understanding of the organizational soci-alization process. Importantly, the authorscite considerable empirical support for manyof the propositions in the model. Further-more, in recent years more empirical reseachhas emerged linking specific socializationtactics to a number of socialization outcomes(e.g., Allen, 2006).

Despite its enduring value, a number ofthings about this model must be kept inmind. First, although the tactics representedby each of the models are described asthough they are discreet forms of socializa-tion, in reality they represent opposite endsof a continuum. As an example, most social-ization efforts are neither completely formalnor completely informal—they fall some-where in between. Also, many of the social-ization tactics described in this section occurin combination. An organization may social-ize new recruits individually, using an infor-mal, serial approach. This highlights thecomplexity of the organizational socializa-tion process and suggests a possible reasonwhy it is often difficult to predict the out-comes of socialization.

Finally, despite the complexity of orga-nizational socialization tactics, making themexplicit is quite useful for organizations.If managers are aware of the tactics thatare available for socializing new recruits,the socialization process can be managedmore effectively. Organizations can choosethose methods of socialization that are likelyto provide the most desirable outcomes toboth the organization and the new recruit.

The Socialization Process: ANewcomer Perspective

Despite the value of early theory and researchon organizational socialization (Feldman,1976, 1978; Van Maanen, 1975; Van Maanen& Schein, 1979), this literature had a majorgap. Specifically, socialization was viewedalmost exclusively from an organizationalpoint of view, or as something the organiza-tion does to the newcomer. Thus, very littleresearch focused on how newcomers makesense of the complex maze of technical andinterpersonal information facing them dur-ing the socialization process. There was alsovery little work investigating how new-comers actively seek out information duringthe socialization process despite the factthat we know that new employees do this.To fill this void, the focus of more recentwork on organizational socialization hasshifted quite dramatically to the organiza-tional newcomer. More specifically, organi-zational psychologists have become quiteinterested in how newcomers gather infor-mation about their new organizations andhow they make sense out of this information.

Information-Seeking Tactics

According to Miller and Jablin (1991), new-comers actively seek information duringorganizational socialization, and they do so


in a number of ways. Figure 3.2 presents amodel developed by Miller and Jablin todescribe the complex process of newcomers’information-seeking processes. As can beseen in the first step in this model, one factorthat initially determines information seekingis the newcomer’s perceptions of uncer-tainty. Generally speaking, newcomers putmore effort into information seeking whenthey perceive a great deal of uncertainty inthe environment. Newcomers’ perceptionsof uncertainty depend on a multitude offactors such as the nature of the informationone is seeking, individual differences andcontextual factors, availability of informationsources, and, ultimately, the level of roleconflict and ambiguity one experiences. In

actual organizations, the degree of uncer-tainty varies considerably.

A second factor influencing the choice ofnewcomers’ information-seeking tactics isthe social costs associated with these tactics.Social costs really center on the image new-comers want to project to others in the orga-nization, such as supervisors and coworkers.Most readers have probably had the experi-ence of beginning a new job and havingcoworkers say, ‘‘If you have any questions,just ask,’’ or ‘‘There’s no such thing as astupid question.’’ Although experiencedemployees may be completely sincere inmaking these statements, newcomers maystill feel uncomfortable when they mustrepeatedly ask questions of supervisors or

FIGURE 3.2Miller and Jablin’s (1991) Model of Newcomer Information-Seeking Behavior

Information-Seeking Tactics

Individual Differences andContextual Factors

• Testing Limits• Indirect• Surveillance• Observing

• Disguising Conversations• Third Parties• Overt

Types ofInformation

• Referent• Appraisal• Relational

Key Sources ofInformation

• Supervisors• Coworkers• Relational

Newcomer’sLevels of:

• Role Ambiguity• Role Conflict

Newcomer’sPerceptions of:

• Uncertainty• Social Costs of Information Seeking

Source: V. D. Miller and F. M. Jablin. (1991). Information seeking during organizational entry: Influences,

tactics, and a model of the process. Academy of Management Review, 16, 92–120. Reprinted by permission of

the Copyright Clearance Center.


coworkers. In doing so, one incurs anobvious social cost: appearing incompetentin the eyes of one’s supervisor and/or co-workers. When the social costs of informa-tion seeking are high, newcomers tend to useless overt information-seeking tactics and aremore likely to seek out nonthreatening in-formation sources.

Based on perceptions of uncertainty, andon the social costs of information seeking,newcomers choose from a variety of infor-mation-seeking tactics. The most straight-forward tactic newcomers use to obtaininformation is overt questioning. If a newemployee does not know how to use a copymachine, he or she can simply ask someonehow to use it. Of all the possible information-seeking tactics, overt questioning is clearlythe most efficient. It is also the most likelyto yield useful information and may evenhelp the newcomer to develop rapport withothers. Despite these advantages, newcomersmay incur considerable social costs by usingovert questioning because they run the riskof appearing incompetent and may beviewed as an annoyance by some coworkers.Such costs obviously depend on the numberof times the same question is repeated and,to some extent, the manner in which thequestions are asked. If an employee contin-ues to ask coworkers how to use a copymachine after 6 months on the job, or rudelydemands such assistance, he or she wouldlikely be seen as an annoyance.

Another information-seeking tactic new-comers may use is indirect questioning: notasking someone to provide the exact infor-mation that is needed, but asking a questionthat gets at it indirectly instead. For example,a new employee hired for a sales positionmay eventually want to move into a positionin the organization’s human resources de-partment. As a new employee, this personmay feel uncomfortable directly asking his

or her supervisor about the possibility ofobtaining a transfer. As an alternative, thenew employee may casually ask a questionthis way: ‘‘I have a friend who works for XYZCorporation and he was initially hired as apurchasing agent but eventually transferredinto market research. Does that type of thinghappen much here?’’ By using this approach,the employee reduces the risk of offendinghis or her supervisor by asking what couldbe perceived as an inappropriate question.Unfortunately, this type of question may notgenerate the most accurate information. Inthe newcomer’s organization, transfers frompurchasing to market research may be com-mon, but going from sales to human re-sources is very rare.

A somewhat riskier information-seekingtactic, testing limits, may also be used bynewcomers. This involves creating situationsin which information targets must respond.For example, if a new employee is uncertainabout whether attendance at staff meetings ismandatory, he or she may deliberately skip ameeting one week and await the supervisor’sreaction. If there is no negative reaction, theemployee may presume that attendance isnot mandatory. On the other hand, if his orher supervisor reprimands the employee,this signals that attendance is importantand should be viewed as mandatory. Assum-ing that the employee attends subsequentmeetings, this one infraction is unlikely tohave a negative impact.

Another technique newcomers may useto seek information is through disguised con-versations. This involves initiating a conver-sation with someone for a hidden purpose.A new employee may be uncertain aboutwhether employees in the organization areexpected to bring work home on the week-ends, but is uncomfortable asking about thisdirectly. To obtain the information, the new-comer may strike up a conversation with a


fellow employee about what he or she didduring the weekend. If the fellow employeestates that he or she spent time on a work-related project, this suggests to the new-comer that the organization expects employ-ees to bring work home.

Disguised conversations can be usefulbecause they save the newcomer from havingto ask potentially embarrassing questions ofothers. In the previous example, the new-comer may worry that he or she will be seenas a slacker by fellow employees if bringingwork home on the weekend is the norm.On the other hand, if this is not the norm,the newcomer may worry that he or she willbe perceived as trying to make others lookbad (i.e., a rate buster) if this question isasked directly. The major disadvantage ofdisguised conversations is that the new-comer has little control over the responseof the information source. That is, the fellowemployee in the previous example may bevery vague and not divulge whether he or shespent any time working.

One of the major nonverbal information-seeking tactics used by newcomers is observa-tion. For example, organizational newcomerstypically become keenly aware of the behav-iors that are rewarded and punished in theorganization. Although newcomers will typi-cally utilize observation to obtain many typesof information, they will rely most heavily onthis tactic when the social costs of asking theinformation source directly are high. A newemployee may be uncomfortable directly ask-ing his or her supervisor what is consideredto be outstanding performance. Observingothers may be the safest route to acquiringthis information.

Closely related to observation is theuse of surveillance to gather information.The primary distinction between surveil-lance and observation is that surveillance ismore dependent on retrospective sense mak-

ing, and is more unobtrusive than observa-tion. A newcomer may use surveillance totry to understand organizational norms withregard to the length of the workday. To dothis, he or she may pay close attention to thebehavior of fellow employees near the end ofthe day. The use of surveillance allows thenewcomer to obtain important informationwhile avoiding the social costs of askingwhat may be an embarrassing question(e.g., ‘‘What hours do we work?’’). Unfortu-nately, this is somewhat risky because thenewcomer has no control over the targetunder surveillance. Thus, newcomers tendto use surveillance in situations of extremelyhigh uncertainty. Newcomers will also tendto use surveillance to a greater degree toobtain information from coworkers ratherthan supervisors. Newcomers typically haveless opportunity to obtain information fromsupervisors in this manner, and the behaviorof coworkers has more information valuethan supervisory behavior.

A final information-seeking tactic con-tained in Miller and Jablin’s (1991) modelis the use of third parties, or seeking infor-mation from those other than the primarysource of information. The use of third par-ties actually encompasses several of theinformation-seeking tactics describedpreviously. For example, an employee whois unsure whether a supervisor is pleasedwith his or her performance may directly orindirectly ask coworkers whether they thinkthe supervisor is pleased. Like other indirecttactics, acquiring information in this wayspares an employee potential embarrassment.In the previous example, if the employee’ssupervisor has not been pleased with his orher performance, asking the supervisor aboutthis directly would obviously be uncomfort-able. As with all indirect information-seekingtactics, however, new employees may receiveinaccurate information by not going directly


to the most relevant information source. Inthe authors’ experience, bitter conflicts inorganizations are often started because peo-ple do not go to each other directly to obtaininformation. As anyone who has played thegame ‘‘Telephone’’ knows, secondhand infor-mation may be highly distorted.

Outcomes of Information-SeekingTactics and Socialization

Having described the major information-seeking tactics used by newcomers, the nextissue addressed in Miller and Jablin’s (1991)model is the various outcomes associatedwith information-seeking tactics. At a gen-eral level, different information-seeking tac-tics provide newcomers with informationthat varies in both quantity and quality.According to Miller and Jablin, the qualityof information is reflected primarily in new-comers’ levels of role ambiguity and roleconflict. Role ambiguity simply means thatan employee is uncertain about his or herrole responsibilities. For example, role ambi-guity may result if a supervisor is very unclearabout performance standards.

Role conflict, on the other hand, occurswhen information obtained from differentsources is inconsistent. This might occur,for example, if a newcomer receives mixedmessages from a supervisor and coworkersregarding performance standards. Levels ofrole ambiguity and conflict are typicallyhighest when newcomers rely on indirector covert tactics to acquire information.Because these tactics are far removed fromthe most relevant information source, theyprovide newcomers with the least opportu-nity to verify the accuracy of the informationthey obtain. Given that both role ambiguityand role conflict are associated with negativeoutcomes (e.g., Jackson & Schuler, 1985),organizations need to create an environment

in which newcomers feel comfortable usingdirect information-seeking tactics, such asovert questioning.

Since Miller and Jablin’s (1991) review,there has been considerable research on themany aspects of newcomers’ informationseeking. Ostroff and Kozlowski (1992), forexample, examined the relationship betweenthe types of information acquired duringsocialization, and the use of different infor-mation sources. These authors proposed thatnewcomers use different informationsources to acquire different types of informa-tion. To acquire task-related information, itwas expected that testing (e.g., proposingdifferent approaches to one’s supervisor) orexperimentation (e.g., performing one’s jobtasks in different ways and evaluating theeffects) would be relied on most heavily. Toobtain information about group processes,however, it was expected that coworkerswould be the most useful informationsource. The most important source of infor-mation about roles was expected to be obser-vation of the behavior of others.

This study also examined a number ofoutcomes of the socialization process, aswell as changes in the socialization processover time. New employees who consideredthemselves more knowledgeable about theirjob-related tasks, role demands, group-leveldynamics, and the organization as a wholewere expected to be more satisfied withtheir jobs; be committed to and feel moreadjusted to their organization; experiencefewer stress-related symptoms; and reportlower levels of turnover intent. Over time,knowledge in all areas was expected toincrease. The authors proposed that knowl-edge of the group would initially be greatestbut knowledge of the task would equal itover time. Knowledge of the organization asa whole was expected to be the slowest todevelop.


Based on data collected at two points intime from 219 individuals who had beenbusiness and engineering majors in college,most predictions in this study were sup-ported. For example, observing the behaviorof others, which was used most for acquiringknowledge, was followed by interpersonalsources (coworkers and supervisors), exper-imentation, and objective referents (e.g.,consulting written manuals). Also, as pre-dicted, different information sources wereused, depending on the type of informationrespondents were trying to acquire. For in-formation about the role being performed,respondents relied more heavily on super-visors than on coworkers, but tended to relymore on coworkers for information aboutthe internal dynamics of their work group.To obtain information about the task, exper-imentation was used to a greater extent thaninterpersonal sources such as supervisors orcoworkers.

In terms of knowledge of different do-mains, at Time 1 respondents reported thatknowledge about the group was greater thanknowledge of the task, role, and organiza-tion. This pattern had changed somewhat atTime 2. At this point, knowledge of the taskhad surpassed knowledge of the role andgroup, and knowledge of the organizationremained the lowest. There was only onearea in which knowledge changed from Time1 to Time 2: Respondents reported becomingmore knowledgeable about the task.

When relationships among informationacquisition, knowledge, and outcomes wereexamined, a number of trends emerged. Atboth points in time, acquiring knowledgefrom one’s supervisor was associated withhigher levels of job satisfaction and commit-ment, and lower levels of turnover intent.Interestingly, acquiring knowledge fromcoworkers was associated with high levelsof satisfaction and commitment, and low

levels of stress and turnover at Time 1, butthese relations were not supported at Time 2.This finding suggests that supervisors are aconstant source of information, whereascoworkers may initially be very influentialbut their influence wanes over time. Acquir-ing information from observing others andthrough experimentation was positively re-lated to stress-related symptoms. This maybe due to the fact that observing othersmay provide unclear information and thusmay result in role ambiguity. Acquiring in-formation through experimentation may bestressful because it may often result in fail-ure, at least when job tasks are first beinglearned.

Respondents who believed they pos-sessed more knowledge about all of thedomains reported higher levels of satisfac-tion, commitment, and adjustment. How-ever, the two that stood out as most stronglyrelated to these outcomes were knowledge oftask and role domains. It was also found thatcorrelations were stronger between level ofknowledge and outcomes than they werebetween sources of information and out-comes. The implication is that, for new-comers to feel adjusted, it is important thatthey feel knowledgeable about both theirjob-related tasks and their work-group role.Where this information is acquired is lessimportant than the fact that it is acquired.

When changes in the relationship amonginformation sources, knowledge, and out-comes were examined, it was found thatnewcomers who increased the informationobtained from supervisors over time alsoexperienced positive changes in satisfaction,commitment, and adjustment. This furtherreinforces the importance of the supervisoras an information source during the social-ization process. It was also found that positivechanges in task knowledge were associatedwith positive changes in both commitment


and adjustment and effected a reduction instress. This finding reinforces the impor-tance of task proficiency to the adjustmentof the newcomer.

Ostroff and Kozlowski’s (1992) study hasa number of important implications. Consis-tent with Miller and Jablin’s (1991) model,the results suggest that newcomers use dif-ferent methods to acquire different types ofinformation. The results also clearly showthat supervisors are important informationsources for employees, although newcomersmay initially rely just as much on coworkers.Perhaps the most important lesson from thisstudy is that acquiring task knowledge is ofparamount importance to the adjustment ofnew employees. Thus, organizations need tomake sure that new employees receiveproper training and, in some cases, on-the-job coaching in order to increase their taskknowledge over time. A related implicationis that organizations should not overloadnew employees with ancillary duties.

In another longitudinal study of the so-cialization process, Morrison (1993) col-lected data, at three points in time, from135 new staff accountants. In this study, itwas proposed that newcomers acquire anumber of types of information, most ofwhich were comparable to those in Ostroffand Kozlowski’s (1992) study. For example,Morrison proposed that newcomers acquireinformation on how to perform their job-related tasks. Newcomers also acquire whatMorrison described as referent information, orinformation about one’s role. Newcomersalso must acquire information about howthey are performing their jobs (labeled Per-formance Feedback). In many cases, new-comers need to acquire what may bedescribed as normative information, or infor-mation about the norms within the organi-zation. Finally, newcomers need to acquiresocial information, or information about their

level of social integration into their primarywork group.

In addition to describing the types ofinformation acquired, this study proposedthat there are multiple ways of acquiringeach type of information. Consistent withpast socialization research, it was proposedthat information could be acquired fromone’s supervisor or from an experiencedpeer; through monitoring others’ behavior;by responses to direct inquiries; or fromavailable written sources. Consistent withOstroff and Kozlowski (1992), the dimen-sions of socialization examined were taskmastery, role clarification, acculturation,and social integration.

Newcomers seeking greater amounts oftechnical information and performance feed-back were expected to exhibit higher levelsof task mastery than newcomers seekinglesser amounts of this information. It wasalso expected that newcomers seekinggreater referent information and perform-ance feedback would report experiencinghigher levels of role clarity. With respect toacculturation, it was expected that thiswould be associated with seeking greateramounts of normative information and socialfeedback from others. Finally, social integra-tion was also expected to be highest amongthose seeking greater amounts of normativeinformation and social feedback.

The results of this study partially sup-ported the hypotheses. For example, it wasfound that technical information (from bothsupervisors and peers) and written feedbackwere statistically significant predictors oftask mastery. Interestingly, though, thedirection of the relation between technicalinformation from peers and task mastery wasnegative. This may be due to the fact thatpeers may not always have an adequate mas-tery of the technical information that issought by newcomers.


It was also found that to facilitate roleclarification, newcomers tended to make useof referent information, performance feed-back (through inquiries), and consultants’written feedback. For example, a personnew to a work group may pay attention tocues from group members as to whether hisor her role performance is satisfactory; infor-mally solicit feedback from the supervisor;and take advantage of written feedback fromthe initial performance review. Using theseinformation sources makes sense becausethey are most likely to be relevant to employ-ees’ role-related activities.

Social integration was related primarilyto the use of normative inquiries and mon-itoring activities. This finding suggests thatnew employees may feel uncomfortable ask-ing for direct feedback from either peers orsupervisors, in their efforts to determinetheir level of social integration. Indeed, it isunlikely that most people would feel com-fortable asking fellow employees directlyabout the degree to which they are likedand whether they fit in with the work groups.Written sources of feedback would not pro-vide this type of information either.

Finally, for the acculturation dimension,the only significant predictor was monitor-ing. To some extent, this finding mirrors thefindings with regard to social integration. Tolearn about the culture of the organization, anew employee must primarily observe othersin the organization and how things are done.This is likely due to the complexity of cul-ture, but may also be due to the potentialsocial costs associated with more direct formsof information seeking. Because culture isgenerally taken for granted or internalized(e.g., Schein, 1990), newcomers may run therisk of embarrassment by asking directlyabout things many experienced organiza-tional members consider to be obvious ormundane. Overall, Morrison’s (1993) study,

like that of Ostroff and Kozlowski (1992),suggests that newcomers use a variety ofinformation-seeking tactics, and they usedifferent tactics for acquiring different typesof information.

Unfortunately, neither of the studies justdescribed examined whether informationseeking during the socialization process hasan impact on the success of newcomers. It isclear, for example, that newcomers seek andacquire information, and that they use differ-ent information sources to acquire differenttypes of information. What is less clear, how-ever, is whether employees who increasetheir knowledge over time are ultimatelymore successful than employees who acquireless information.

Chao et al. (1994) addressed these issuesin a longitudinal study of 182 engineers,managers, and professionals, conducted overa 3-year period. Career success in this studywas measured by respondents’ levels of per-sonal income and career involvement. Withrespect to personal income, the only social-ization dimension that was predictive wasknowledge about the politics of the orga-nization. Employees who developed thegreatest knowledge of organizational politicstended to have the highest incomes. This maybe due to the fact that those who become veryknowledgeable about the politics of the orga-nization may be most likely to make thecontacts and form the alliances needed toreach levels in the organizational hierarchythat are associated with high levels of income.

In terms of career involvement, the onlysocialization dimension that was predictivewas knowledge of the goals and values of theorganization. Specifically, those who indi-cated having a greater deal of knowledge ofthe goals and values of the organizationreported higher levels of career involvementthan those with less knowledge. Lookingat this another way, it is difficult for new


employees to become highly involved intheir careers if they are unsure of what theiremployers are trying to accomplish. Takentogether, these findings suggest that certainaspects of socialization may contribute toaffective outcomes (Ostroff & Kozlowski,1992), and other aspects may be moreimportant in determining success.

Another interesting finding from thisstudy was that changes in the socializationdimensions were related to changes in bothmeasures of career success. Thus, for em-ployees to sustain a high level of success overtime, they must continually increase theirknowledge in crucial areas of socialization.This finding suggests that, to sustain a highlevel of success over time, one must neverstop learning. Thus, organizations shouldprovide learning opportunities for employ-ees and, when possible, design work in a waythat allows employees to learn (Parker &Wall, 1998).

Given the shift in focus of recent social-ization research to newcomer information-seeking strategies (e.g., Miller & Jablin,1991), the influence of other socializingagents and methods has been de-emphasized(e.g., Van Maanen & Schein, 1979). As aresult, much less is known about the com-bined effect of newcomers’ information-seeking strategies and the behavior of others(e.g., peers and supervisors) in explainingthe socialization of organizational new-comers. Bauer and Green (1998) examinedthis issue in a very ambitious longitudinalstudy of 205 newcomers, 364 of their cow-orkers, and 112 of their managers. Like pastsocialization research, this study examinednewcomer information seeking, severaldimensions of socialization (feelings of taskproficiency, role clarity, and feelings of beingaccepted by one’s manager), and socializa-tion outcomes such as performance, job sat-isfaction, and organizational commitment.

What makes the study unique, however, isthat the behaviors of managers designed tofacilitate socialization were also examined.Thus, this study addresses the need for adual perspective.

As with the research previously dis-cussed, it was expected that the type of in-formation sought by newcomers andprovided by managers would match social-ization outcomes. For example, it was pre-dicted that task-oriented informationseeking and managers’ clarifying behaviorswould be related to feelings of task profi-ciency and role clarity. For feelings of accept-ance by one’s manager, it was expected thatthe best predicators would be social infor-mation sought by the newcomer, as well asmanagers’ supporting behaviors. For out-comes, it was expected that feelings of taskproficiency would predict performance, andfeelings of acceptance by one’s managerwould be predictive of both job satisfactionand organizational commitment. A final pre-diction examined in this study was that theeffects of both information-seeking tacticsand managerial behavior on socializationoutcomes would be mediated by newcomers’perceived level of socialization.

The results of this study showed that onlymanagerial clarifying behavior at Time 2 pre-dicted role clarity at Time 3. This same resultoccurred for predicting performance efficacyat Time 3. These findings are interestingbecause they seem to contradict recent so-cialization research that has placed sucha strong emphasis on the information-seeking tactics of newcomers. Rather, thesefindings suggest that the behaviors of man-agers are the most important factor, at leastfor these outcomes. With respect to feelingsof acceptance by one’s manager at Time 3,the only variable that was predictive wasmanagers’ supportive behaviors at Time 2.Again, employees’ information seeking had


no impact on this measure. With respect tothe mediational hypotheses, no support wasfound for the mediating role of socializationon the relation between newcomers’ infor-mation seeking and outcomes. There was,however, evidence that feelings of task pro-ficiency and role clarity fully mediated therelationship between managerial behaviorand performance. Role clarity and feelingsof acceptance partially mediated the relationbetween managerial behaviors and organiza-tional commitment. These findings suggestthat behaviors of managers, such as provid-ing clarification and support, have a positiveimpact on things such as newcomers’ per-formance and affective outcomes, but only tothe extent that they facilitate the socializationprocess.

The broader implication of Bauer andGreen’s (1998) study is that the behaviorof individual managers toward new employ-ees is a critical factor in employee socializa-tion. As stated earlier, this study is alsonoteworthy because the recent organiza-tional socialization literature has focusedso heavily on information-seeking tacticsand knowledge acquisition of newcomers.Earlier work on organizational socializationfocused heavily on the organizationalattempts to socialize newcomers. This sug-gests that a more balanced view of organiza-tional socialization is needed—that is,socialization is the result of a complex inter-action between socialization tactics used byorganizations and the information-seekingand sense-making processes of newcomers.Ignoring either the organizational or the new-comer perspective provides a limited pictureof the organizational socialization process.

A final issue regarding the newcomerperspective is the expectations that new-comers bring to the socialization process.As Feldman’s (1981) model showed, thereis a period of anticipatory socialization prior

to newcomers’ formal entry into the organi-zation. One way that prior expectations havebeen examined is through the study of real-istic job previews (RJPs) (Wanous, 1989;Wanous, Poland, Premack, & Davis, 1992).As was stated in the earlier section on recruit-ing, the basic idea behind realistic job pre-views is that, prior to organizational entry,the newcomer is given a realistic preview ofwhat the job will entail, even if some of thisinformation is negative. Despite the intuitiveappeal of RJPs, meta-analyses have shownthat they have a very small impact on turn-over (McEvoy & Cascio, 1985; Reilly, Brown,Blood, & Malatesta, 1981).

Another approach to dealing with new-comers’ expectations is to focus informationat a more general level. For example, Buck-ley, Fedor, Veres, Weise, and Carraher(1998) conducted a field experiment thatevaluated the effect of what they describedas an expectation lowering procedure (ELP)among a sample of 140 employees recentlyhired by a manufacturing plant. The ELPconsisted of lecturing the new employeeson the importance of realistic expectations,and how inflated expectations can lead to anumber of negative outcomes. This studyalso included one condition in whichemployees were provided with a RJP. Thisallowed the researchers to test the impact ofan RJP against the more general ELP.

The results of this study indicated thatboth the RJP and ELP had positive effects.For example, employees in those two con-ditions initially had lower expectations thanthose who received neither intervention,although there was no difference after 6months. Most importantly, lower levels ofturnover and higher levels of job satisfactionwere found in the RJP and ELP conditions,compared to those receiving neither inter-vention. It was also found that expectationsmediated this effect; that is, both RJP and


ELP interventions lowered turnover becausethey first lowered employees’ expectations.

An important implication of this study isthat organizations may not have to developjob-specific realistic previews for newcomersin order to facilitate realistic expectations.Rather, the expectations of newcomers canbe changed to be more realistic by moregeneral interventions of the type conductedby Buckley et al. (1998). From a practicalpoint of view, this is encouraging becausedeveloping RJPs is more time-consuming

than more general interventions such asELP. RJPs must be job-specific; thus, manyRJPs must be developed, depending on thenumber of jobs in an organization. The moregeneral point to be gleaned from this study isthat newcomers do much better when theycome into a new organization with realisticexpectations of both their jobs and theirfuture lives within the organization. Thus,it’s always a good idea to have as muchinformation as possible before choosing ajob or career (see Comment 3.6).

COMMENT 3.6

HOW TO DEVELOP REALISTIC EXPECTATIONS

THERE IS CONSIDERABLE research evidence to

support the value of having a realistic picture

of the job one will be performing, as well as

life in the organization in which one will be

working. Despite the value of realistic expec-

tations, many readers might be wondering

how to gain this type of information while still

in college. Many students do so throughinternships, participation in cooperative edu-

cation programs, and summer employment.

Many university placement offices post (and

keep records of ) these types of jobs at local,

national, and international levels. Recent

research has also shown that college students,

and job seekers in general, make considerable

use of company websites (Braddy, Meade, &Kroustalis, 2006), as well as their experiences

with an organization as a consumer (Lievens

& Highhouse, 2003).

A somewhat less conventional way to

obtain information is to set up an informational

interview with a member of the profession you

wish to pursue, or an employee of the organi-

zation you would like to work for. Thisinvolves simply contacting such an individual

and asking for about 30 minutes of his or her

time. Before the meeting, it’s a good idea to

prepare a list of questions about the profession

or the organization. Although time may not

always permit an informational interview,

professional people are often very willing to

talk about their profession to an eager college

student.

What, then, is the best way to develop

realistic expectations? In this author’s opinion,direct experience with a member of an orga-

nization is most likely to provide the most

accurate and timely information. Despite the

accessibility of company websites, one must

remember that organizations control the infor-

mation on these sites, and therefore they may

provide an overly optimistic picture. It has also

been shown that users of company websitesform impressions of an organization based not

only on the actual content of the website, but

also by it’s design (e.g., how easy it is to use,

how visually appealing it is; Cober, Brown,

Levy, Cober, & Keeping, 2003). Relying on

one’s experiences as a consumer may also lead

to inaccurate information about a job and

particular organizations. For example, manyorganizations are highly customer focused so

they often may treat consumers much better

than they do their employees.


The Socialization Process: AnInteractionist Perspective

As previously stated, early research onemployee socialization was focused pri-marily on the tactics organizations use tosocialize newcomers. The focus then shiftedsomewhat; that is, researchers became inter-ested in understanding the manner in whichnewcomers acquire and use informationduring the socialization. While both typesof research continue, the most recent trendin socialization research has been to under-stand the interaction between organizationaltactics and the methods newcomers use toacquire information. This is an exciting de-velopment in socialization research becauseit represents the most realistic view of whatactually happens when employees first enteran organization. In this section we brieflyreview research that has been guided by thisinteractionist perspective to socialization.

One issue that researchers have recentlyexamined is whether the impact of organiza-tional socialization tactics differs dependingon the degree to which newcomers seekinformation. Gruman et al. (2006), for exam-ple, found in a study of 140 university stu-dents that organizational socialization tacticswere more strongly related to socializationoutcomes among newcomers who exhibitedless feedback-seeking and information-seeking behaviors. Kim, Cable, and Kim(2005) found similar results in a study of279 employee-supervisor pairs from sevenorganizations located in South Korea. Theseresearchers also found that socialization tac-tics had little effect on employees who pro-actively developed strong relationships withtheir supervisors.

Taken together, both of these studiessuggest that organizational socialization tac-tics may in fact not be needed in many cases.When newcomers are highly proactive about

seeking information and feedback, and whenthey develop strong working relationshipswith their immediate supervisors, organiza-tions may simply not need to do a great dealto socialize them.

Another issue that has been examined inrecent socialization research is the manner inwhich employees are socialized into workteams. Given the prevalence of teams in orga-nizations (Kozlowski & Bell, 2003), and thefact that an employee’s ties to his or her workgroup are often stronger than they are to theorganization as a whole (Meyer & Allen,1997), this is clearly an important issue. Asan example of this research, Chen G. (2005)examined the socialization process in 104project teams from three high-technologyorganizations. The most important findingsof this study were that employee empower-ment and the expectations of team memberswere both positively related to newcomerperformance. More specifically, new employ-ees performed the best when they feltempowered and when their teams expectedthem to perform well.

It has also been shown that the manner inwhich team members are socialized into theorganization as a whole may have an impacton the overall effectiveness of teams. Forexample, Oh, Chung, and Labianca (2004)conducted a study of 77 work teams from 11organizations in Korea and found that teamswhose members had more social relation-ships with members of other teams weremore effective than teams whose membershad fewer such relationships. While thisfinding may be somewhat specific to theKorean context, it suggests that employeesshould not only be socialized into a team butalso to the organization as a whole.

In summary, this section provideda small sampling of recent socializationresearch examining the dynamic interactionbetween organizational socialization tactics


and newcomer information seeking. Whilemore research clearly needs to be done, thisresearch has already contributed to a greaterunderstanding of employee socializationthan examining organizational socializationtactics or newcomer information seekingin isolation. Future organizational socializa-tion research will likely follow this interac-tionist framework.

THE IMPACT OF DIVERSITYON ORGANIZATIONALSOCIALIZATION

Given increasing levels of diversity in theworkplace, organizations are understand-ably concerned about how this may impactthe socialization process. Research has in factshown that women and ethnic minorities aremore strongly attracted to organizations theyperceive as managing diversity more effec-tively (Ng & Burke, 2005) and providingmore opportunities to minorities (Perkins,Thomas, & Taylor, 2000). Thus, such con-cerns are well founded.

Jackson, Stone, and Alvarez (1992)reviewed the literature on the impact ofdiversity on socialization into groups andcame up with a number of propositions,many of which are relevant to the broaderissue of organizational socialization. Accord-ing to these authors, the primary dilemmaposed by diversity is that many individualswho are perceived as different must still besocialized and assimilated into organiza-tions. This would be the case, for example,if a female executive were promoted to an all-male top-management group. Not only doessuch an individual tend to stand out, but itmay also be difficult for such individuals tobecome accepted and seen as ‘‘part of theteam.’’ According to Jackson et al., thisoccurs simply because people tend to like,and feel more comfortable around, persons

perceived to be similar to themselves (e.g.,Byrne, 1971).

Because people are attracted to, and feelmore comfortable with, those who are sim-ilar to themselves, those perceived as differentare often at a disadvantage during the social-ization process. According to Jackson et al.(1992), newcomers who are dissimilar areoften less likely to form the social ties andreceive, from experienced organizationalmembers, the feedback necessary to assim-ilate well into organizations. Experiencedorganizational members may not deliber-ately exclude those who are demographicallydifferent, but there is a subtle tendency toshy away from such individuals. This oftenputs women and racial minorities at a dis-advantage because, in many organizations,the most influential members are white males.

How can organizations facilitate the so-cialization of a demographically diverseworkforce? Jackson et al. (1992) suggesteda number of strategies that might help tofacilitate the socialization process. For exam-ple, they recommended that when severalminority employees enter an organization atthe same time, collective socialization proc-esses should be used, if possible. Recall fromVan Maanen and Schein’s (1979) descriptionof organizational socialization tactics thatcollective socialization has the benefit of gen-erating communication and support amongthose socialized in the same cohort. This typeof socialization may help women, racialminorities, and perhaps older employees feelless isolated, and may facilitate the develop-ment of social support networks within theorganization. In professional baseball, forexample, the socialization of players fromLatin American countries is greatly enhancedby the formation of these types of socialsupport networks (see Comment 3.7).

Another recommendation of Jacksonet al. (1992) is for organizations to develop


training programs aimed at both newcomersand established organizational members. Fornewcomers, such training programs mightbe aimed at increasing awareness of someof the problems they may face in the social-ization process, and further helping them todevelop coping strategies. For establishedorganizational members, such training mayhelp to increase awareness of some of thechallenges women and racial minorities facewhen they are being assimilated into theorganization. As Jackson et al. point out,however, such diversity training programsmay have the unintended consequence ofhighlighting the differences rather than thesimilarities between people. It is also possi-ble that if such programs are forced onemployees, they may create less favorableattitudes toward diversity.

A third recommended strategy is the useof valid procedures in the selection of female

and minority employees. Assimilating anynewcomer into an organization will be mucheasier if the individual has the skills andabilities needed to do the job (Ostroff &Kozlowski, 1992). Although socialization offemale and minority employees may initiallybe difficult, organizations are typically prag-matic enough to accept those who are capa-ble of performing their jobs well and makinga positive contribution. This also suggeststhat no one benefits when organizations hireand promote unqualified individuals on thebasis of gender or racial preferences. Thisbecame very obvious to one of the authorsseveral years ago when teaching a coursecomposed primarily of African Americans,many of whom worked in professional posi-tions in the auto industry. When the issue ofracial quotas was discussed during class, thevast majority of these African American stu-dents were strongly opposed to this method

COMMENT 3.7

‘ ‘THE CHAIN’ ’

IMAGINE BEING 17 years old and living on your

own in a foreign country where you don’t

speak the language and have little familiarity

with the culture. Worse yet, you are undertremendous pressure on your job—in fact, if

you don’t perform well you will be sent back

to your native country and more than likely

live in poverty. Sounds like a pretty far-

fetched scenario, right? Actually it’s a pretty

common experience for many talented young

baseball players who come to the United

States from Latin American countries such asthe Dominican Republic, Panama, Venezuela,

and Puerto Rico with the dream of becoming

major league stars.

How do Latin players cope with their

difficult circumstances? Actually, many do

not adjust well and end up returning to their

native countries. For those who do survive, an

important coping mechanism is support from

each other. As Omar Vizquel of the San Fran-cisco Giants explains, ‘‘Latin players tend to

band together in the States and help each other.

Some people call it La Cadena—Spanish for

The Chain. It’s nothing formal; just a feeling

of obligation the older players have toward

the younger ones.’’ Examples of this might

be helping a younger player secure housing,

or simply explaining to them how to order ameal in English.

Source: Vizquel, O. (with Bob Dyer). (2002). Omar! My life

on and off the field. Cleveland, OH; Gray & Company

Publishers.

The Impact of Diversity on Organizational Socialization 91

of addressing past racial discrimination.Most expressed a strong desire to be seenas having made it on the basis of their owntalents, and not because of a government-mandated program.

Finally, performance appraisal and re-ward systems can go a long way towardassimilating female and racial minorityemployees into organizations. For example,managers in organizations should be eval-uated, at least to some degree, on the extentto which they develop all of their subordi-nates. If female and minority employee sub-ordinates continually have a difficult timeadjusting, this should reflect poorly on theevaluation of a manager. If an organizationrewards on the basis of the performance ofwork groups, it is in a group’s best interest tomaximize the talents of all group members,regardless of gender, race, or age. This mayexplain the relative success of the military, atleast in comparison to civilian organizations,in providing opportunities for racial minor-ities (Powell, 1995). Mission accomplishmentis the highest priority in military organiza-tions, and those who contribute positively tothe mission are likely to be rewarded andaccepted, regardless of race.

CHAPTER SUMMARY

In this chapter, we examined the ways thatorganizations attract new organizationalmembers, and the process by which theyare socialized. Organizations utilize a varietyof methods to recruit potential newcomers;the choice of method is dependent on anumber of factors such as the nature of thejob, cost, relative quality of candidates gen-erated, and time considerations. Regardless ofthe method chosen, recruiting research sug-gests that organizations are best served byproviding recruits with accurate information,and treating them with respect and courtesy.

Recruiting obviously is not a one-sidedprocess. Job seekers, who are the targets oforganizational recruiting efforts, evaluate themessages put out by organizations and makesome judgment as to the attractiveness of theorganization. Research suggests that judg-ments of organizational attractiveness aremade primarily on the basis of job seekers’judgments of fit with the organization. Thatis, job seekers make some judgment as towhether several aspects of the organizationfit with their abilities, values, and person-ality. It has also been shown more recentlythat job seekers evaluate organizations inmuch the same way that consumers evaluatedifferent products. The major implication fororganizations is that it is in their best interestto provide an accurate portrayal of their cul-ture to potential employees.

Once an individual is hired, the processof organizational socialization begins.Although many definitions of socializationhave been provided, most see it as the extentto which a new employee is able to do his orher job, get along with members of the workgroup, and develop some understanding ofthe culture of the organization. Organizationsmay use a variety of tactics to socialize orga-nizational newcomers. The choice of tacticsdepends, to a large extent, on the nature of thejob a newcomer will assume in the organiza-tion and the ultimate goals of the socializationprocess.

Like recruiting, socialization is a two-wayprocess. Organizational newcomers activelyseek information about the organization andmay use a variety of tactics in order to obtaininformation. The choice of tactics dependslargely on the level of uncertainty, the natureof the information being sought, and the per-ceived social costs of obtaining it. A consistentfinding in recent socialization research is thatnewcomers initially put their efforts intoobtaining information that will help them to



DEREK CHAPMAN AND THE MECHANISMS UNDERLYING RECRUITMENT

I think one of the main things that attracted

me to doing research on recruiting was how it

seemed to be relevant for just about everyone

and yet we knew very little about it. We all

look for jobs at some point in our lives andhave to make hard choices about what to do

and where to work. Furthermore, every com-

pany has to worry about finding and compet-

ing for good employees so the broad appeal of

recruiting was compelling to me. Despite this

fact, the area seemed to be nearly wide open

for investigation, and I found the opportunity

to explore the unexplored to be very interest-ing. For me, the route to conducting recruiting

research was somewhat indirect.

In the mid 90’s I was doing my Ph.D. in I/O

Psychology under the supervision of Pat Rowe

at the University of Waterloo. Pat had been

doing interview research since the mid 1950’s

and quickly interested me in looking at the

dynamic interpersonal processes in employ-ment interviews. In the course of doing inter-

view research with Pat I noticed that many

applicants did not like the rigorous and highly

valid structured interview formats that most

academics recommend. I became more and

more interested in how applicants reacted to

those interviews as well as how they reacted to

various interview media (telephone, videocon-

ference etc.). I was also curious about howthose reactions affected their impressions of

organizations. This ultimately led me to spend

more and more time thinking about recruiting

and job choice. This interest led me to sign up

for a session with Sara Rynes in the late 1990’s

at a doctoral consortium with a small group

of Ph.D. students interested in recruiting

research. Rynes pointed out that the recruitingfield was still in its infancy, and I was imme-

diately struck by the large number of un-

explored areas that really needed attention.

She recommended a recruiting book by Allison

Barber that had just been published. The com-

bination of Rynes’ encouragement and Barber’s

book motivated me to focus more on my

recruiting research stream.I think I have learned a lot about recruiting

from conducting research, speaking with pro-

fessional recruiters, and from talking to people

who are going through the job choice process.

Oneof the things that has intriguedme themost

has been the role that recruiters play in our job

choice decisions. For a long time the collective

wisdom of researchers had been that recruitersdidn’t matter much—it was all about pay and

location. Some of my work has refuted this

premise by taking a look at the mechanisms by

which recruiters influence our thinking. Rather

than having direct effects on our decisions,

recruiters appear to play a role in influencing

our perceptions of job and organizational

attributes as well as providing us with signalsabout what it might be like to work for their

organization. We also found that recruiters had

the most influence on the best candidates—

those who had to choose among multiple job

offers. Now, with my students at the University

of Calgary, I am looking at integrating the(Continued)


perform their job tasks competently, and thatwill enable them to get along with members oftheir immediate work group. Once they areable to perform their job tasks competently,the focus of information seeking shifts tobroader issues such as the culture of the orga-nization. Recent research has begun to exam-ine the interaction between organizationalsocialization tactics and newcomer informa-tion seeking.

A final issue examined was the impact ofdiversity on the socialization of organiza-tional newcomers. Those perceived as differ-ent by established organizational membersmay face a number of unique challenges inthe socialization process. In the extreme,such individuals run the risk of being margi-nalized and never really fitting in. Organiza-tions can, however, take steps to facilitate thesocialization of older employees, females,and racial minorities. Through facilitatingthe development of support networks, pro-viding training programs, using valid selec-tion procedures, and placing an emphasis onperformance and employee development,organizations can make sure that these indi-viduals are accepted and their talents arefully utilized.


Allen, D.G. (2006). Do socialization tacticsinfluence newcomer embeddedness andturnover? Journal of Management, 32, 237–256.

Barber, A. E. (1998). Recruiting employees:Individual and organizational perspectives.Thousand Oaks, CA: Sage.

Braddy, P. W., Meade, A. W., & Kroustalis,C. M. (2005). Organizational recruitmentwebsite effects on viewers’ perceptions oforganizational culture. Journal of Businessand Psychology, 20, 525–543.

Chen, G. (2004). Newcomer adaptation inteams: Multilevel antecedents and out-comes. Academy of Management Journal, 48,101–116.

Kim, T., Cable, D. M., & Kim, S. (2005).Socialization tactics, employee productivity,and person-organization fit. Journal ofApplied Psychology, 90, 232–241.

Slaughter, J. E., & Zickar, M. J. (2006). Anew look at the role of insiders in the new-comer socialization process. Group & Orga-nization Management, 31, 264–290.

persuasion literature with the recruiting liter-

ature. I believe there is considerable overlap inthese areas that could help us understand what

employers can do to improve recruiting.

I now live and work in Calgary, Canada, a

beautiful city near the Canadian Rockies that is

home to a booming oil and gas industry.

Unemployment here is approaching zero and

finding skilled tradespeople in this city of 1

million people is extremely difficult. With

billions of dollars in oil and gas projects

depending on finding qualified people to dothe work, I get a great deal of satisfaction seeing

how the fundamental research in recruiting

can be used to help real companies develop

scientifically based strategies to compete in one

of the tightest labor markets in North America.

Derek S. Chapman


University of Calgary


Chapter Four

ProductiveBehavior inOrganizationsA

s new employees gradually be-come acclimated to their workenvironments, they eventuallyreach a point where they arecapable of engaging in behavior

that contributes positively to organizationalgoals and objectives. As examples, anaccountant becomes capable of handlingthe tax returns of several clients of anaccounting firm, a retail store employeebecomes capable of operating a cash registerwith minimal supervision, and a scientistbecomes capable of independently carryingout his or her own original research inves-tigations. The behaviors described in theexamples may be thought of collectively asproductive behavior, which is the focus of thischapter.

After thoroughly defining productivebehavior, the chapter shifts to a discussionof job performance. This is, by far, the mostcommon form of productive behavior inorganizations, and organizational psycholo-gists have devoted considerable attention toits study. Much work, for example, has beendevoted to simply understanding what ismeant by job performance, and in determin-ing performance dimensions that are com-mon across jobs.

After addressing models describing thedifferent aspects of basic job performance,we then address the critical issue of how jobperformance has been measured by organi-zational psychologists. The effective meas-urement of job performance is critical if wehope to better understand the predictors ofemployees who excel versus falter at work.Organizational psychologists encounter anumber of challenges in trying to effectively

measure job performance, including suchissues as all employees being given high rat-ings (restriction of range) and the perfor-mance of employees varying over time as aresult of personal and environmental factors.

Having addressed the basic dimensionsof job performance and how to assess thesedimensions, we then move to the importantarea of the causes of job performance. Con-siderable work has been devoted to deter-mining the relative contribution of abilities,skills, motivation, personality, and situa-tional factors in explaining performance dif-ferences across employees. As researchershave found, the interaction among all of

95

these predictors is complex. Fortunately, theamount of research done allows us to drawsome fairly definitive conclusions aboutwhat predicts employees who perform wellversus poorly.

The chapter then addresses the secondmajor form of productive behavior at work,which is when employees do things that arenot required in their formal job descriptions.For example, organizations may at timesneed employees to provide assistance to eachother, even though this activity is not part oftheir formal job descriptions. These types ofbehaviors have been defined as organiza-tional citizenship behaviors (OCBs). Researchinto OCB has focused primarily on under-standing the factors that lead employees toperform OCBs.

Finally, we address a third form of pro-ductive behavior at work: innovation. Forexample, to remain competitive, a computermanufacturer may need employees to consis-tently design new computer models that haveinnovative designs and features. There is con-siderable research on creativity in the generalpsychological literature, but organizationalpsychologists have also examined organiza-tion-specific innovation and creativity. Likeother forms of productive behavior, innova-tion and creativity result from a complexinteraction between characteristics of indi-vidual employees and the organizational envi-ronments in which they work.

DEFINING PRODUCTIVEBEHAVIOR

For the purposes of this chapter, productivebehavior is defined as employee behavior thatcontributes positively to the goals and objec-tives of the organization. When an employeefirst enters an organization, there is a tran-sition period during which he or she is not

contributing positively to the organization.For example, a newly hired managementconsultant may not be generating any bill-able hours for his or her consulting firm.From an organizational perspective, a newemployee is actually a liability because he orshe is typically being compensated duringthis unproductive period. The organizationis betting, however, that, over time, the newemployee will reach a point where his or herbehavior contributes positively to the orga-nization. When productive behavior isviewed in financial terms, it represents thepoint at which the organization begins toachieve some return on the investment ithas made in the new employee. In the sec-tions that follow, we take an in-depth look atthree of the most common forms of produc-tive behavior in organizations: job perfor-mance, organizational citizenship behavior(OCB), and innovation.

JOB PERFORMANCE

Defining Job Performance

Job performance is a deceptively simple term.At the most general level, it can be definedsimply as ‘‘all of the behaviors employeesengage in while at work.’’ Unfortunately, thisis a rather imprecise definition becauseemployees often engage in behaviors at workthat have little or nothing to do with job-specific tasks. For example, in a study ofenlisted military personnel, Bialek, Zapf,and McGuire (1977) found that less thanhalf of the work time of these individualswas spent performing tasks that were partof their job descriptions. Thus, if perfor-mance were defined simply in terms ofemployee behaviors performed while atwork, many behaviors that have no relationto organizational goals would be included(e.g., talking with coworkers about last

96 Productive Behavior in Organizations

night’s game). On the other hand, if jobperformance were confined only to behav-iors associated with the technical aspects ofthe basic tasks employees perform, muchproductive behavior in the workplace wouldbe excluded.

According to Campbell (1990), job per-formance represents behaviors employeesengage in while at work that contribute toorganizational goals. This definition is obvi-ously more precise than simply defining per-formance as all behaviors that employeesperform at work. It is also not too restrictive;job performance is not confined only tobehaviors directly associated with task per-formance. One other important aspect of thisdefinition is that job performance representsbehaviors that are formally evaluated by theorganization as part of the employee’sresponsibilities and duties. This aspect ofthe definition distinguishes job performancefrom the other forms of productive behaviorwe address later in the chapter.

In defining job performance, it is impor-tant that we distinguish it from severalrelated terms. According to Campbell(1990), job performance should be distin-guished from effectiveness, productivity,and utility. Effectiveness is defined as theevaluation of the results of an employee’sjob performance. This is an important dis-tinction because employee effectiveness isdetermined by more than just job perfor-mance. For example, an employee who isengaging in many forms of productivebehavior may still receive a poor perfor-mance rating (a measure of effectiveness)because of performance rating errors, or sim-ply because he or she is not well liked by theperson assigned to do the rating.

Productivity is closely related to both per-formance and effectiveness, but it is differentbecause productivity takes into account thecost of achieving a given level of performance

or effectiveness. For example, two salespeo-ple may perform equally well and ultimatelygenerate the same level of commissions in agiven year. However, if one of these indi-viduals is able to achieve this level of salesat a lower cost than the other, he or shewould be considered the more productiveof the two. A term that is closely related toproductivity, and is often used interchange-ably, is efficiency. This refers to the level ofperformance that can be achieved in a givenperiod of time. If a person is highly efficient,he or she is achieving a lot in a relativelyshort period of time. Given that ‘‘time ismoney,’’ one can consider efficiency a formof productivity. Some organizations, in fact,are highly concerned with efficiency. UnitedParcel Service (UPS), for example, places astrong emphasis on the efficiency of thetruck drivers who deliver packages to cus-tomers.

Finally, utility represents the value of agiven level of performance, effectiveness, orproductivity for the organization. This defi-nition may seem redundant alongside thedescription of effectiveness. Utility is some-what different, however. An employee mayachieve a high level of effectiveness (i.e., theresults of his or her performance are judgedto be positive), but utility still could be low.An organization simply may not place a highvalue on the level of effectiveness achievedby the employee. In large research univer-sities, for example, faculty research produc-tivity and grant writing are typically givenhigher priority than teaching performance.Consequently, it is possible to be deniedtenure at such universities even though oneis a superb teacher.

At first glance, distinguishing among per-formance, effectiveness, productivity, effi-ciency, and utility may appear to be arather trivial exercise. On the contrary, thesedistinctions are extremely important if one is

Job Performance 97

interested in understanding and ultimatelypredicting performance. Many studies inorganizational psychology purport to predict‘‘performance’’ when they are actually pre-dicting ‘‘effectiveness’’ or ‘‘productivity’’ (Jex,1998). Employees typically have more con-trol over performance than they do overeffectiveness or productivity, so studies oftenfail to adequately explain performance differ-ences among employees. This gap may ulti-mately lead to erroneous conclusions aboutthe determinants of performance differences.

Models of Job Performance

Efforts to model job performance are aimedat identifying a set of performance dimen-sions that are common to all jobs. Given thevast number of jobs that exist in the world ofwork, trying to determine a relatively smallnumber of dimensions underlying job per-formance is a challenging task. However,modeling job performance is vitally impor-tant because so much research and practicein organizational psychology centers aroundperformance prediction. A major reason forstudying many of the variables that we do(e.g., motivation, leadership, stress) is theirpotential impact on performance. Eventhough models of job performance containmany different dimensions, two major cate-gories of job performance can be foundacross models: in-role (task) performance andextra-role (contextual) performance (Borman& Motowidlo, 1993; Conway, 1999). In-roleperformance refers to performance on thetechnical aspects of an employee’s job. Forexample, a nurse would be required to per-form specific tasks such as drawing blood,ensuring proper delivery of medication, andso on. Likewise, a truck driver has to knowhow to effectively load and handle cargo,operate complex machinery, and performother technical types of tasks. Extra-role per-

formance refers to nontechnical abilities suchas being able to communicate effectively,exhibiting motivation and enthusiasm atwork, and being a good team member.

The distinction between in-role and extra-role performance can be seen in Campbell’s(1990,1994) comprehensive model of jobperformance. Campbell developed his modelof job performance by analyzing a diverse setof jobs performed by soldiers in the U.S.Army. Based on an analysis of the perfor-mance dimensions of multiple jobs, heargued that performance on all jobs can bebroken down into the eight dimensionslisted in Table 4.1. We would argue thatthe first two dimensions in Campbell’smodel reflect the importance of in-role ortask performance. The first dimension is job-specific task proficiency, and it includesbehaviors associated with the core tasks thatare unique to a particular job. For example,behaviors such as counting money, record-ing deposits, and cashing checks would rep-resent some of the job-specific tasks of abank teller. On the other hand, examplesof the core job tasks of a teacher at a day-care center may include scheduling activ-ities, maintaining discipline, and communi-cating with parents.

The second dimension reflective of in-role performance in this model is nonjob-specific task proficiency. This dimension isrepresented by behaviors that must be per-formed by some or all members of an orga-nization, but that are not specific to aparticular job. For example, the primaryjob-related activities of a college professorare teaching and research in a given substan-tive area (e.g., physics). However, regardlessof one’s specialty, most professors are re-quired to perform common tasks such asadvising students, serving on universitycommittees, writing grants, and occasionallyrepresenting the university at ceremonial


events such as commencement. In themilitary, soldiers must not only be able tocarry out the technical aspects of their job(e.g., fire patriot missiles, ensure adequatequantities of helmets and ammunition), butthey must also be proficient on tasks com-mon to all soldiers (e.g., knowing how toreact upon a chemical weapons attack,knowing how to read a map and navigatein an unfamiliar environment).

Both the first and second dimensions ofCampbell’s model reflect tasks that must beperformed within a given occupational posi-tion and therefore represent in-role perfor-mance. The remaining six dimensionsaddress the extra-role or contextual dimen-sions of performance that tend to cut acrossdifferent jobs.

The third dimension is labeled writtenand oral communication task proficiency.Inclusion of this dimension acknowledgesthat incumbents in most jobs must commu-nicate either in writing or verbally. Forexample, a high school teacher and an at-torney obviously perform very differentjob-specific tasks. Both, however, mustperiodically communicate, both orally andin writing, in order to do their jobs effec-

tively. A high school teacher may need tocommunicate with parents regarding stu-dents’ progress, and an attorney may needto communicate with a client in order toverify the accuracy of information to be con-tained in a legal document such as a trust ordivorce agreement.

The fourth and fifth dimensions arelabeled demonstrating effort and maintainingpersonal discipline, respectively. Demonstrat-ing effort represents an employee’s level ofmotivation and commitment to his or her jobtasks. Regardless of whether one performsthe job of dentist, firefighter, or professionalathlete, it is necessary to exhibit commitmentto one’s job tasks. It may also be necessary attimes to demonstrate a willingness to persistin order to accomplish difficult or unpleas-ant tasks. Professional athletes, at times, mayhave to play through nagging injuries in orderto help their teams. Maintaining personaldiscipline is abiding by specified rules andrefraining from negative behaviors such assubstance abuse or other forms of unproduc-tive behavior. Taken together, these twodimensions essentially represent the degreeto which an employee is a good citizen in theworkplace.

TABLE 4.1Campbell’s (1990, 1994) Model of Job Performance According to the In-Role Extra-Role Distinction

Performance Dimensions Description

In-Role

1. Job-Specific-Task Proficiency Technical aspects of job performance

2. Non-Job-Specific Task Proficiency Common tasks performed by different employees

Extra-Role

3. Written and Oral Communication Ability to write and communicate effectively

4. Demonstrating Effort Going the ‘‘extra mile’’ at work

5. Maintaining Personal Discipline Refraining from negative behaviors, following through on tasks

6. Facilitating peer and team performance Being a good team member; working well with other members

7. Supervision/Leadership Effectively supervising and leading others

8. Management/Administration Effectively organizing and keeping track of critical information

Job Performance 99

The sixth dimension is labeled facilitatingpeer and team performance. One aspect of thisdimension is the degree to which anemployee is helpful to his or her coworkerswhen they need assistance. This couldinvolve assisting a coworker who is havingtrouble meeting an impending deadline, orperhaps just providing encouragement orboosting the spirits of others. This dimen-sion also represents the degree to which anemployee is a team player, or is working tofurther the goals of his or her work group. AsCampbell (1990) points out, this dimensionwould obviously have little relevance if oneworked in complete isolation. Today, whenso many companies place strong emphasison teamwork, working alone is more theexception than the rule (see Comment 4.1).

The seventh and eighth dimensions arelabeled supervision/leadership and manage-ment/administration, respectively. Both ofthese dimensions represent aspects of jobperformance that obviously apply only tojobs that carry some supervisory responsibil-ities. Whether one is a supervisor in a retailoutlet, a hospital, or a factory, certain com-mon behaviors are required. For example,supervisors in most settings help employeesset goals, teach employees effective workmethods, and more generally attempt tomodel good work habits. Many supervisorypositions also require a multitude of ad-ministrative tasks such as monitoring andcontrolling expenditures, obtaining addi-tional resources, and representing one’s unitwithin an organization.

When we consider each of these dimen-sions of job performance, it becomes clearthat all eight dimensions would not be rele-vant for all jobs. In fact, Campbell (1990)argued that only three (core task proficiency,demonstrating effort, and maintenance ofpersonal discipline) are major performancecomponents for all jobs. This model is still

quite useful because it provides a commonmetric for examining performance acrossjobs. For example, using this model, wecould compare employees from two com-pletely different jobs on the dimension ofdemonstrating effort. In addition, the perfor-mance of different types of jobs could becompared across the dimensions. Havingsuch a common metric is tremendously help-ful in trying to understand the general deter-minants of job performance.

A second model of job performance wasproposed by Murphy (1994). His model wasspecifically developed to facilitate an under-standing of job performance in the U.S.Navy, but the performance dimensionsare also relevant to many civilian jobs. Ascan be seen in Table 4.2, this modelbreaks performance down into four dimen-sions instead of eight. The first of these islabeled task-oriented behaviors, which closelymirrors the job-specific task proficiencydimension in Campbell’s (1990, 1994)model. It is also reasonable to assume that,for supervisory jobs, this label would includethe dimensions related to supervision andmanagement/administration. In essence, thisrepresents performing the major tasks asso-ciated with one’s job. This dimension is alsothe only component of Murphy’s model thatrefers explicitly to in-role performance. Theremaining dimensions refer to extra-role per-formance. The second dimension, labeledinterpersonally oriented behaviors, representsall of the interpersonal transactions thatoccur on the job. These might include a retailstore clerk answering a customer’s question,a nurse consulting a doctor about a patient’smedication, or an auto mechanic talking to aservice manager about a repair that must bedone on a car. Because many interpersonaltransactions in the workplace are task-related, this dimension mirrors facilitatingpeer and team performance in J. Campbell’s


TABLE 4.2Murphy’s (1994) Model of Job Performance According to the In-Role Extra-Role Distinction

Performance Dimensions Description

In-Role

1. Task-Oriented Behaviors Performing major tasks associated with job

Extra-Role

2. Interpersonally oriented Behaviors All interpersonal transactions that occur on job

3. Down-time Behaviors Behaviors outside of work that affect job performance (e.g. drug/alcohol

use, extra jobs)

4. Destructive/Hazardous Behaviors Safety violations, sabotage, accidents

COMMENT 4.1

BEING A GOOD TEAM MEMBER

OF THE EIGHT dimensions of job performance

described in Campbell’s (1990) model, one of

the most interesting, and potentially most

important, is ‘‘Facilitating peer and team per-formance.’’ One obvious reason is that more

and more organizations are making use of

teams for both projects and even as a basis

for organizational structure. Given this greater

use of teams, it is not surprising that much

recent organizational research has focused on

team effectiveness. However, one aspect of

team effectiveness that has not been givengreat attention is identifying the characteris-

tics of a good team member. According to

Susan Wheelan, in her book Creating Effective

Teams: A Guide for Members and Leaders, there

are a number of behavioral characteristics of

effective team members. These include:

� not blaming others for group problems� encouraging the process of goal, role, and

task clarification� encouraging the adoption of an open

communication structure� promoting an appropriate ratio of task

and supportive communications� promoting the use of effective problem-

solving and decision-making procedures� encouraging the establishment of norms

that support productivity, innovation,

and freedom of expression

� going along with norms that promote

group effectiveness and productivity� promoting group cohesion and cooper-

ation� encouraging the use of effective conflict-

management strategies� interacting with others outside of the

group, in ways that promote group inte-

gration and cooperation within the larger

organizational context� supporting the leader’s efforts to facilitate

group goal achievement

This list is obviously not meant to beexhaustive, but it illustrates the specific

behaviors that contribute to effective team

performance. As is evident from the list,

most of these behaviors transcend technical

specialties and even organization types.

This is consistent with Campbell’s notion

that there is a general set of performance

dimensions.

Source: J. P. Campbell. (1990). Modeling the perfor-

mance prediction problem in industrial and or-

ganizational psychology. In M. D. Dunnette and L. M.

Hough (Eds.), Handbook of industrial and organizational

psychology (2nd ed., Vol. 1, pp. 687–732). Palo Alto,

CA: Consulting Psychologists Press; and S. A. Wheelan.

(1999). Creating effective teams: A guide for members and

leaders. Thousand Oaks, CA: Sage.

Job Performance 101

model. Not all interpersonal transactions inthe workplace are task related. For example,employees may start off Monday morningswith small talk about what they did overthe weekend. This dimension therefore alsorepresents the extent to which employeesgenerally maintain positive interpersonalrelations with coworkers. This aspect of jobbehavior is not explicitly part of Campbell’smodel, although it is clearly an importantaspect of performance (see Comment 4.2).The distinction between task-oriented perfor-mance and performance that occurs in thecontext of the job has also been highlightedby Van Scotter and Motowidlo (1996).

The third dimension, down-time behaviors,represents behaviors that may lead the jobincumbent to be absent from the worksite.These include counterproductive behaviors,such as drug and alcohol abuse, and otherviolations of the law. They are consideredaspects of performance because anemployee with a substance abuse problem,for example, may be frequently absent fromwork and is therefore not performing well.A closely related set of behaviors is includedin the fourth category, destructive/hazardousbehaviors. These would include such thingsas safety violations, accidents, and sabotage.The down-time behaviors and destructive/

COMMENT 4.2

MAINTAINING POSITIVE INTERPERSONAL RELATIONS AT WORK

MAINTAINING POSITIVE INTERPERSONAL relation-

ships with others is a performance dimension

that is rarely noticed unless someone is unable

to do it. Research over the years has shown,

relatively consistently, that interpersonal con-

flict is perceived negatively by employees and

leads to a number of negative outcomes (e.g.,

Spector & Jex, 1998). Specifically, when thereare frequent interpersonal conflicts in the

work environment, employees tend to dislike

their jobs and feel anxious and tense about

coming to work.

Another aspect of interpersonal relations

that has been explored less frequently, but

may be just as important, is the impact of

interpersonal relations on promotions in orga-nizations. Having worked in different organi-

zations and taught many courses over the

years, a frequent theme we have heard is that

relatively few individuals fail to get promoted

due to lack of technical skills. More often than

not, a lack of mobility in organizations is due

to an inability to get along with others. In fact,

many organizations invest considerable

amounts of money in individual coaching

programs that are often aimed at individuals

who have a great deal of technical prowess but

are lacking in interpersonal skills. Why is it so

important to get along with others in organi-

zations? The likely reason is that much of whatgets done in any organization gets done

through people. If someone has a hard time

getting along with others, it is quite possible

that he or she will have a hard time gaining

others’ cooperation and assistance—factors

that are often necessary to get things done in

organizations.

Source: P. E. Spector and S. M. Jex. (1998). Development

of four self-report measures of job stressors and strain:

Interpersonal Conflict at Work Scale, Organizational Con-

straints Scale, Quantitative Workload Inventory, and

Physical Symptoms Inventory. Journal of Occupational

Health Psychology, 3, 356–367.


hazardous behaviors dimensions are mostclosely related to the dimension of main-taining personal discipline in Campbell’s(1990, 1994) model. In some cases, though,destructive/hazardous behaviors may resultfrom a lack of effort (e.g., not taking thetime to put on safety equipment), so thisdimension may overlap with the demon-strating effort dimension in Campbell’smodel.

Compared to Campbell’s (1990, 1994)eight-dimension model, Murphy’s (1994)four-dimension model is somewhat lessuseful, for two reasons. First, this modelwas developed to explain job performance—specifically, among U.S. Navy personnel.Campbell’s objective was to describe perfor-mance in a broader spectrum of jobs,although his model could certainly be usedto describe job performance among militarypersonnel. Second, the performance dimen-sions described by Murphy are considerablybroader than those described by Camp-bell. Because they are so broad, it is moredifficult to determine the factors that led todifferences among employees on these per-formance dimensions. Despite these disad-vantages, this model again provides us with aset of dimensions for comparing performanceacross jobs. The importance of looking forcross-job dimensions of performance has alsobeen emphasized by Viswesvaran (2002),who has argued that most models of jobperformance include a more general measureof performance followed by more specificmeasures depending on the theorist.Although not all researchers agree on theexact subdimensions of job performance,there is a general recognition that job perfor-mance is more than just the technical aspectsof performing narrowly defined tasks, andthat how employees interact with othersand in other ways contribute to the organi-zation need to be considered.

MEASUREMENT OF JOBPERFORMANCE

The prior section dealt with models organi-zational psychologists have developed tocapture the important dimensions on whichjob performance should be assessed. Thenext critical step in understanding whatcauses employees to perform well or poorlyis developing reliable and valid measures ofjob performance. Having good measures ofjob performance allows us to better under-stand those variables related to performance.In the present section we examine how jobperformance has been measured, as well aschallenges researchers have faced in devel-oping reliable and valid measures of jobperformance. Specifically, three areas arediscussed: (1) measures of job performance,(2) restriction in the variability of job perfor-mance, and (3) instability in job perfor-mance over time.

Measures of Job Performance

By definition, job performance is behavior,so job performance is rarely measureddirectly. More typically, what is measuredis some external assessment of job perfor-mance. According to Murphy (1989a), per-formance can be assessed in eight differentways: (1) paper/pencil tests, (2) job skillstests, (3) on-site hands-on testing, (4) off-sitehands-ontesting,(5)high-fidelitysimulations,(6) symbolic simulations, (7) task ratings,and (8) global ratings. By far, the two mostcommon methods of performance assess-ment in organizations are ratings of employ-ees’ performance on specific tasks andratings of overall performance on the job.

An example of the latter performancerating was used by Van Dyne and LePine(1998). These authors assessed in-roleand extra-role performance by having


employees, their coworkers, and supervisorsrespond to items assessing different aspectsof performance. Sample items from themeasure are presented in Table 4.3. In-roleperformance was assessed by items referringto the extent that employees met their per-formance expectations and performed wellat the tasks that made up the employees’job. Extra-role performance was assessedthrough nontask behaviors that neverthelessare expected of the employee, such as help-ing others in his or her group and attendingscheduled functions.

The literature on performance rating isvast (e.g., Landy & Farr, 1980; Murphy,2004; Murphy & Cleveland, 1990), and willnot be reviewed in detail here. However, twogeneral points can be made. First, there aremany potential sources of error in perfor-mance ratings. For example, a rater maynot have an adequate opportunity to observeperformance, ratings may be biased bythe degree to which the rater likes or dislikesthe ratee, or different raters may employdifferent internal performance standards.These are just three of many potentialsources of error. Rating errors are problem-atic because they ultimately mask meaningfuldifferences in actual job performance, andthus may weaken the relationship betweenjob performance and other variables.

A second point is that steps can be taken toreduce error in performance ratings. Forexample, rater training has been shown toincrease accuracy in performance ratings(Pulakos, 1984). Another way to circumventthe problems with performance ratings is toseek more objective performance measures,such as output produced or sales commis-sions. Unfortunately, these more objectiveperformance measures may have seriousflaws of their own. The most obvious flaw isthat most are really measures of effectivenessor productivity and not actual job perfor-mance (Campbell, 1990). Another disadvant-age is that employees may lack control overobjective performance indicators. For exam-ple, even a very skilled real estate salespersonwould probably not sell many houses if themortgage interest rates rose to 20%.

Some researchers have attempted to dealwith the biases inherent in performance rat-ings by constructing behaviorally anchoredrating scales that clearly reflect those em-ployee behaviors that constitute poor, aver-age, and superior performance (seeCampbell, 1990). For example, consider aparalegal’s ability to prepare a legal brief for acourtroom session. Instead of rating theparalegal on a scale from ‘‘very poor’’ to‘‘very good,’’ a behaviorally anchored ratingscale would include such terms as ‘‘fails to

TABLE 4.3Van Dyne and LePine’s Measure of In-Role and Extra-Role Performance

Sample of In-Role Performance Items

1. ‘‘This particular worker fulfills the responsibilities specified in his/her job description’’

2. ‘‘This particular worker performs the tasks that are expected as part of the job’’

Sample of Extra-Role Performance Items

3. ‘‘This particular worker assists others in this group with their work for the benefit of the group’’

4. ‘‘This particular worker speaks up in the group with ideas for new projects or changes in procedure’’

5. ‘‘This particular worker attends functions that help the work group’’

Source: Van Dyne, L.V., & LePine, J.A. (1998). Helping and voice extra-role behaviors: Evidence of construct and predictive

validity. Academy of Management Journal, 41, 108–119.


incorporate necessary details into the legalbrief’’ and ‘‘produces a report that summa-rizes the major points of the brief withoutadditional unnecessary material.’’ Anchor-ing the scale with specific behaviors ispresumed to take some of the subjectivityout of the performance assessment.

The major point of considering perfor-mance measurement is simply that we mustalways keep in mind that performance is notthe same thing as the measurement of per-formance. Furthermore, because measuringanything will inevitably involve some degreeof error, our understanding of performanceand our ability to predict it will alwaysremain imperfect.

Restriction in the Variability of JobPerformance

Researchers face a number of challengesin adequately assessing job performanceamong a population of employees. We firstconsider the issue of restriction in the varia-bility of performance among employees.For a variety of reasons, the variability inperformance levels within organizations isoften restricted. To better understand restric-tion in performance variability, it is useful todistinguish between artifactual restriction inperformance variability and true restriction.Artifactual restriction in performance varia-bility results from factors such as errorsin performance ratings or the performancemeasurement system. Even though theremay be real differences among employees’levels of job performance, these may bemasked because of an error in the perfor-mance rating process. For example, employ-ees within an organization may really differfrom each other in how well they are per-forming, with some doing very well andsome doing very poorly. However, supervi-

sors may give all employees high ratingsbecause of a concern they will be laid off ifratings are less than stellar. True restrictionin performance variability, on the otherhand, occurs when measures of performanceare relatively accurate but there is a true lackof meaningful variation in actual job perfor-mance. It may simply be the case that allemployees are performing at a high or lowlevel within an organization because of suchfactors as leadership or the inherent diffi-culty of the tasks being performed. In thissection, reasons for true restriction in per-formance variability are discussed.

According to Peters and O’Connor(1988), there are four reasons why variationin individual performance may be restricted.First, organizations simply may have verylow performance standards. If organizationsdo not expect much, this standard will tendto discourage high levels of performance,and employees will gravitate toward mini-mally acceptable levels of performance. Theend result of this process is often a greatreduction in the variability of performance.A good example is the commonly held ster-eotype that performance standards for gov-ernment employees are low. Many readershave probably heard the expression ‘‘Goodenough for government work,’’ whichimplies that work must only be done at aminimally acceptable level.

A second factor, which is related to lowperformance standards, is that organiza-tions vary in the degree to which they valuehigh levels of individual job performance.Organizations either may fail to recognizethe contributions of those who performwell or tolerate individuals who consis-tently perform poorly. Some organizationsmay even inadvertently create situations inwhich low levels of performance areactually rewarded, and high levels of per-formance are punished. For example, in


many organizations, employees who per-form well are often rewarded with greaterresponsibility and heavier workloads, butreceive no additional compensation or pro-motions. The current authors have also seenmanagers rid themselves of incompetentemployees by recommending that they bepromoted to positions in other departments.

A third factor restricting performancevariability is the degree to which organiza-tions excuse employees for low levels ofperformance. This factor is related tolow performance standards but operatessomewhat differently. According to Petersand O’Connor (1988), organizations maydevelop what they describe as a ‘‘culture ofjustification’’ (p. 117); that is, employees areroutinely allowed to ‘‘explain away’’ instan-ces of poor performance. A somewhat moreirreverent way of describing this is the famil-iar acronym CYA, otherwise known as ‘‘coveryour a��!’’Such a culture takes away theincentive to perform well and ultimatelyrestricts performance to mediocre levels.

A final cause of restriction in perfor-mance variability is variation in organiza-tional resources. Having limited resourcesoften leads to situational constraints thatultimately reduces the variability in per-formance (Peters & O’Connor, 1980). Forexample, it is difficult for an auto mechanicto perform well if he or she has no tools. Onthe other hand, if organizational resourcesare extremely plentiful, this may also reducethe variability in performance. In this case,everyone in an organization may perform upto his or her full potential and, as a result, thevariability in performance will be restricted.

A somewhat different explanation as towhy the variation in actual performance lev-els may be restricted is that selection andretention in organizations are not randomprocesses. According to Johns (1991), mostorganizations require that employees pass

through relatively rigorous screening proc-esses before they are hired. For example,those who wish to become police officerstypically must pass through a series of testsbefore even being selected for academy train-ing. In many other occupations, such as law,medicine, and engineering, much of thisscreening is done by universities during pro-fessional training. As a result of these screen-ing processes, the variation in skill andability levels among employees may be quiterestricted, which may ultimately restrict thevariability in job performance. Employeeswho perform poorly or who simply do notfit well with an organization’s culture oftenselect themselves out and leave voluntarily(Ployhart, Weekly, & Baughman, 2006;Schneider, 1987). Like formal socializationprocesses, this again tends to create uniform-ity in job performance.

Despite all of the factors that may restrictperformance variability, empirical evidencesuggests that performance variability in orga-nizations is still meaningful. For example,Schmidt and Hunter (1998) point out thateven though performance variability in orga-nizations is somewhat restricted, a substan-tial portion still remains. If this were not thecase, it is unlikely that selection tools such ascognitive ability tests, personality measures,and biodata instruments would be related toperformance.

Instability in Job Performanceover Time

A second challenge in assessing the job per-formance of employees includes the extent towhich performance is stable versus fluctuat-ing. There has been considerable debate,over the years, concerning the relative stabil-ity of performance criterion measures (e.g.,Ackerman, 1989; Austin, Humphreys, &


Hulin, 1989; Barrett, Caldwell, & Alexander,1985; Henry & Hulin, 1987, 1989). Somecontributors have claimed that performanceis relatively stable over time; many othershave argued (quite forcefully at times) thatperformance is more dynamic. The weight ofthe evidence seems to support the positionthat performance criteria are dynamic. Forexample, Deadrick and Madigan (1990)examined the stability in performance of sew-ing machine operators over time and foundthat the correlations between performancelevels were quite strong when the time inter-val was very short. However, the correlationbetween performance at one point in timeand 23 weeks later was considerably weaker.Thus, because of a variety of factors, em-ployee performance tends to fluctuate overtime. In fact, this inconsistency may explainwhy people are so impressed when a high

level of consistency is displayed. In sports,for example, great honors are bestowedon athletes for breaking records that indi-cate consistency and longevity (see Com-ment 4.3).

Ployhart and Hakel (1998) pointed outthat although evidence supports thedynamic nature of performance, correlationsbetween levels of performance at differentpoints in time provide little insight intohow the performance of individuals changesover time. Furthermore, we know very littleabout variables that predict distinct patternsof change in performance over time. Toaddress this issue, these researchers exam-ined 8 years’ worth of performance criteriondata from a sample of 303 securities analysts.

Using a statistical procedure known aslatent growth curve modeling, which allows themodeling of patterns of change over time,

COMMENT 4.3

THE INTERACTION BETWEEN DECLARATIVE AND PROCEDURAL

KNOWLEDGE

IS DECLARATIVE KNOWLEDGE a necessary precon-

dition to obtaining procedural knowledge?

That is, do you have to know about something

in order to know how to do something? For

some tasks, it is fairly obvious that declarative

knowledge is a precursor to procedural

knowledge. For example, it would be verydifficult to fly a jet airplane if one had abso-

lutely no knowledge of jet propulsion.

For some types of human performance,

however, it is unclear whether declarative

knowledge must precede procedural knowl-

edge. For example, it is not unusual for ath-

letes to understand how to do things but not

necessarily know the principles behind whatthey are doing (perhaps that’s where Nike

came up with the slogan ‘‘Just Do It’’). There

are also instances of great musicians who are

unable to read music but are able to play

musical compositions based on their auditory

memory.

Perhaps those instances when one can

achieve procedural knowledge without firstobtaining declarative knowledge are relatively

rare. However, it would be useful to develop a

greater understanding of the interaction

between these two forms of knowledge. Many

training and educational programs are based

on the premise that declarative knowledge

must come first, so a greater understanding of

this interaction may pave the way for interest-ing new training and educational methods.


they found that, on average, performanceamong these securities analysts approxi-mated a basic learning curve. Initially, per-formance rose steadily; eventually, it reacheda leveling-off point. They also found that,within the sample, not all curves were thesame. For example, there were differences inhow quickly performance initially acceler-ated. There were also differences in howquickly performance reached a leveling-offpoint. Most importantly, they found thatpatterns of change in performance over timewere predictable; for example, those whodescribed themselves as persuasive andempathetic exhibited the quickest initial rateof acceleration in sales. They also found thatthese two variables predicted whether therewould be a drop in performance. Those whodescribed themselves as persuasive weremore likely to exhibit a drop in performanceearly in the second year of employment, andthose describing themselves as empatheticwere less likely to exhibit this drop. At apractical level, this finding suggests thatexhibiting empathy toward clients may be amore effective sales technique than trying topersuade them.

Ployhart and Hakel’s (1998) study pro-vides important insight into the issue of per-formance stability. At least for the sampleemployed, it suggests that although perfor-mance is not stable over time, it does notfluctuate randomly. More importantly, thisstudy suggests that it is possible to identifyand statistically model patterns of change inperformance over time. It also suggests thatthere may be individual differences that pre-dict patterns of performance variability overtime. An important practical implication ofthis possibility is that an organization may beable to identify a desired temporal pattern ofperformance and select individuals who arelikely to exhibit that pattern. For example, it

may be possible to screen out individualswhose performance peaks very quickly andthen declines.

Job performance variability over time canalso be explained by characteristics of the jobitself. Murphy (1989b) proposed that jobsare characterized by what he termed main-tenance stages and transition stages. Duringmaintenance stages, the tasks comprising thejob become somewhat routine and automaticfor the job incumbent. For example, once aperson learns to drive an automobile, thesteps necessary to perform this task becomeso routine that little conscious thought isrequired. When this level of proficiency isachieved, it is as if people are on automaticpilot when they are performing the task. Thismay explain why, during morning com-mutes over the years, the author has wit-nessed drivers applying makeup, eatingbreakfast, or reading newspapers!

When a job is in the transition stage, thetasks comprising the job become novel andthe incumbent cannot rely on automatic rou-tines while performing them. Transitionperiods in jobs may occur during the intro-duction of new technology or perhaps whena major change in laws impacts the jobbeing performed. For example, due to newmanufacturing technology, the jobs of manyproduction employees have changed dra-matically in the past 10 years (Parker & Wall,1998). Also, many employees in nursinghomesandotherlong-termhealthcarefacilitieshave recently experienced profound changesin their jobs because of changes in Medicarebilling procedures (Campbell, 1999).

Murphy (1989b) notes that, because tran-sition periods require adjustments on the partof the employee, they lead to some level ofdisruption and instability in performance.Another consequence of transition pointsis that general cognitive ability is a more


important determinant of performance dur-ing these periods (compared to performanceduring the maintenance period). This makessense, given the well-established finding thatgeneral cognitive ability is a stronger predictorof performance in complex jobs. If this is true,it follows that general cognitive ability shouldbe more strongly related to performance dur-ing these periods. Unfortunately, this propo-sition has not yet received empirical scrutiny.

More recent research by Sturman, Cher-amie, and Cashen (2005) has also empha-sized the importance of job characteristics inexamining the stability of performance overtime. These authors also examined whetherjob performance was assessed with subjec-tive ratings or more objective indicators. Theauthors found that test-retest correlations forjob performance over the course of a yearwere highest (r ¼ .83) for jobs that were lowin complexity and assessed through subjec-tive ratings, and were lowest (r ¼ .50) forjobs that were high in complexity andassessed through objective indicators. Theauthors pointed out that even in the lattercase the correlation between job perfor-mance assessed at different time periodswas relatively high.

DETERMINANTS OF JOBPERFORMANCE

Having discussed how organizational psy-chologists define and measure job perfor-mance, we now consider those factorspredictive of whether employees performwell or poorly. In trying to explain behaviorsuch as job performance, organizational psy-chologists have at times engaged in heateddebates over the relative impact of the personversus the environment (e.g., nature versusnurture). In such cases, these debates areresolved by the rather commonsense notion

that most behaviors are the result of a com-plex interaction between characteristics ofpeople and characteristics of the environ-ment.

Generally speaking, differences in jobperformance are caused by the interactionamong ability, motivation, and situationalfactors that may facilitate or inhibit perfor-mance. Thus, for an employee to performwell, he or she must possess job-relevantabilities. Ability alone will not lead to highlevels of performance, though, unless theemployee is motivated to perform well anddoes not experience severe situational con-straints. Of course, in some cases, a highlevel of one of these three factors will com-pensate for low levels of the others (e.g., ahighly motivated employee will overcomesituational constraints), but usually all threeconditions are necessary.

This section begins with an examina-tion of a well-known theoretical model ofthe determinants of job performance, fol-lowed by an exploration of empirical evi-dence on determinants of job performance.Given the vast number of factors that influ-ence job performance, the exploration ofthe empirical literature will currently belimited to individual differences or charac-teristics of persons that explain perfor-mance differences. Environmental factorsthat influence job performance (e.g., lead-ership, motivation, situational constraints)will be covered in more detail in subse-quent chapters.

Campbell’s Model of JobPerformance

Campbell (1990, 1994) proposed that jobperformance is determined by the interac-tion among declarative knowledge, proceduralknowledge/skill, and motivation (see Fig. 4.1).


Declarative knowledge is simply knowledgeabout facts and things. An employee with ahigh level of declarative knowledge has agood understanding of the tasks that arerequired by his or her job. As an example,a medical technician with a high level ofdeclarative knowledge knows the steps nec-essary to draw blood from a patient. Accord-ing to Campbell, differences in declarativeknowledge may be due to a number of fac-tors, such as ability, personality, interests,education, training, experience, and theinteraction between employee aptitudesand training. Many forms of professionaland academic training, at least in the earlystages, stress the acquisition of declarativeknowledge. The first year of medical school,for example, requires considerable memori-zation of information about human anatomyand physiology.

Once an employee has achieved a highdegree of declarative knowledge, he or sheis in a position to acquire a high level ofprocedural knowledge/skill. When this isachieved, the employee understands notonly what needs to be done but also howto do it, and is able to carry out these behav-iors. A medical technician who has achieveda high level of procedural skill or knowledgenot only knows the steps involved in draw-

ing blood, but is also able to perform thistask. According to Campbell, differences inthe acquisition of procedural knowledge/skill are determined by the same factors thatlead to differences in declarative knowledge.In academic and professional training, theacquisition of procedural knowledge/skilltends to be emphasized at later stages or,typically, after a sufficient degree of declara-tive knowledge has been acquired. Medicaltraining, for example, becomes more handson during the third and fourth years.

When an employee has reached a highlevel of procedural knowledge/skill, he orshe is capable of high levels of job perfor-mance. Stated differently, the employee hasperformance potential. Whether this po-tential actually leads to high levels of jobperformance depends on motivation. Ac-cording to Campbell (1990, 1994), motiva-tion reflects an employee’s choices regarding(1) whether to expend effort directed atjob performance, (2) the level of effort toexpend, and (3) whether to persist with thelevel of effort that is chosen. Thus, even if anemployee has achieved a very high level ofprocedural knowledge/skill, low motivationmay prevent the skill from being translatedinto a high level of performance. For exam-ple, a highly capable employee may simplydecide not to put forth any effort, may notput in enough effort, or may put forth theeffort but lack the willingness to sustain itover time.

The primary value of Campbell’s (1990,1994) model is that it states, in precise terms,the factors within the person that determineperformance, and the interplay among thosefactors. Furthermore, it has received empiri-cal support (e.g., McCloy, Campbell, &Cudeck, 1994). The model also reminds usthat the interaction among the factors thatdetermine performance is complex. Forexample, a high level of motivation may

FIGURE 4.1Campbell’s (1990, 1994) Model of theDeterminants of Job Performance

ProceduralKnowledge/Skill

DeclarativeKnowledge

MotivationX X

Adapted from J. P. Campbell. (1990). Modeling the perfor-

mance prediction problem in industrial and organizational

psychology. In M. D. Dunnette and L. M. Hough (Eds.), Hand-

book of industrial and organizational psychology (2nd ed., Vol.

1, pp. 687–732). Palo Alto, CA: Consulting Psychologists Press.

Modified and reproduced by permission of the publisher.


compensate for a moderate level of proce-dural knowledge/skill. On the other hand, alow level of motivation may negate thepotential benefits of a high level of proce-dural knowledge/skill. This model can alsobe used to generate ideas and hypothesesabout performance and its determinants(see Comment 4.4).

Given all the factors that have been pro-posed to explain differences in job perfor-mance, a logical question may be: What isthe relative contribution of all of these factorsto performance? Indeed, so much researchhas examined this question over the yearsthat a comprehensive review of this literatureis clearly beyond the scope of the chapter. It

is possible, however, to draw some conclu-sions, at least with respect to individual dif-ference predictors of performance. As statedearlier, situational factors that impact perfor-mance will be covered in other chapters.

General Mental Ability as aPredictor of Job Performance

By far the one individual difference variablethat has received the most attention as adeterminant of job performance is generalmental ability. Numerous definitions havebeen offered, but the common element inmost definitions of general mental ability isthat it reflects an individual’s capacity to

COMMENT 4.4

WHAT IS JOB EXPERIENCE?

JOB EXPERIENCE IS a variable that is used sofrequently in organizational psychology that

it is easy to take its importance for granted.

Typically, most researchers don’t pay too

much attention to job experience because they

are measuring it either for descriptive pur-

poses, or to use as a control variable in statis-

tical analyses. In the vast majority of studies,

experience is measured simply as the numberof months or years that a person has been

employed in a particular job or in a particular

organization.

Tesluk and Jacobs (1998) pointed out that

organizational or job tenure is not likely to

capture the complexity of job experience.

They point out, for example, that the same

length of tenure may be very different in termsof both the density and the timing of job-

related experiences. A good example of the

density dimension would be a surgeon per-

forming in a war zone. This individual would

typically be doing surgeries around the clock,

and would thus acquire more surgical experi-

ence in three months than a surgeon at aregular civilian hospital would acquire in

twice the time. A good example of timing

would be a manager’s having to take over a

poorly performing department immediately

after completing his or her training. Such an

experience would undoubtedly have a greater

impact on this individual now than it would

later in his or her career.Many organizations recognize complexity

of experience and attempt to structure the

assignments of high-potential managers in a

way that maximizes their developmental

value. For the most part, however, researchers

have treated experience in a very simplistic

fashion. In the future, this is likely to be a very

fruitful area of research in organizational psy-chology.

Source: P. E. Tesluk and R. R. Jacobs. (1998). Toward an

integrated model of work experience. Personnel Psychol-

ogy, 51, 321–355.


process and comprehend information (Mur-phy, 1989b; Waldman & Spangler, 1989).Research has consistently shown, over manyyears, that general mental ability predictsperformance over a wide range of jobs andoccupations. The most comprehensive dem-onstration of this was a meta-analysis con-ducted by Schmidt and Hunter (1998), inwhich nearly 85 years of research findings onvarious predictors of job performance weresummarized. Their analysis indicated thatthe corrected correlation between generalmental ability and performance across jobswas .51—that is, over 25% of the variance inperformance across jobs is due to differencesin general mental ability. A recent meta-analysis of 283 independent samples con-ducted in the United Kingdom also revealedvalidity coefficients between general mentalability and performance in the .5 to .6 range(Bertua, Anderson, & Salgado, 2005). Inaddition, recent research suggests that meta-analyses of the relationship between generalmental ability and job performance have notsufficiently adjusted for the problem of rangerestriction in mental ability (i.e., individualsvery low in mental ability are likely not tobe hired), and may actually underestimatethe correlation by 25% (Hunter, Schmidt,& Huy, 2006). When one considers thenumerous other factors that can influencejob performance (e.g., motivation, leader-ship, situational constraints), the fact thatmental ability is such a strong predictor istruly impressive.

Why is general mental ability such a keyto explaining differences in job performance?According to Schmidt, Hunter, and Outer-bridge (1986), the intermediate link betweengeneral mental ability and job performance isjob knowledge; that is, employees who possesshigher levels of general mental ability tend todevelop a greater understanding of their jobduties than individuals with lower levels. For

example, a very intelligent airplane pilotwould likely possess greater knowledge ofall that goes into flying a plane than a pilotwho was less intelligent. In essence, thosewith high levels of mental ability are able toextract more relevant information from thejob environment than those with lower levelsof general mental ability.

Another consistent finding in this litera-ture is that general mental ability is a betterpredictor of performance in jobs that have ahigh level of complexity compared to jobslower in complexity (Bertua et al., 2005;Hunter, Schmidt, & Judiesch, 1990).Although there is no standard definition,most researchers agree that job complexityis strongly influenced by the mental de-mands and information-processing require-ments placed on job incumbents (Wood,1986). For example, the job of a corporateexecutive requires the use of higher-ordercognitive skills such as planning and synthe-sizing large amounts of information. On theother hand, the job of a convenience storeclerk typically requires what might be con-sidered lower-level cognitive skills such asfollowing established guidelines and pro-cedures. General mental ability predictsgood performance in complex jobs, pri-marily because such jobs place higher-levelinformation-processing demands on in-cumbents. Thus, compared to those withlower levels, incumbents who possess highlevels of general mental ability are betterable to meet those demands.

Although mental ability is a strongpredictor of job performance, researchershave also found that there are large ethnicgroup differences on such tests, which maycontribute to biased selection and sub-sequent litigation (Van Rooy, Dilchert, &Viswesvaran, 2006). Therefore, researchershave become interested in assessing otherforms of intelligence, such as emotional


intelligence. Baron, Handley, and Fund(2006) describe emotional intelligence asbeing able to understand and express your-self; understand and relate with others; man-age and control emotions; change, adapt,and solve problems of a personal and inter-personal nature; and finally the ability togenerate positive mood and to be self-motivated. Although emotional intelligenceshows fewer ethnic group differences, meta-analyses reveal it only accounts for 2% of thevariance in performance after controlling forgeneral intelligence (Van Rooy & Viswesvaran,2004).

Job Experience as a Predictor of JobPerformance

Job experience is an individual difference var-iable that has been examined frequently as ageneral predictor of job performance. Itwould seem logical that a person with ahigher level of relevant job experience wouldperform better than others who possess littleor no job experience. Empirical evidencehas, in fact, shown that experience, like gen-eral mental ability, is positively related to jobperformance over a wide range of job types(McDaniel, Schmidt, & Hunter, 1988;Schmidt & Hunter, 1998). Like generalmental ability, the relation between experi-ence and job performance appears to bemediated by job knowledge (Schmidt et al.,1986). Researchers have also found that therelationship between experience and jobperformance depends on job complexity.For example, McDaniel et al. (1988) foundthat experience was a better predictor of per-formance in low- rather than high-complexityjobs. They attributed this difference tothe fact that experience is really the onlypreparation for low-complexity jobs. Forexample, there is no way to learn how toperform the job of convenience store clerk

other than by actually working at it. Withhigh-complexity jobs, however, educationmay compensate for a lack of experience.Note that the form of this interaction effectis exactly the opposite of that found for gen-eral mental ability.

There is also evidence that the impor-tance of job experience in explaining perfor-mance differences tends to diminish overtime. For example, McDaniel et al. (1988)found that the correlation between experi-ence and performance was strongest in sam-ples in which the average level of jobexperience was less than 3 years, but thecorrelation was considerably less for samplesin which the average level of experience washigher. This suggests that there is a law ofdiminishing returns with respect to the influ-ence of job experience on job performance.

Research on the influence of job experi-ence on job performance should be viewedcautiously, however, because most studieshave measured job experience as the numberof years in an organization or job. QuinonesFord, and Teachout (1995) pointed out thatjob experience can be viewed not only interms of quantity but also quality. Years ofexperience is a quantitative measure of expe-rience. If job experience is viewed qualita-tively, this has to do with the job tasksperformed and the relevance of situationsone has been exposed to on the job. Forexample, if an individual has several yearsof experience as an accountant, but has con-ducted few field audits, that person will notnecessarily perform better in an auditingposition than an individual who has lessgeneral accounting experience.

Building on the work of Quinones et al.(1995), Tesluk and Jacobs (1998) proposedthat job experience can also be viewed interms of both the density and timing ofjob-related experiences. When experiencehas high density, the employee is exposed


to many developmental experiences in a rela-tive short period of time. These may includeincreased responsibilities, and perhaps evenbeing required to perform under very diffi-cult conditions. The timing dimension has todo with the fact that certain experiencesmight have more, or less, developmentalvalue, depending on whether they occur atthe beginning, middle, or latter stage of one’scareer. For most employees, mistakes have agreater developmental impact when theyoccur at the early (as opposed to later) stagesof one’s career. The more important pointfrom the work of Quinones et al. (1995) andTesluk and Jacobs (1998) is that job experi-ence is a complex variable, and much theo-retical and empirical work needs to be donebefore we fully understand and appreciate it(see Comment 4.5).

Personality as a Predictor of JobPerformance

Along with general mental ability and jobexperience, there has been a recent wave ofresearch on personality as a predictor of jobperformance. The personality trait consis-tently found to predict job performance overa wide range of jobs is conscientiousness (Bar-rick & Mount, 1991, 2005; Dudley, Orvis,Lebiecki, & Cortina, 2006; Ones, Viswes-varan, & Schmidt, 1993; Thoresen, Bradley,Bliese, & Thoresen, 2004). A person who isconscientious can be described as depend-able, goal oriented, planful, and achieve-ment oriented. Barrick and Mount (1991)found that the corrected correlation betweenconscientiousness and performance, acrossa wide variety of jobs, was .22. Ones et al.

COMMENT 4.5

CONSISTENCY OF PERFORMANCE IN BASEBALL

SOME OF THE most highly regarded records in

the world of sports reflect consistency of per-

formance. In baseball, for example, a record

that has stood for over 50 years is New York

Yankee Joe DiMaggio’s 56-game hitting streak.More recently, Cal Ripken, Jr., of the Baltimore

Orioles, made history by playing in 2,632

consecutive games. Why are these two records

so highly regarded? DiMaggio’s record is

remarkable when one considers all of the fac-

tors that work against obtaining a base hit in

that number of consecutive games. One would

think that the skill of pitchers at the majorleague level, minor injuries, and general fatigue

would make such a streak highly unlikely.

Thus, this record is a reflection of DiMaggio’s

skill as a hitter, and his determination.

One reason Ripken’s streak is so unusual is

simply that few players last that long at the

major league level. It is also unusual for players

to avoid serious injuries for that period of time.

Furthermore, because of the number of games

played in a major league season (162), and

minor injuries, most players want an occasionalday off. Thus, Ripken’s streak is a reflection of a

number of factors, including consistency in

performance, rigorous off-season conditioning,

and a high level of motivation.

What do these baseball records tell us

about stability and consistency of perfor-

mance? If anything, they highlight the fact that

stability and consistency, over time, are morethe exception than the rule. Because of exter-

nal constraints, fluctuations in motivation,

and just plain good/bad luck, performance

in most domains is often quite variable. How-

ever, when it does remain consistent for a long

period of time, it is often highly rewarded.


(1993) found that the mean correctedcorrelation between integrity tests (whichmany presume are measures of conscien-tiousness) and job performance, across jobs,was .34.

There are three explanations for whyconscientiousness is a robust predictor ofperformance. According to Schmidt andHunter (1998), the variable that links con-scientiousness and job performance is jobknowledge. Recall that this was the samevariable proposed to mediate the relationbetween both general mental ability andexperience and performance. In this case,however, the process has to do primarilywith motivation rather than with ability.Individuals who are highly conscientiouspresumably put time and effort into acquir-ing high levels of job knowledge, and hencewill perform better than those who are lessconscientious.

Another explanation for the relationbetween conscientiousness and performanceis goal setting. In a study of sales personnel,Barrick, Mount, and Strauss (1993) foundthat goal setting mediated the relationbetween conscientiousness and job perfor-mance. Specifically, those who were highlyconscientious exhibited a greater tendencyfor setting performance-related goals thanthose who were less conscientious. This pro-clivity for setting goals facilitated, in turn,higher levels of job performance. This addsto the findings of Schmidt and Hunter(1998) regarding why highly conscientiouspeople tend to perform well, regardless ofthe job.

A final explanation for the relationshipbetween conscientiousness and performanceis motivational. Barrick, Stewart, and Pio-trowski (2002) found that conscientiousnesswas related to performance ratings ofemployees through the employee striving toaccomplish more at work and reach a higher

status. After these motivational variableswere controlled, conscientiousness no lon-ger predicted job performance. This researchsupports a model in which conscientious-ness gives rise to increased motivation,which leads employees to perform better.

It is also worth noting that, in addition toconscientiousness, other personality traitsare more likely to predict performance inparticular types of jobs. Barrick and Mount(2005) point out that the personality traitsof extraversion and agreeableness (the ten-dency to avoid conflict and be easy to getalong with) are especially predictive ofjob performance for jobs requiring employ-ees to interact with other people on a fre-quent basis. Openness to experience, thetendency to be open to new ideas and expe-riences, is an important predictor for jobsthat require employees to continuouslyadapt to change.

A more recent approach to personalityand job performance involves combiningtraits into a larger factor that predicts jobperformance. Erez and Judge (2001) arguedthat self-esteem, locus of control, generalizedself-efficacy, and neuroticism all tap a person’score self-evaluation, which they define as‘‘basic conclusions or bottom-line evalua-tions that represent one’s appraisal of people,events, or things in relation to oneself’’(p. 1,270). Self-esteem is a person’s overallattitude toward himself or herself. Locus ofcontrol refers to whether individuals believethe causes of their behavior are either due totheir own actions (internal locus of control)or the environment (external locus of con-trol). Generalized self-efficacy refers towhether people think they can generallyaccomplish the tasks they face. Finally, neu-roticism refers to a lack of emotional stabilityand the tendency to experience negativeaffective states. Neuroticism contributes neg-atively to an individual’s core self-evaluation.


Erez and Judge (2001) found that thefour separate personality traits all contrib-uted to the one larger trait of core self-evaluation. More importantly, the authorsfound that this larger trait was predictive ofperformance in both student samples(r ¼ .35) and insurance salesmen (r ¼ .34for sales volume and r ¼ .44 for rated per-formance). The correlations between coreself-evaluations and performance weregreater than any of the correlations betweenthe four traits that made up the larger trait.Finally, core self-evaluations are related toperformance even when controlling for con-scientiousness. In a recent review, Judge,Van Vianen, and De Pater (2004) providedsupport for the core self-evaluation person-ality trait, and also provided a shorter scalethat can assess the variable.

Summarizing the state of knowledgeregarding the predictors of job performance,the most important individual differencevariables influencing job performance aregeneral mental ability, job experience, andconscientiousness. Furthermore, the pri-mary mechanisms linking these variables tojob performance are job knowledge and, to alesser extent, goal setting and motivation.Finally, many of these relations appear tobe influenced by job complexity. Figure 4.2summarizes these propositions.

Readers will undoubtedly note thatFigure 4.2 does not contain a number ofsituational factors such as motivation, lead-ership, and organizational climate. This wasdone largely for pedagogical reasons, be-cause the link between these situational fac-tors and performance will be covered in laterchapters. It is important to note, however,that although few studies have examined thejoint effect of individual differences and sit-uational factors, it has been demonstratedempirically that both do contribute to jobperformance (see Barrick & Mount, 2005;

Colarelli, Dean, & Konstans, 1987; Day &Bedeian, 1991). Thus, organizations must domore than simply hire smart, experienced,conscientious people in order to facilitatehigh levels of employee performance.

ORGANIZATIONALCITIZENSHIP BEHAVIOR

The second form of productive behavior to bediscussed in this chapter is organizationalcitizenship behavior (OCB) (Organ, 1977,1994). Generally speaking, OCB refers tobehaviors that are not part of employees’formal job descriptions (e.g., helping a co-worker who has been absent; being courteousto others), or behaviors for which employeesare not formally rewarded. Even though suchbehaviors are not formally mandated by orga-nizations, in the aggregate they are believed toenhance the effectiveness of groups and orga-nizations (George & Bettenhausen, 1990;Katz & Kahn, 1978; Podsak-off, Ahearne, &

FIGURE 4.2Summary of the Most Important IndividualDifference Predicators of Job Performance

JobKnowledge

GoalSetting

JobComplexity

JobPerformance

GeneralCognitive

Ability

JobExperience

Conscientiousness


MacKenzie, 1997). Recall from the previousmodels of job performance (Campbell, 1990,1994; Murphy, 1994) the distinctionbetween in-role and extra-role performance,and how extra-role performance representedthose aspects of job performance not tied tospecific tasks relevant to one’s primary areaof expertise (e.g., teamwork, dedication,communication ability). The distinctionbetween extra-role performance and OCB israther blurred. Technically, the key distinc-tion is that OCBs are not evaluated as part ofthe formal appraisal system used to assessemployees. In addition, the antecedents ofOCB are different from those of in-role andextra-role performance.

One way of classifying OCB has beenadopted by Organ (1977, 1994), whereOCB in organizations can be categorized intofive different types:

1. Altruism represents what we typicallythink of as ‘‘helping behaviors’’ in theworkplace. This form of OCB is some-times referred to as prosocial behavior.An example of altruism would be anemployee’s voluntarily assisting a co-worker who is having difficulty operat-ing his or her computer.

2. Courtesy represents behaviors that reflectbasic consideration for others. An exam-ple of behavior within this categorywould be periodically ‘‘touching base’’with one’s coworkers to find out howthings are going, or letting others knowwhere one can be reached.

3. Sportsmanship is different from otherforms of OCB because it is typicallyexhibited by not engaging in certainforms of behaviors, such as complainingabout problems or minor inconveniences.

4. Conscientiousness involves being a goodcitizen in the workplace and doing thingssuch as arriving on time for meetings.

5. Civic virtue is somewhat different fromthe others because the target is theorganization—or, in some cases, thework group—rather than another indi-vidual. An example of this form of OCBwould be attending a charitable func-tion sponsored by the organization.

Although this classification schemerepresents a reasonable way of carving upOCB, other researchers have organizedOCB differently. For example, Organ andKonovsky (1989) distinguished OCB thathelped others at work when they had aproblem (altruism) from following rulesand doing whatever is needed to get thejob done (compliance). Finally, McNeelyand Meglino (1994) distinguished OCBsthat are directed at helping others (OCBI)from those that are directed toward theorganization as a whole (OCBS). These lat-ter types of distinctions are typically guidedby a researcher’s specific interest in com-paring the predictors of different types ofOCB.

Reasons for OCB

Why do employees engage in OCB? Thereare actually three different explanations.According to the first, the primary deter-minant is positive affect, typically in theform of job satisfaction. Theoretically, thisview comes from a fairly long history ofsocial–psychological research showingthat a positive mood increases the fre-quency of helping and of other forms ofspontaneous prosocial behavior (seeGeorge & Brief, 1992). Furthermore, pos-itive mood and helping behavior areactually mutually reinforcing becausehelping others usually makes people feelgood. Bettencourt, Gwinner, and Meuter(2001) found that positive job attitudes


were related to different types of OCB inservice-oriented employees. Researchershave also found that job involvement, acorrelate of job satisfaction, is positivelycorrelated with supervisor ratings of OCB(Diefendorff, Brown, & Kamin, 2002).

A second explanation for OCB has to dowith cognitive evaluations of the fairness ofemployees’ treatment by an organization.This view is theoretically rooted in EquityTheory (Adams, 1965), which states thatemployees evaluate their work situations bycognitively comparing their inputs to theorganization with the outcomes they receivein return. (Equity Theory will be covered inmore detail in Chapter 8.) If employees per-ceive that the organization is treating themfairly or justly, then they are likely to recip-rocate the organization by engaging in OCB.It seems, however, that certain forms of fair-ness or justice predict OCB better thanothers. For example, Moorman (1991) foundthat the best predictor of OCB was interac-tional justice, or the manner in which super-visors treat employees as they carry outorganizational policies and procedures. Incontrast, other studies have found that proce-dural justice is a better predictor of OCB thanis distributive justice (e.g., Konovsky & Pugh,1990). Procedural justice refers to employees’perceptions of the fairness of proceduresused to make decisions such as pay raises;distributive justice refers to perceptions offairness of the outcomes one receives as aresult of those procedures. Recent researchsuggests that perceptions of organizationaljustice are especially important predictors ofOCB for employees who are classified asentitled (Blakely, Andrews, & Moorman,2005). These types of employees like theiroutcomes to be greater than other employeeseven when inputs are comparable. Theseindividuals may be especially likely to basetheir decision to engage in OCB on the extent

to which they feel they are being treated fairlyby the organization.

A third explanation for OCB is that it isdue to dispositions. According to this view-point, certain personality traits predisposeindividuals to engage in OCB. In other words,some people are naturally more helpful thanothers are. Compared to the first two explan-ations of OCB, the dispositional viewpointhas received much less attention in the OCBliterature because proponents of this viewhave been vague as to the specific personalitytraits that should be related to OCB. This hasbeen a criticism of dispositional explanationsof other forms of employee attitudes andbehavior (Davis-Blake & Pfeffer, 1989).

Other than affect, fairness, and disposi-tions, a handful of other factors have beenproposed to explain the performance ofOCB, although none of these has receivedextensive empirical scrutiny. For example,Chattopadhyay (1998) found evidence thatOCB is predicted by the demographic com-position of work groups. It has also beenfound that the performance of OCB may beinfluenced by factors such as job-relatedstressors (Jex, 1998; Jex, Adams, Bachrach,& Rosol, 2001) and employees’ level of orga-nizational commitment (Williams & Ander-son, 1991). Wang, Law, Hackett, Wang, andChen (2005) have recently shown that lead-ership is an important predictor ofOCB. Using employees from organizationsthroughout the People’s Republic of China,these authors found that positive percep-tions of and trust in leaders were related toa greater tendency to perform OCB. Finally,Finkelstein and Penner (2004) found thatmotives surrounding the desire to help co-workers and possessing a citizenship-roleidentity (e.g., ‘‘helping the company is animportant part of who I am’’) were morestrongly related to OCB than motives asso-ciated with impression management.


To evaluate the relative impact of variousantecedents of OCB, Organ and Ryan (1995)conducted a meta-analysis of 55 studies.Their results suggest that job satisfactionand perceived fairness were correlated withOCB at approximately the same magnitude.The results for dispositional predictors ofOCB were rather disappointing, however. Forexample, personality traits such as conscien-tiousness, agreeableness, positive affectivity,and negative affectivity were all unrelated toOCB. As Figure 4.3 summarizes, the mostlogical conclusion to be drawn from Organand Ryan’s meta-analysis is that affective andcognitive influences combine in an additivefashion to determine OCB.

Special Issues in OCB Research

Since Organ (1977) first introduced the con-cept of OCB, there has been considerableresearch on the topic. As with most well-researched topics, many issues have generat-ed controversy and debate among research-ers in this area. In this section, four of theseissues are discussed briefly.

The underlying premise behind OCBresearch is that this form of productivebehavior is necessary in order for organiza-tions to be effective (Katz & Kahn, 1978).

What is typically argued is that if employeesperformed their jobs exactly as written, anddid nothing beyond that, organizationswould not be able to function effectively.Surprisingly, this claim had received virtu-ally no empirical scrutiny until very recently.It has now been shown empirically, at leastfor groups, that OCB is positively related toeffectiveness (Karambayya, 1989; Podsakoffet al., 1997). As would be expected, groupsin which members engage in more OCBstend to be more effective than groups inwhich members engage in fewer of thesebehaviors. For example, researchers havefound that OCB is related to aspects of orga-nizational effectiveness (e.g., profit, cus-tomer satisfaction) among bank branches inTaiwan (Yen & Niehoff, 2004).

What is still not clear from research onOCB and its effectiveness is the direction ofcausality underlying this relationship. Re-searchers have largely operated under theassumption that OCB has a causal impacton group and organizational effectiveness.However, it is also possible that the directionof causality could be reversed. Members ofeffective groups may report high levels ofOCB, regardless of whether they actuallyexist. When a group is successful, groupmembers may perceive high levels of OCBas they bask in the glow of this success. In arelated study, Staw (1975) found that groupmembers’ retrospective reports of groupcohesiveness could be manipulated basedon false feedback about group performance.In this study group, members who were toldthat their group had been successfulreported higher levels of cohesiveness thandid group members who were told that theirgroup had been unsuccessful. Using thesame paradigm as Staw (1975), Bachrach,Bendoly, and Podsakoff (2001) recentlyfound evidence that retrospective percep-tions of OCB may be influenced by group

FIGURE 4.3Summary of the Major Antecedents ofOrganizational Citizenship Behavior (OCB)

PositiveAffect

Fairness/EquityCognitions

OrganizationalCitizenship Behavior


performance. This issue will undoubtedly beaddressed in future OCB research.

A second issue concerns the validity ofthe OCB concept itself. As originally definedby Organ (1977), OCB represents behaviorthat is above employees’ formal job respon-sibilities, and for which there are no formalrewards. With regard to the first issue, it isbecoming increasingly questionable that, inperforming their day-to-day activities,employees make the job descriptive versusnonjob descriptive distinctions upon whichOCB is based. This suggests that manyemployees view activities, such as helpingother employees, being courteous to others,and occasionally attending functions onbehalf of their organization, as part of theirformal role responsibilities. This reasoning issupported by Morrison (1994), who found,in a sample of clerical employees, that manybehaviors that are considered OCB wereclassified by these employees as part of theirnormal job responsibilities. She also foundthat there was very little correlation betweenemployees’ and supervisors’ classifications ofOCBs. Thus, many of the behaviors thatsupervisors consider OCB may simply rep-resent employees’ doing things that they con-sider to be part of their jobs.

Another interesting finding from Morri-son’s (1994) study was that employees weremost likely to classify OCBs as in-role behav-iors when they reported high levels of bothjob satisfaction and affective organizationalcommitment. Building on this finding, Bach-rach and Jex (2000) conducted a laboratorystudy in which they used a mood-inductionprocedure to investigate the impact of moodon the categorization of OCB for a simulatedclerical position. In this study, it was foundthat inducing a positive mood state had noimpact on classification of OCB. Interest-ingly, though, subjects who experienced anegative mood-induction procedure classi-

fied fewer of the OCBs as being part of theirregular roles, compared to those in the pos-itive or neutral mood conditions. These find-ings suggest that negative affect may result ina more narrow definition of one’s role. Takentogether with Morrison’s study, these find-ings call into question the in-role versusextra-role distinction that has been implicitin OCB research.

A third issue in OCB research is whetheremployees really engage in OCB without theexpectation that such behaviors will berewarded. Despite Organ’s (1977) initialclaim, recent evidence suggests that thisassumption may be rather questionable.For example, it has been shown empiricallythat performing OCB positively influencesformal performance appraisals (Eastman,1994), and it is doubtful that employeesare unaware of this. According to Bolino(1999), when OCB is performed with theexpectation of future rewards, it thenbecomes a form of impression management(see Chapter 10) rather than truly altruisticbehavior. Impression management behav-iors are simply tactics people use to influenceothers’ views of them. According to Bolino,OCB is most likely to be used as an impres-sion management tool when it is highly visi-ble to others, particularly those responsiblefor the dispensation of rewards. As an exam-ple, an employee may help other employeesonly when his or her supervisor is around toobserve.

One could certainly argue that as long asOCB is performed, the motivation is irrele-vant. However, the reasons behind suchbehavior are important if organizations wantto influence the performance of OCB. Ifemployees perform OCB primarily becausethey are satisfied with their jobs, or feel thatthey have been treated fairly, organizationscan influence the performance of OCB bytreating employees fairly and taking steps


to enhance satisfaction. On the other hand, ifOCB is performed with the expectation ofrewards, or for impression management pur-poses, organizations should directly or indi-rectly link rewards to the performance ofOCB. In essence, this suggests that OCBshould be explicitly recognized as anotherform of job performance.

A final issue in OCB research is whetherOCB will remain a viable concept in theworkplace of the future. Bridges (1994),among others, has pointed out a clear trendin recent years: Organizations have beenmoving away from formal job descriptions.In fact, Bridges has predicted that the con-cept of a job will eventually cease to exist (see

Comment 4.6). This has not occurred as yet,but it is true that the work of employees inmany organizations has become increasinglyproject driven, and their activities revolvemore and more around project completionrather than fulfilling their job duties. Giventhis trend, one may ask whether the in-role/extra-role distinction upon which OCB restswill be relevant in the workplace of thefuture. Behaviors considered to be OCB willstill be necessary in a dejobbed environment,but employees in the future will tend toconsider them ‘‘part of the job,’’ at least tothe extent that they facilitate project comple-tion. As Morrison’s (1994) study shows, thisis already occurring but will probably

COMMENT 4.6

A WORLD WITHOUT JOBS

WILLIAM BRIDGES, IN his 1994 book JobShift: How

to Prosper in a Workplace without Jobs, argues

that, in the near future, the concept of a ‘‘job’’

will cease to exist. That is, rather than having a

formalized job description that lays out one’s

duties, each person in an organization will be

given project-based objectives and expected to

accomplish them. One of the implications ofhaving no formalized jobs is that organiza-

tions will be able to make much greater use

of temporary and contingent employees; that

is, an organization will be able to bring in

specialists on an ‘‘as needed’’ basis to complete

specific projects. This will give organizations

considerable flexibility and allow them to

operate with much lower labor costs. Anotherimplication of this trend is that more and more

people will become ‘‘independent contrac-

tors’’ rather than permanent employees of a

given organization.

According to Bridges, this trend toward

doing away with jobs has thus far been most

evident in organizations that operate in high-

technology sectors. This is largely due to the

speed at which things are done in these sec-

tors, and the need for constant innovation.

Will other types of organizations eventually

do away with jobs? Although it’s certainly

possible, there are reasons to believe that

many organizations will not do away with

jobs. For example, defending the legal sound-ness of selection and promotion procedures

depends, to a large degree, on the job-related-

ness of those procedures. Thus, an organiza-

tion without job descriptions would be in a

very difficult position if its selection and pro-

motion procedures were challenged. One

would also assume that unions would be very

wary of doing away with job descriptionssince they help in establishing wage rates

and essentially serve as a ‘‘contract’’ as to the

job duties employees are expected to perform.

Source: W. Bridges. (1994). JobShift: How to Prosper in a

workplace without Jobs. Reading, MA: Addison-Wesley.


become a more pronounced trend becausemany employees may not have formal jobdescriptions to guide their behaviors.

INNOVATION INORGANIZATIONS

The third and final form of productivebehavior to be examined in this chapter isinnovation. Like OCB, innovation is really anaspect of job performance, but it is uniqueenough that a distinct literature examiningits antecedents has developed. Although nostandard definition of innovation exists, thisform of productive behavior may be thoughtof as instances in which employees come upwith very novel ideas or concepts that furtherthe goals of the organization. Hellstrom andHellstrom (2002) coined the term organiza-tion ideation to refer to ‘‘the process of creat-ing useful conceptual novelty, and thecirculation and taking on of that novelty inan organization’’ (p. 108). The most visibleforms of these types of employee innovationin organizations are new products and serv-ices, and there are many examples of these.The Dell Computer Company, for example,has been an innovator in the marketing anddistribution of personal computers. Saturnhas been an innovator in both the distribu-tion and service of automobiles. Not all inno-vations, however, take the forms of productsand services. For example, an employee oremployees may come up with a unique orga-nizational structure, a more efficient produc-tion method, or some other cost-savingadministrative procedure.

In the organizational innovation litera-ture, there are four distinct streams ofresearch (Damanpour, 1991). For the firststream, some researchers have examined theprocess by which employees come up withinnovative ideas; others are more interestedin determining the characteristics that dis-

tinguish highly innovative employees fromothers. Note that, in both cases, the focus ison the employee or employees responsiblefor the innovation. This view is also reason-ably congruent with the definition of inno-vation proposed in this chapter. For thesecond stream, innovation is viewed from amore macro perspective; that is, many inno-vation researchers focus on what is describedas the diffusion of innovations throughout anorganization (see Greenhalgh, Robert, &McFarlane, 2004). An example of this mightbe the manner in which computers come tobe utilized companywide. For the thirdstream, innovation researchers tend to focuson what can be described as the adoption ofinnovations (Frambach & Schillewaert,2002). Viewed from this perspective, thefocus is on an organization’s initial decisionon whether to adopt some innovation.Finally, a fourth stream emphasizes theimportance of individuals and organizationsin innovation. Hellstrom and Hellstrom(2002) have recently emphasized how indi-vidual workers and facilitating organiza-tional conditions come together to breedinnovation. The authors argue that ‘‘organi-zational highways, alleys, and by-lanes (p.107)’’ can be created within an organizationto spread innovation throughout the organi-zation. See Comment 4.7 for an example oforganizational innovation in the business ofbaseball.

Employee Attributes that Contributeto Creativity and Innovation

If innovation is viewed from the individualemployees’ perspective, a logical question is:Are there predictors of whether employeeswill be innovative? According to Amabile(1983), several variables are predictive ofcreative production in individuals. Becausecreativity and innovation are closely linked,


these variables are also relevant to predictinginnovation in organizational settings. Accor-ding to Amabile, creativity is due to task-relevant skills, creativity-relevant skills, andtask motivation.

The area of task-relevant skills is relatedto the previously discussed variable of gen-eral mental ability, but it is more than that.To be creative, an individual must have ahigh level of general mental ability, but mustalso have more specific abilities. For exam-ple, a scientist developing a new vaccinemust not only be intelligent, but must also

know specific information about the behav-ior of microorganisms and be able to applythis knowledge in his or her work. Specificknowledge and technical skills are depend-ent on a certain level of general mental abil-ity. Often, however, individuals mustacquire these through some type of formaleducation; for example, most successful sci-entists have completed graduate training intheir respective fields. Creative talent mayalso be developed apart from formal educa-tion. In the creative arts, for example, manysuccessful people learn through informal

COMMENT 4.7

MONEYBALL APPLIED TO INNOVATION IN ORGANIZATIONS

MICHAEL LEWIS, IN his 2004 book ‘‘Moneyball:

The Art of Winning and Unfair Game,’’

describes the art of managament and winning

in the world of baseball. Although the bookwas written about the business of recruiting

and managing talented baseball players, a

number of organizational scholars recognized

the relevance of the principles described in the

book to the overall field of human resource

management and organizational innovation. A

lead article in Human Resource Management

(2006) was devoted to the implications of thebook for organizations, and several other

organizational scholars commented on this

lead article.

Perhaps the biggest lesson to be gained

from ‘‘Moneyball’’ is the importance of not

maintaining practices that are based purely

on speculation, intuition, and tradition, but

recognizing the need to be innovative to dealwith the demands facing the organization.

Lewis tells the story of how general manager

Billy Beane turned the Oakland Athletics into

a highly effective team with a third as much

money as the wealthiest baseball teams. One

of the underlying themes is how Beane had to

be innovative in order to identify, acquire, and

keep talent in ways that other baseball teams

with more money did not. Beane used a num-

ber of principles from psychology in order tomake decisions for which baseball players to

recruit, including the best predictor of future

performance being past performance and sta-

tistical indicators of performance being supe-

rior to hunches or intuition in identifying

players who would excel. Although this

approach seems reasonable, at the time it rep-

resented a strong deviation from the statusquo of identifying talented players, and was

viewed as a highly radical approach. Of

course, the method ultimately paid off and

the Oakland A’s thrived as a baseball team

and organization. The story of Moneyball

has implications for how organizations in gen-

eral recruit and retain employees, emphasiz-

ing the use of objective data as opposed tosubjective perceptions when making deci-

sions.

Source: M. Lewis. (2004). Moneyball: The art of winning

an unfair game. New York, NY: W.W. Norton & Com-

pany.


means such as one-on-one tutoring, or theymay even be self-taught.

Despite the importance of task-relevantskills, many people possess them but do notproduce creative, innovative work. Forexample, despite the large number of indi-viduals holding the doctoral degree in indus-trial/organizational psychology and relatedfields (e.g., Organizational Behavior, HumanResource Management), a relatively smallproportion become highly productive re-searchers (e.g., Long, Bowers, Barnett, &White, 1998; Ones & Viswesvaran, 2000).Keep in mind that individuals holdingPh.D. degrees in these fields all have reason-ably equivalent education and training, andhave achieved a certain level of competencein their specialty. Why, then, are somehighly productive while others are not? Theanswer to this question may lie in the area ofcreativity-relevant skills and task motivation.

Creativity-relevant skills are essentiallymeta-skills that individuals use in the creativeprocess. One crucial skill in the creative pro-cess is a cognitive style that is conduciveto creativity. According to Amabile (1983),creative people are able to understand thecomplexities in a problem and are able tobreak set during problem solving. Stated dif-ferently, being creative requires being able tosee a problem from multiple perspectivesand having the willingness needed to breakthe mold in order to solve a problem. A goodhistorical example of this principle not beingapplied can be seen in retrospective accountsof the Vietnam War (McNamara, Blight,Brigham, Biersteker, & Schandler, 1999).In hindsight, it is clear that American andNorth Vietnamese decision makers viewedthe conflict from completely different per-spectives and were unwilling to deviate fromthese perceptions. On the American side,Vietnam was viewed as the ‘‘First Domino’’in a Communist plan to dominate Southeast

Asia. The North Vietnamese, on the otherhand, equated American intervention withthe colonialism of the French. If either sidehad been willing to deviate from these per-spectives, it is possible that the conflict couldhave been settled before the war escalated toa level that was so destructive for both sides.

Another important creativity-relevantskill is a work style that is conducive tocreativity. Creative people are able to con-centrate their efforts on a given problem forlong periods of time. Stated differently, crea-tivity requires hard work. Creative people,for example, are often able to work longhours at a time without stopping. Anotheraspect of work style is that creative peopleare able to engage in what Amabile (1983)described as productive forgetting—the abilityto abandon unproductive searches, andtemporarily put aside stubborn problems.Clear examples of this can be found in thesciences, where breakthroughs are typicallyachieved only after many failures.

The creativity-relevant skills describedup to this point may be acquired from train-ing, but there are more dispositional factorsthat contribute to creativity. Although resear-chers have been unable to isolate a creativepersonality, some personality traits do seemto be associated with creative activity.These include self-discipline, ability to delaygratification, perseverance in the face offrustration, independence, an absence ofconformity in thinking, and lack of depend-ence on social approval.

The issue of task motivation has not beenexamined extensively in creativity research,largely because of the strong focus on intrin-sic factors associated with creativity. It islikely, however, that at least some of thevariation in creativity can be explained bythe level and nature of the motivation onehas toward the task being performed. Ac-cording to Amabile (1983), creativity


requires that individuals genuinely enjoywhat they are doing, and perceive that theyare performing the task because they want to,rather than because of external pressures.These perceptions of enjoyment and intrin-sic motivation depend on one’s initial level ofintrinsic motivation toward the task beingperformed, the presence or absence of exter-nal constraints in the social environment,and the individual’s ability to block out orminimize external constraints.

Organizational Determinantsof Creativity and Innovation

Given our discussion of the determinants ofcreativity in individuals, what can organiza-tions do to foster creativity and innovationamong employees? The short answer to thisquestion is to hire creative people. Althoughthis suggestion makes sense, there are otherthings organizations can do. For example,to enhance creativity-relevant skills, organi-zations can provide training in the use ofcreative problem-solving methods such asbrainstorming. A typical activity in such atraining program might be for participantsto come up with as many different uses fora paper clip as they can think of in 5 minutes(there are actually quite a few, if you thinkabout it!). Such forms of training will obvi-ously not completely compensate for a lackof innate ability; however, they may helptalented employees realize their creativepotential.

Another way that organizations can fostercreativity and innovation is through influ-encing task motivation. A more comprehen-sive discussion of motivation is contained inChapter 8 but in the present context, thereappear to be things organizations can doto enhance task enjoyment and intrinsicmotivation. One way is to attempt to placeemployees into jobs that they genuinely

enjoy. This is not always possible, but if itcan be done, it can lead to higher levels ofcreativity. Another way organizations canenhance task motivation is through the iden-tification and removal of external constraints(Peters & O’Connor, 1988). Even thoughsome individuals may be able to temporarilycircumvent external constraints, employeesstand a greater chance of developing theintrinsic motivation necessary to be creativeif they are not there in the first place.

Kauffeld, Jonas, and Grote (2004)attempted to develop a structured measure-ment tool to assess an organization’s climatefor innovation. The authors found fourmajor factors: activating leadership (havingleaders that encouraged and modeled inno-vation), continuous questioning (encourag-ing employees to always question currentpractices), consequential implementation(seeing that the implementation of inno-vations had real consequences for theemployee and organization), and profession-al documentation (clearly indicating anddescribing the innovation). The authorsfound that a climate for innovation wasrelated to variables such as quality of devel-oped solutions for organizational problemsand product innovations.

Amabile and Conti (1999) have alsoextended Amabile’s work on individual cre-ativity into organizations, and explicitly rec-ognizes the importance of the organizationalcontext in facilitating the creativity ofemployees. She argues that five environmen-tal factors in organizations can contributeto creativity in employees: encouraging crea-tivity, autonomy and freedom, resources(the opposite of removing constraints), pres-sures (increasing positive challenges andreducing such factors as workload), andobstacles to creativity (e.g., conservatismand conflict). Amabile, Conti, Coon, Laz-enby, and Herron (1996) found that the


presence of facilitating factors is related tothe creativity of Research and Developmentprojects in a high-tech company.

As stated earlier, much of the innovationliterature has adopted a macro focus; that is,researchers have focused on identifyingcharacteristics of organizations that facilitateor impede the adoption or diffusion of inno-vation in those organizations. The most com-prehensive examination of organizational-level predictors of the adoption of innovationwas a meta-analysis by Damanpour (1991),in which he combined data from 23 studies.Before describing the findings from thismeta-analysis, it is important to note thatDamanpour distinguished between technicalinnovations and administrative innovations.Technical innovations pertain to innovationsin products, services, and production pro-cess technology. An organization adopting anew production process would be adopting atechnological innovation. Administrativeinnovations focus on organizational struc-ture and administrative processes. An exam-ple of this would be an organization’sdecision to switch to a team-based organiza-tional structure.

The results of this study suggest there areseveral organizational-level predictors ofinnovation. The strongest predictor, not sur-prisingly, was technical knowledge re-sources. Organizations are more likely toadopt innovations when they have employ-ees who possess the technical expertise tounderstand and facilitate the implementa-tion process. A possible explanation for thisfinding is that without technical expertise,there would be no innovations for organiza-tions to adopt in the first place. Thus, anorganization needs to hire individuals withhigh levels of technical knowledge.

The second most powerful predictor ofinnovation was the organization’s level ofspecialization. An organization that is highly

specialized, such as the manufacturer of asmall number of products, likely has indi-viduals with high levels of technical exper-tise. Having many technical specialistssimply brings more talent to bear on impor-tant problems and may facilitate the cross-fertilization of ideas, both of whichultimately lead to innovation.

A third notable predictor of innovationidentified in this meta-analysis was the levelof external communication in an organiza-tion. Examples of this predictor would betechnical experts’ presenting their researchfindings at conferences and sharing theirideas with individuals in other organiza-tions. Organizations that encourage fre-quent communication with the externalenvironment are likely to increase the chan-ces of bringing in innovative ideas from theoutside. External communication also pro-vides members of organizations with anopportunity to test the validity of their ideason those outside of the organization. Forthose in many technical specialties, externalcommunication may in fact be the only wayto obtain unbiased feedback on their ideas.One recent study found that employees inproject-based and knowledge-intensive orga-nizations are more innovative at work whenthey are embedded in social structures rele-vant to a given project or area of knowledgeoutside the organization (Staber, 2004).

A fourth predictor of innovation wasidentified as functional differentiation. A highlevel of functional differentiation simplymeans that distinct and identifiable func-tional specialties exist within an organiza-tion. As an example, an organization with ahigh degree of functional differentiation mayhave a research and development divisionwith a departmental structure based on tech-nical specialties. A high level of functionaldifferentiation leads to innovation becausegroups of employees who belong to the same


functional specialty are better able to elabo-rate on ideas and hence to develop innova-tions. In many cases, this is helpful becausespecialty-based coalitions may help to facili-tate administrative changes and innovations.

The four variables described previouslywere the strongest predictors of innovationidentified in this meta-analysis. Other lesspowerful, though statistically significant,predictors of innovation were professional-ism (.17), centralization (�.16), managerialattitudes toward change (.27), administra-tive intensity (.22), slack resources (.14),and internal communication (.17). These re-sults suggest that innovation is fostered byemployees who have a strong identificationwith their profession, a low level of central-ization, positive managerial attitudes towardchange, a high concentration of administra-tive employees, available slack resources,and a high level of communication.

More recent research has supported thevalidity of these factors as predictors. Thomas,McDonnell, McCulloch, While, Bosanquet,and Ferlie (2005) examined innovation inPrimary Care Groups within the healthcaresystem in England. These authors found thatan organization’s capacity for innovation wasdriven by such factors as having multipleopportunities for employees to reflect andlearn and communicate with other employ-ees within the organization, having bothclinicians and managers in positions of lead-ership to provide multiple perspectives, andbeing sure to time a particular innovativeinitiative correctly given the demands facinga particular work group. Caldwell andO-Reilly (2003) conducted interviews withsenior executives and found that factors suchas support for risk-taking, being tolerant ofmistakes, high levels of teamwork, and beingable to implement decisions quickly wereseen to lead to increased innovation (see alsoMohamed, 2002).

Given these findings, organizationswishing to encourage innovative behaviorcertainly need to recruit and hire the besttechnical talent possible. It is also importantthat organizations allow talented individualsto communicate with others outside of theorganization, to develop and test ideas. Thiscan be done through a variety of mecha-nisms: attending professional conferences,publishing in peer-reviewed journals, and,in some cases, bringing in experts from theoutside. Ironically, some organizations arehesitant to do this, for fear that externalcommunication will compromise proprie-tary information. This is particularly truefor organizations operating in highly com-petitive industries (e.g., consumer products,food). This is a valid concern, but one couldargue that the potential benefits of suchforms of external communication far out-weigh the risks.

Influencing managerial attitudes towardchange is a complicated issue, but an orga-nization can approach it in several ways. Oneway is to select management employees whohave positive attitudes toward change. Thismay be difficult if the assessment must bedone during the hiring process. Anotherapproach may be to influence managementattitudes through training and developmentactivities. Ultimately, the most powerfulinfluence on attitudes toward change is theway managers are treated. In many organi-zations, employees are punished for or dis-couraged from trying new things. Thus, thebest way to improve attitudes toward changemay be to encourage managers to try newthings and to take risks. By doing this, orga-nizations can take the threat out of change.Consequently, managers themselves may bemore receptive to change and innovation.

This latter point emphasizes the impor-tance of leadership in positive attitudes andactions toward innovation. Mumford and


Licuanan (2004) discussed the importance ofleaders in facilitating the teamwork necessaryfor creative innovation to occur, andreviewed research showing that leaderlessteams are less effective because of this lackof facilitation. These authors point out thatleaders who are effective at encouraging inno-vation and creativity tend to have high levelsof technical expertise and creative thinkingskills. Finally, the authors note that becauseemployees involved in innovative work tendto be more intrinsically motivated to performtheir tasks, the primary role of leaders is tocreate the conditions that can funnel motiva-tion to a given area of creativity.

CHAPTER SUMMARY

In this chapter, we examined productivebehavior, or employee activities that contrib-ute to the goals of the organization. The mostcommon form of productive behavior inorganizations is job performance, and thishas been studied extensively for a numberof years. There have even been attempts todescribe dimensions of performance that arecommon to most jobs. Such efforts to modeljob performance continue to evolve, andthey hold great promise in helping us tounderstand the substantive nature of jobperformance. Although there are differencesbetween proposed models, one distinctionthat cuts across all of them is in-role (tech-nical aspects of a given job) versus extra-role(skills that transcend the specific content of ajob such as communication skills and being ateam player) performance.

Because of its complexity, a number offactors complicate the attempts to measurejob performance. These include the amountof instability in job performance over timeand the fact that a number of forces tend torestrict the variability in job performance

within organizations. Despite all of thesecomplicating factors, organizational research-ers have still learned a great deal about thedeterminants of job performance. Researchaccumulated over the years has led to theconclusion that three variables stand out aspredictors of performance, regardless of thejob: (1) general cognitive ability, (2) level ofjob experience, and (3) the personality trait ofconscientiousness. Furthermore, these varia-bles appear to influence performance largelythrough the acquisition and utilization of jobknowledge and the motivation to performwell.

Organizational Citizenship Behavior(OCB) represents the second form of pro-ductive behavior examined in the chapter.Although it can take several forms, OCB isdefined as behavior that is not part ofemployees’ formal job responsibilities.Research has shown that employees engagein OCB primarily because of positive affectand perceptions of the level of fairness withwhich they are treated by the organization.Only recently have researchers begun toempirically examine the assumption thatOCB enhances organizational performance,to question the in-role/extra-role distinctionthat lies at the heart of OCB, and to probe theunderlying motivation for the performanceof OCB.

The third form of productive behaviordiscussed was innovation. We examinedthe characteristics of individuals who arelikely to engage in innovative or creativebehavior, and we explored macro influenceson the innovation process. Drawing onindividual-level studies of creativity, itappears that creativity and innovation canbe explained on the basis of domain-relevantskills, creativity-relevant skills, and taskmotivation. Macro-level studies suggest sev-eral influences on the innovation process in



JOHN CAMPBELL AND THE MODELING OF JOB PERFORMANCE

For 100 years the focus of I/O Psychology hasbeen on two major dependent variables, indi-

vidual job performance and individual job

satisfaction. While hundreds of studies have

been devoted to the modeling and measure-

ment of job satisfaction, no such literature

existed for performance itself, until recently

(i.e. after 1990). We had no ‘‘theory of perfor-

mance.’’ Performance measures typically were‘‘criteria of convenience,’’ and virtually all

available indicators (e.g. college professor

grant dollars or realtor sales volume) are sus-

ceptible to many sources of variation besides

the actions of the individual. An outlandish

example was a suggestion by a researcher to

use ‘‘death rates’’ as a measure of physician

performance. Over a 20-year period my goalwas to correct this situation and give our most

important dependent variable its due as a

scientific construct.

Two projects made this possible. First, my

colleagues and I compiled every study ever

done on the determinants of, and the meas-

urement of, individual manager performance.

It resulted in the 1970 book, Managerial

Behavior, Performance, and Effectiveness, whichmade a clear distinction between (a) the things

managers actually do, and (b) the outcomes of

what they do (i.e. the bottom line), which are

virtually always influenced by many things

besides the manager’s behavior.

The second was the largest project in the

history of applied psychology (Project A),

which dealt with the U.S. Army’s selectionand classification system. Three years were

devoted to development of multiple measures

of performance, using every known measure-

ment technology. Two cohorts of 10,000 new

recruits were followed for six years. Perfor-

mance was assessed at the end of technical

training, after 3 years, and after 6 years on the

job.With this much data it was possible to

model empirically the substantive nature of

performance in this population of occupa-

tions. I then proposed a comprehensive model

of individual performance for all jobs (e.g.

Campbell, McCloy, Oppler, & Sager, 1992).

This was the first model of its kind, and

the intent was to provide a meaningful spec-ification of performance that can guide

research, inform human resource practices,

and provide a framework for integrating the

existing literature. It felt good then. It still

does.

John P. Campbell


University of Minnesota

Chapter Summary 129

organizations. The most general predic-tors of innovation appear to be technicalknowledge resources, external communica-tion, and managerial attitudes towardchange. As with individual-level attributes,organizations have several levels of influenceat the macro level in order to encourageboth the development and adoption of inno-vation.


Barrick, M. R., & Mount, M. K. (2005). Yes,personality matters: Moving on to moreimportant matters. Human Performance, 18,359–372.

Hellstrom, C., & Hellstrom, T. (2002). High-ways, alleys and by-lanes: Charting thepathways for ideas and innovation in orga-nizations. Creativity and Innovation Manage-ment, 11, 107–114.

Podsakoff, P. M., & MacKenzie, S. B. (1997).Impact of organizational citizenship behav-ior on organizational performance: A reviewand suggestions for future research. HumanPerformance, 10, 133–151.

Ree, M. J., Earles, J. A., & Teachout, M. S.(1994). Predicting job performance: Notmuch more than g. Journal of Applied Psychol-ogy, 79, 518–524.

Sackett, P. R., Gruys, M. L., & Ellingson,J. L. (1998). Ability-personality interactionswhen predicting job performance. Journal ofApplied Psychology, 83, 545–556.

Viswesvaran, C. (2002). Assessment of indi-vidual job performance: A review of the pastcentury and a look ahead. In N. Anderson,D. S. Ones, H. P. Sinangil, & C. Viswesvaran(Eds.), Handbook of industrial, work, and orga-nizational psychology, Volume 1: Personnelpsychology. Thousand Oaks, CA: Sage Publi-cations.


Chapter Five

Job SatisfactionandOrganizationalCommitmentE

veryday experience suggests thathumans are evaluative creatures;they look at much of their experi-ence in terms of liking and dislik-ing. Most of us, for example, have

developed very clear preferences regardingthe people we socialize with, the activities weengage in, and even the foods we choose toeat. In the workplace, this propensity forevaluation leads employees to develop feel-ings of liking or disliking toward the jobsthey are performing. Most people have someopinion, be it positive or negative, about theirjob and the organization in which they work.

One could argue that another human ten-dency is to develop feelings of attachment orcommitment. Indeed, many of us developfeelings of commitment toward other people,ideas, and even institutions. In the workplace,this tendency is manifested as employees’level of commitment toward the employingorganization. Employees may be committedto their employing organizations for varyingreasons, but there is no doubt that suchfeelings of commitment have important con-sequences for both the individual employeeand the organization as a whole.

In this chapter, we cover two topics thatmany believe lie at the core of organizationalpsychology: (1) job satisfaction and (2) orga-nizational commitment. Job satisfactionessentially represents employees’ overall eval-uation of their jobs, including feelings of pos-itive affect toward their jobs or job situations.Organizational commitment, which is closelyrelated to job satisfaction, represents employ-ees’ feelings of attachment and loyalty towardan organization. Both of these variables have

been studied extensively in organizationalpsychology, largely because they are relatedto a number of outcomes that are importantfor both theoretical and practical reasons.

JOB SATISFACTION

Job satisfaction is, without a doubt, one ofthe most heavily studied topics in organiza-tional psychology, as well as in the broaderfield of industrial/organizational psychology.To emphasize this point, many authors havereferred to Locke’s chapter in the Handbookof Industrial and Organizational Psychology

131

(1976), where he reported that studies deal-ing with job satisfaction numbered in thethousands. That was approximately 25 yearsago, so the figure cited by Locke has grownconsiderably. A recent search of Psychinfowith the keywords ‘‘Job Satisfaction’’ revealed21,375 references on the topic. Incidentally,this high level of research attention has notescaped the notice of many inside and out-side the field of I/O psychology. For example,the first author can remember as a graduatestudent one of the nonI/O faculty stating thatI/O psychology was defined as ‘‘One hun-dred and one ways to ask people how theylike their jobs. . . . ’’ Although this individ-ual was being a bit facetious (he was actuallybeing very facetious), there is certainly a grainof truth in his statement.

Defining Job Satisfaction

Job satisfaction can be defined as an employ-ee’s overall evaluation of his or her job asfavorable or unfavorable (e.g., Locke, 1976;Spector, 1997a). In this sense, job satisfac-tion is basically an employee’s attitudetoward his or her job (see Eagly & Chaiken,1993). Like all attitudes, job satisfaction con-sists of feelings, thoughts, and behaviorsindividuals have relevant to their jobs. Mostauthors emphasize the emotional com-ponent of job satisfaction. However, cogni-tive and behavioral components of jobsatisfaction are important aspects of the con-struct as well.

The cognitive aspect of job satisfactionrepresents an employee’s beliefs about his orher job or job situation; that is, an employeemay believe that his or her job is interesting,stimulating, dull, or demanding—to namea few options. Note that although these re-present cognitive beliefs, they are notcompletely independent of the previouslydescribed affective component. For example,

a statement or belief that ‘‘My job is interest-ing’’ is likely tobe strongly related to feelings ofpositive affect.

The behavioral component represents anemployee’s behaviors or, more often, behav-ioral tendencies toward his or her job. Anemployee’s level of job satisfaction may berevealed by the fact that he or she tries toattend work regularly, works hard, andintends to remain a member of the organiza-tion for a long period of time. Compared tothe affective and cognitive components ofjob satisfaction, the behavioral componentis often less informative because one’s atti-tudes are not always consistent with one’sbehavior (Fishbein, 1979). It is possible, forexample, for an employee to dislike his orher job but still remain employed therebecause of financial considerations.

Measurement of Job Satisfaction

Given the importance of job satisfaction toorganizational psychologists, it is crucial todevelop measures of the construct that arereliable and valid. As we learned with thestudy of job performance, it is impossibleto study something if you can’t measure it.Fortunately for organizational psychologists,several reliable and valid measures of jobsatisfaction are available for their use. In thissection, fourof the mostwidelyusedmeasuresare described. However, before describingspecific measures, we briefly review the pro-cess by which measures come to be seen asvalid.

Although the measures that are describedin this section are viewed by many as con-struct valid measures of job satisfaction, it isreally incorrect to say that any measure is oris not construct valid. Construct validity is amatter of degree. Research has strongly sup-ported the construct validity of the measuresdescribed in this section—in most cases, this

132 Job Satisfaction and Organizational Commitment

evidence has been accumulated over severaldecades. Because of this accumulated evi-dence, researchers can use these measureswith a great deal of confidence that they areindeed measuring employees’ levels of jobsatisfaction.

How do we provide evidence for the con-struct validity of a measure? In general, thereare three tests of construct validity (Campbell& Fiske, 1959; Nunnally & Bernstein, 1994).First, for a measure to be construct valid, itmust correlate highly with other measures ofthe same construct. Stated differently, a meas-ure must exhibit convergence with other meas-ures of the same construct. For example, iftwo different organizational researchersdevelop separate measures of job satisfactionwith different items, these two measuresshould correlate highly with each other. Asecond test of construct validity is that ameasure must be distinct from measures ofother variables. Another name for this is dis-crimination. For example, we may want toshow that job satisfaction is not simply afunction of differences between people intheir tendency to evaluate everything in theirworld either positively or negatively. A thirdway that researchers typically show evidenceof construct validity is through theoreticallygrounded predictions; that is, researcherstypically develop a theoretically based nomo-logical network of proposed relationshipsbetween the measure being developed andother variables of interest. To the extent thatthese relations are supported, the constructvalidity of the measure is supported. Forexample, much of the research we review inthe following paragraphs illustrates the theo-retical relationships between job satisfactionand other variables, such as job characteristicsand personality. Given the thousands of stud-ies conducted on job satisfaction, the nomo-logical network for the construct is quiteelaborate.

Several measures are widely considered tobe construct valid measures of job satisfac-tion. Again, they are not construct valid in anabsolute sense. Rather, so much favorableevidence has accumulated over the years thatthey are widely accepted measures of the jobsatisfaction construct. Given the large numberof construct valid job satisfaction measurescurrently in use, a comprehensive coveragewould be beyond the scope of this chapter.However, a handful of job satisfaction meas-ures have been used widely over the years.Four of these are described in this section.

One of the first measures of job satisfac-tion that enjoyed widespread use was theFaces Scale developed by Kunin in themid1950s (Kunin, 1955). As can be seen inFigure 5.1, this scale consists of a series offaces with differing emotional expressions.Respondents are asked simply to indicatewhich of the five faces best represents theirfeelings of overall satisfaction toward the job.The primary advantages of the Faces Scale areits simplicity and the fact that respondentsneed not possess a high reading level in orderto complete it. This would be an excellentscale to use, for example, if a researcher weresurveying a sample of employees who wereknown to have a very low level of education.

FIGURE 5.1The Faces Scale of Job Satisfaction

Put a check under the face that expresses how you feel aboutyour job in general, including the work, the pay, the supervision,the opportunities for promotion, and the people you work with.

Source: T. Kunin. (1955). The construction of a new type of

attitude measure. Personnel Psychology, 8, 65–67. Reprinted

by permission.


A potential disadvantage of the Faces Scaleis that it does not provide the researcher withany information about an employee’s satisfac-tion with different facets of the job. If anemployee endorses one of the lower valueson the Faces Scale (a ‘‘Frown’’), this does nottell the researcher whether the source of thisdissatisfaction is pay, supervision, or the con-tent of the work itself. Thus, if a researcher isinterested in pinpointing the source of satis-faction or dissatisfaction, the Faces Scale is ofmore limited value.

Another scale that has enjoyed extremelywidespread use is the Job Descriptive Index(JDI) developed in the late 1960s by PatriciaCain Smith and her colleagues at CornellUniversity (Smith, Kendall, & Hulin,1969). Sample items from the JDI are pre-sented in Table 5.1. One thing to noticeimmediately is that the JDI is aptly namedbecause the scale does require that respond-ents describe their jobs. Also, in contrast tothe Faces Scale, users of the JDI obtain scoresfor various facets of the job and the workenvironment. The JDI provides scores for theindividual facets of work, pay, promotionopportunities, supervision, and coworkers.Although some users of the JDI combine

the facet scores to form an overall satisfactionindex, this practice is not recommended bythe developers of the JDI.

The primary advantage of the JDI is that agreat deal of data supports its construct va-lidity. Furthermore, research still continuesin an effort to improve this scale (see Com-ment 5.1). Thus, the initial development andcontinued research on the JDI are exemplary.One consequence of this longstandingresearch effort is that considerable normativedata on the JDI have been accumulated overthe years. Thus, if a researcher or consultantwere to use the JDI to measure job satisfac-tion among a sample of nurses, he or shewould be able to compare their scores to anormative sample from the same occupation.Norm group comparisons can often beextremely useful if top managers want toknow how the satisfaction levels of theiremployees compare to those of employeesin similar occupations or employees withinthe same industry.

Given the vast amount of research asso-ciated with the JDI, not many disadvantagesare associated with this scale. However, oneissue has come up with the JDI: the lack of anoverall satisfaction scale. As stated earlier, in

TABLE 5.1Sample Items from the Job Descriptive Index (JDI)

Think of your present job. In the blank beside each word of phrase, write:

Y for ‘‘Yes’’ if it describes your job

N for ‘‘No’’ if it does not describe your job

? if you cannot decide

Work Pay Promotions

——Fascinating ——Barely live on income ——Opportunities somewhat limited

——Pleasant ——Bad ——Promotion on ability

——Can see results ——Well paid ——Regular promotions

Source: P. C. Smith, The Job Descriptive Index, Revised. Copyright, 1975, 1985, 1997, Bowling Green State University. Licensing

for the JDI and related scales can be obtained from: Department of Psychology, Bowling Green State University, Bowling

Green, Ohio 43403. Reprinted by permission.


some cases, researchers merely wish to meas-ure employees’ levels of overall satisfaction,and the JDI does not allow for this. Toaddress this issue, developers of the JDI cre-ated what is termed the Job in General (JIG)Scale (Ironson, Smith, Brannick, Gibson, &Paul, 1989). The JIG is modeled after the JDIexcept that it consists of a number of adjec-tives and phrases about the job in generalrather than about specific job facets. The JIGhas been used less frequently than the JDI,but some recent research has contributedto the scale’s validity (Skibba & Tan,2004). In addition, Russell and his col-leagues (2004) have recently developed ashorter form of the measure that should beappealing to organizational researchers.

A third job satisfaction measure thathas enjoyed widespread acceptance and usewithin organizational psychology is theMinnesota Satisfaction Questionnaire (MSQ).The MSQ was developed by a team of re-searchers from the University of Minnesotaat roughly the same time the JDI was beingdeveloped (Weiss, Dawis, England, & Lof-quist, 1967). The long form of the MSQconsists of 100 items that are designed tomeasure the 20 facets of work that are pre-sented in Table 5.2. There is also a short formof the MSQ, consisting of 20 items. The shortform, however, is not designed to providefacet satisfaction scores.

The items comprising the MSQ consist ofstatements about various facets of the job,

COMMENT 5.1

THE LEGACY OF THE JDI

THE JOB DESCRIPTIVE Index (JDI) is undoubtedly

one of the most popular and widely used

measures of job satisfaction. One of the rea-

sons for such wide use is the considerableresearch and development that has gone,

and continues to go, into this instrument.

The JDI was developed by Patricia Cain Smith

and colleagues at Cornell University in the

early 1960s. When Dr. Smith relocated from

Cornell to Bowling Green State University in

the mid-1960s, she founded the JDI research

group, which consisted of both faculty andgraduate students.

Over the years, the JDI research group has

conducted research aimed at further refine-

ment of the instrument, as well as develop-

ment of national norms. This group continues

such efforts to the present day, and has devel-

oped a number of other measures that are

based on the JDI. The JDI research group atBowling Green has also served as a clearing-

house for research data using the JDI. As a

result, the group has assembled an impressive

data archive consisting of dozens of data sets,

collected over a 25-year period and containingover 12,000 cases. Recently, the group has

begun to make a portion of these data avail-

able to researchers outside of Bowling Green

who are interested in job satisfaction and

related areas such as occupational stress,

retirement, and job design.

Overall, the JDI represents one of the most

comprehensive and exemplary scale develop-ment efforts ever conducted in the field of

industrial/organizational psychology. It also

serves as a good example of how scale devel-

opers, and the entire field, can benefit by

making data available to the research com-

munity.

Source: http://www.bgsu.edu/departments/psych/JDI.


and the respondent is asked to indicate his orher level of satisfaction with each. For exam-ple, a respondent is presented with an itemhaving to do with activity level, such as‘‘Being able to keep busy all the time,’’ andis asked to indicate his or her level of sat-isfaction with the statement. Compared tothe JDI, the MSQ is more of an affectbasedmeasure; that is, responses indicate liking ordisliking rather than description.

Like the JDI, considerable research hasgone into the development and constructvalidation of the MSQ. The MSQ also pro-vides quite extensive information on employ-ees’ satisfaction with various facets of the jobor work environment. As stated earlier, thistype of information may be especially usefulwhen organizations are conducting internalemployee opinion surveys. For example, if itis found that satisfaction with a certain facetis much lower compared to the others, thissuggests that an organization may need tomakes changes in this area. The only majordisadvantage of the MSQ is its length. At 100items, the full version of the MSQ is verydifficult to administer, especially if a re-searcher wishes to measure other variables.

Even the shortened version (20 items) is stillconsiderably longer than many other meas-ures of satisfaction available.

A final job satisfaction measure that hasnot been used as extensively as the othersdescribed, but that has considerable evi-dence supporting its psychometric proper-ties, is the Job Satisfaction Survey (JSS). Thisscale was originally developed by Spector(1985) as an instrument to measure job sat-isfaction levels of Human Service employees.The JSS consists of 36 items designed tomeasure nine facets of the job and workenvironment. The facets measured by theJSS are listed in Table 5.3.

Compared to the other measuresdescribed in this section, the JSS is fairlytypical; that is, the items represent state-ments about a person’s job or job situation.Respondents are then asked to indicate theextent to which they agree with each item.Given this type of scaling, the JSS is moresimilar to the JDI because it is more descrip-tive in nature than the MSQ. Unlike the JDI,however, an overall satisfaction score can becomputed for the JSS by summing the facetscores.

TABLE 5.2A Listing of the Facets Measured by the Minnesota Satisfaction Questionnaire (MSQ)

Activity Ability utilization

Independence Company policies and practices

Variety Compensation

Social status Advancement

Supervision (human relations) Responsibility

Supervision (technical) Creativity

Moral values Working conditions

Security Coworkers

Social service Recognition

Authority Achievement

Supervision (human relations) Company policies and practices

Source: D. J. Weiss, R. V. Dawis, G. W. England, and L. H. Lofquist. (1967). Manual for the Minnesota

Satisfaction Questionnaire (Minnesota Studies in Vocational Rehabilitation, No. 22). University of

Minnesota, Minneapolis.


Compared to the JDI and MSQ, not asmuch supporting data are available for theJSS, but the evidence supporting the psycho-metric properties of this scale is still impres-sive (Spector, 1997a). Furthermore, Spectorhas assembled a fairly comprehensive nor-mative database for the JSS; it includes avariety of job types, many different organiza-tions, and even different countries.

In conclusion, a number of valid meas-ures exist to measure overall job satisfactionand satisfaction with different aspects ofone’s job. The normative information onmany of these scales is especially useful fororganizational psychologists who want to beable to compare their employees to employ-ees from similar occupations and organiza-tions. Now that we have addressed theprediction and measurement of job satisfac-tion, we address the important question ofoutcomes predicted by job satisfaction.

Predictors of Job Satisfaction

A substantial portion of the research con-ducted on job satisfaction over the yearshas been devoted to explaining what exactlydetermines employees’ levels of job satisfac-tion. Understanding the development of jobsatisfaction is certainly of theoretical impor-tance to organizational psychologists. It is

also of practical interest to organizations asthey attempt to influence employees’ levels ofjob satisfaction and, ultimately, other impor-tant outcomes.

There are three general approaches to ex-plaining the development of job satisfaction:(1) job characteristics, (2) social informationprocessing, and (3) dispositional approaches.According to the job characteristics app-roach, job satisfaction is determined primar-ily by the nature of employees’ jobs or by thecharacteristics of the organizations in whichthey work. According to this view, employ-ees cognitively evaluate their jobs and orga-nizations and make some determination oftheir relative level of satisfaction.

Over the years, several models have beenproposed to explain the precise manner inwhich job satisfaction develops in responseto job conditions (see Hulin (1991) for asummary). There are differences amongthese models, but the common theme run-ning through most of them is that job sat-isfaction is largely determined by employees’comparison of what the job is currently pro-viding them and what they would like it toprovide. For each facet of a job—pay, work-ing conditions, supervision—employeesmake some assessment of what they arecurrently receiving. These assessments aremeaningful only when they are comparedwith what an employee feels he or she shouldbe receiving from a particular facet. Theseperceptions are based on a number of factors:the employees’ skills, the amount of timethey have put into the job, and the availabil-ity of other employment opportunities.

If employees perceive what they are cur-rently receiving to be at or above what theyfeel they should be receiving, then they aresatisfied. If not, then feelings of dissatis-faction are evoked. As a relatively simpleexample of how this works, suppose an em-ployee’s current yearly salary is $42,000. If

TABLE 5.3A Listing of the Facets Measured by the JobSatisfaction Survey (JSS)

Pay Promotion

Supervision Benefits

Contingent rewards Operating procedures

Coworkers Nature of work

Communication

Source: P. E. Spector. (1997). Job satisfaction: Application,

assessment, causes, and consequences. Thousand Oaks, CA:

Sage. Reprinted by permission of Sage Publications, Inc.


the employee believes that he or she should bereceiving an annual salary of approximately$40,000, then the salary will evoke feelings ofsatisfaction. On the other hand, if the employ-ee believes, for some reason, that he or shedeserves an annual salary of $100,000, thenthe current salary will evoke feelings of dis-satisfaction. The importance of comparisonsin pay satisfaction can be seen in a recentmeta-analysis by Williams, McDaniel, andNguyen (2006). These authors examined theantecedents and consequences of pay levelsatisfaction by combining the results of over203 studies. The authors found that one of thebiggest predictors of pay level satisfaction washow well an employee’s current pay com-pared to other employees in the same organi-zation. Clearly, employees base theirsatisfaction with their pay on comparisonswith others. What makes these findings evenmore convincing is that the actual pay ofemployees was controlled in the analyses.

The notion that job satisfaction dependson an employee’s comparison of what he orshe is currently receiving versus what isdesired is reasonable. However, accordingto Locke (1976), this is an oversimplificationbecause it does not account for the fact thatemployees differ in the importance theyplace on various facets of work. For oneemployee, it may be extremely important tohave pay and fringe benefits that meet his orher expectations; for another, it may beessential to have a job that provides anopportunity for challenging assignments.

To explain how such differences influ-ence the development of job satisfaction,Locke (1976) proposed what has becomeknown as range of affect theory. The basicpremise of range of affect theory is that facetsof the work are differentially weighted whenemployees make their assessments of jobsatisfaction. For example, if pay is veryimportant to an employee, the fact that his

or her current pay is close to what wasexpected would have a large positive impacton his or her overall assessment of job sat-isfaction. In contrast, if pay is relativelyunimportant, the fact that expectations weremet or unmet would have a relatively smallimpact on employee job satisfaction.

The job characteristics approach to jobsatisfaction is strongly ingrained in organiza-tional psychology (e.g., Campion & Thayer,1985; Griffin, 1991; Hackman & Oldham,1980). Furthermore, the weight of empiricalresearch from a variety of areas stronglysupports the idea that characteristics of thejob and the job situation are robust predic-tors of employees’ levels of job satisfaction(e.g., Ellickson, 2002; Fried & Ferris, 1987;Williams et al., 2006). Thus, by the mid1970s,the job characteristics approach had clearlybecome entrenched as the dominant approachto job satisfaction within organizational psy-chology.

The first major challenge to the job char-acteristics approach came in the late 1970s inthe form of Social Information Processing(SIP) theory (Salancik & Pfeffer, 1977,1978). Salancik and Pfeffer criticized thejob characteristics approach to job satisfac-tion on two counts. First, they proposed thatthe job characteristics approach was inher-ently flawed because it was based on theassumption that job characteristics wereobjective components of the work environ-ment. According to these authors, jobs are‘‘social constructions’’ that exist in the mindsof employees and are not objective entities.Second, they pointed out that the job char-acteristics approach was based on the idea ofneed satisfaction. The problem with this,according to Salancik and Pfeffer, is that littleevidence has supported the utility of needsin the prediction of employee outcomes.

Salancik and Pfeffer (1978) proposed twoprimary mechanisms by which employees


develop a feeling of satisfaction or dissatisfac-tion. One of these mechanisms states thatemployees look at their behaviors retrospec-tively and form attitudes such as job satisfac-tion in order to make sense of it. This view isbased on Bem’s (1972) Self Perception Theory,which is a more general social psychologicaltheory of attitude formation. According tothis view, for example, an employee whohas been working in an organization for30 years may say to himself or herself, ‘‘I’veworked here for a long time, therefore I mustreally like my job . . . ’’ This view challengeswhether employees actually hold stable atti-tudes toward their jobs based on objectivefeatures of the environment, and insteadsuggests that employees’ job satisfaction mayfluctuate and be determined based on fea-tures of the situation.

The other explanation—the one mostclosely linked to social information process-ing theory—is that employees develop atti-tudes such as job satisfaction throughprocessing information from the social envi-ronment. This view is based largely onFestinger’s (1954) Social Comparison Theory,which states that people often look to othersto interpret and make sense of the environ-ment. According to this view, for example, anew employee who happened to interactwith other employees who were dissatisfiedwith their jobs would also likely becomedissatisfied. The practical implication of this,of course, is that organizations must be care-ful not to allow new employees to be taintedby dissatisfied employees during the social-ization process.

Within organizational psychology andother related fields, the initial developmentof Social Information Processing theory hada strong impact. This was undoubtedly dueto the fact that the job characteristics ap-proach had been dominant up to that point.As evidence of this impact, a flurry of

research activity designed to test this theorywas conducted in the late 1970s andthroughout the 1980s, (e.g., Adler, Skov, &Salvemini, 1985; O’Reilly & Caldwell, 1979;Weiss & Shaw, 1979; White & Mitchell,1979). Most of these investigations foundthat social information, usually in the formof verbal comments about task characteris-tics, had at least as powerful an impact on jobsatisfaction and perceptions of task charac-teristics as the objective characteristics of thetask. Field tests of Social Information Proc-essing theory, however, have been much lesssupportive than laboratory investigations(e.g., Jex & Spector, 1989).

Given the inability to demonstrate SocialInformation Processing effects outside of lab-oratory settings, it is tempting to concludethat this is nothing more than an interestinglaboratory phenomenon (e.g., Jex & Spector,1988). However, common sense and every-day experience suggest that social informationdoes play a role in the formation of our atti-tudes. If the impact of social influence isubiquitous, why then is the influence of socialinformation on job satisfaction so difficult todemonstrate outside of laboratory settings?According to Hulin (1991), laboratory inves-tigations of social information processingeffects are typically more successful than fieldstudies because they grossly simplify thesocial influence process. For example, in mostlaboratory studies, participants are giveneither ‘‘positive’’ or ‘‘negative’’ social informa-tion about the task they are being asked toperform. In organizational settings, employ-ees rarely receive such discrete levels of socialinformation about their jobs or organizations.For example, employees may receive socialinformation covering a variety of levels offavorability, and may at times receive conflict-ing information from the same source. In thefuture, organizational psychologists mustdevelop more creative ways of studying the


impact of social information on job satisfac-tion (see Comment 5.2).

One recent attempt to examine the influ-ence of the social context on ratings ofjob satisfaction examined how the variabilityin personality within an organization is relat-ed to employee reports of job satisfaction(Ployhart, Weekley, & Baughman, 2006).These authors found that employees work-ing in organizations where people differedwidely in their personalities reported lowerlevels of job satisfaction than employees whoworked in organizations where people weresimilar in their personalities, especially onthe traits of agreeableness and extraversion.The average level of a personality trait in anorganization was also related to job satisfac-tion. Employees reported higher job satisfac-

tion when they were in organizations whereemployees reported higher levels of emo-tional stability, conscientiousness, and extra-version. This study illustrates that the typesof people and similarities between people inthe employees’ environment are predictive ofjob satisfaction.

The most recent approach to explainingjob satisfaction is based on internal disposi-tions. The basic premise of the dispositionalapproach to job satisfaction is that someemployees have a tendency to be satisfied(or dissatisfied) with their jobs, regardlessof the nature of the job or organization inwhich they work. The use of dispositionsto explain behavior and attitudes is oftenportrayed as a very recent phenomenon,but the dispositional approach to job

COMMENT 5.2

SOCIAL INFORMATION PROCESSING RESEARCH

WHEN GERALD SALANCIK and Jeffrey Pfeffer

introduced the Social Information Processing

(SIP) approach to job satisfaction in the late

1970s, they caused a great deal of controversyamong job satisfaction researchers. The reason

for this controversy is that Salancik and Pfeffer

challenged the widely held belief that job

satisfaction was due primarily to characteris-

tics of the jobs and organizations in which

employees work. One of the results of this con-

troversy was that different camps developed—

those who favored the job characteristics ap-proach, and those who favored the social in-

formation processing approach.

As so often happens when different camps

develop, each tried to provide empirical evi-

dence supporting its position. Thus, in the late

1970s and early 1980s, a number of the lab-

oratory studies conducted essentially pitted

the job characteristics and social information

processing approaches against each other.

More specifically, researchers manipulated

characteristics of laboratory tasks and, at the

same time, provided social cues (usually byusing a confederate) about the desirability of

the task. The objective was then to see which

of these manipulations explained the most

variance in task satisfaction.

What many of these so-called race horse

design studies showed, not surprisingly, was

that the task satisfaction of laboratory subjects

was influenced by both task design and thesocial cues that were provided about the task.

Since that time, researchers have generally ac-

cepted the fact that both job characteristics and

social information have an impact on job sat-

isfaction. In the future, the key is to determine

the situations in which each of these (along with

dispositions) may exert the greatest impact.


satisfaction can actually be traced back to thework of Weitz (1952). Weitz was interestedin whether an individual’s general affectivetendencies would interact with job satisfac-tion to impact turnover. Thus, Weitz was notinterested in explaining job satisfaction bydispositions per se, but his work was clearlysuggestive of that notion.

The study that brought about renewedinterest in dispositions was Staw and Ross’s(1985) investigation of the stability of jobsatisfaction among a national sample ofworking males. This study found that therewas a statistically significant correlationbetween job satisfaction at one point in time,and job satisfaction 7 years later. Becausemany of those in the sample had changedjobs—and, in some cases, careers—theauthors argued that the level of stability thatwas found suggested that job satisfaction wasat least partially determined by dispositions.Subsequent research by Staw, Bell, andClausen (1986) provided even more impres-sive evidence of stability by showing that jobsatisfaction in adolescence was predictive ofjob satisfaction in adulthood.

Perhaps the most interesting evidence forthe dispositional approach to job satisfactionwas provided in a study conducted by Arvey,Bouchard, Segal, and Abraham (1989). Inthis study, the authors examined job satis-faction among pairs of monozygotic twinsand estimated the extent to which job satis-faction was similar within pairs. Using astatistic called the intraclass correlation coef-ficient, these authors found that approxi-mately 30% of the variance could beattributed to genetic factors. Although thisstudy was subsequently criticized on meth-odological grounds (e.g., Cropanzano &James, 1990), it is nevertheless consistentwith a dispositional approach to job satisfac-tion. In addition, other researchers haverecently replicated the heritability of job

satisfaction using a stronger methodology.Ilies and Judge (2003) estimated a heritabil-ity coefficient of .54 for job satisfaction, againindicating approximately 29% of the vari-ance in job satisfaction was a function ofgenetic factors.

A major limitation of early work on thedispositional approach to job satisfaction wasthat it was imprecise as to exactly which dis-positions are related to job satisfaction (Davis-Blake & Pfeffer, 1989). Recall that Staw andRoss (1985) demonstrated that job satisfac-tion was stable over time, but they did notspecify which dispositional traits accountedfor this consistency. Subsequent dispositionalresearch focused on documenting relationsbetween specific traits and job satisfaction.As an example, Levin and Stokes (1989)found that negative affectivity was negativelyassociated with job satisfaction, andexplained variance that was independent ofjob characteristics (see also Connolly & Vis-wesvaran, 2000). Negative affectivity is a dis-positional trait having to do with thepredisposition to experience negative emo-tionality and distress (Watson & Clark,1984). It has also been found that positiveanalogues to negative affectivity, such as dis-positional optimism and positive affectivity, arepositively related to job satisfaction (e.g.,Connolly & Viswesvaran, 2000; Jex & Spec-tor, 1996). Bono and Judge (2003) also exam-ined the relationship between the personalitytrait of core self-evaluation (discussed inChapter 4) and job satisfaction, and found acorrelation of .48 between the two variables.

Judge and his colleagues have conductedsome of the most comprehensive research onthe dispositional contributors to job satisfac-tion. Judge, Heller, and Mount (2002) con-ducted a meta-analysis on over 163 samplesand found that the big five personality traits ofextraversion, conscientiousness, openness toexperience, agreeableness, and neuroticism


collectively correlated .41 with job satisfac-tion. The highest correlations were betweenjob satisfaction and neuroticism (such thathigher neuroticism was related to low jobsatisfaction), extraversion, and conscien-tiousness. Furthermore, Ilies and Judge(2003) examined whether the Big Five per-sonality traits and Positive and NegativeAffectivity would account for the geneticinfluences on job satisfaction. According tothis approach, genetic factors produce differ-ences in personality and affectivity, whichthen leads individuals to report different lev-els of job satisfaction. The authors found thatPositive and Negative Affectivity accountedfor 45% of the genetic influence on ratings ofjob satisfaction, and that the Big Five person-ality traits accounted for 24%. These types ofanalyses do a good job of unifying researchon the dispositional approach to job satisfac-tion, showing how various factors within theindividual can influence ratings of job satis-faction in the absence of any additional in-formation about a person’s job!

One issue that dispositional researchersas yet have failed to resolve is determiningthe practical implications of dispositionaleffects. At first glance, it might be assumedthat if job satisfaction is linked to specifictraits, organizations would be justified inusing that information to select individualswho are likely to be satisfied. This recom-mendation, however, ignores the fact thatsituational effects still exert a stronger impacton job satisfaction than dispositions (e.g.,Gerhart, 1987; Levin & Stokes, 1989). Also,given the fact that, in many instances, jobsatisfaction is not strongly related to perfor-mance (Podsakoff & Williams, 1986), select-ing employees who are most likely to besatisfied may have adverse legal ramifica-tions. More research is needed beforedispositional findings are applied in organi-zational settings. Of course, it may be helpful

for managers to be aware of this research, sothey realize that some individuals are proneto report high or low job satisfaction irre-spective of job conditions.

In this section, we have covered threegeneral approaches to explaining employees’levels of job satisfaction in organizations: jobcharacteristics, social information processing,and dispositions. After examining each ofthese approaches, it is tempting to ask:‘‘Which of these approaches is correct?’’ Theweight of empirical evidence favors the jobcharacteristics approach, yet it would be pre-mature to conclude that this approach is‘‘right’’ and the other two approaches are‘‘wrong.’’ As was pointed out earlier, modelingsocial influence with a high degree of fidelityin laboratory settings is extremely difficult(Hulin, 1991). Furthermore, in the case ofdispositions, research is still in its infancy,and much still needs to be learned. Thus,the most appropriate conclusion, which issummarized in Figure 5.2, is that job satisfac-tion is a joint function of job characteristics,social information processing, and disposi-tional effects. This figure also illustrates thatby viewing these approaches as predictingdifferent aspects of job satisfaction, weimprove our ability to understand this impor-tant job attitude. Finally, we would like to

FIGURE 5.2Summary of the Determinants of Job Satisfaction

JobCharacteristics

Social InformationProcessing

JobSatisfaction

Dispositions


point out that organizational researchershave recently recognized the importance ofnational culture in the prediction of job sat-isfaction.

Job Satisfaction: A Cross-CulturalPerspective

Like many issues in organizational psychol-ogy, the study of job satisfaction has takenplace primarily in the United States andWestern European countries. This is obvi-ously a blind spot in our knowledge becausework is a universal activity, and, presumably,so is the development of positive or negativefeelings toward work. In this section, webriefly examine recent evidence on the pos-sibility of cross-cultural differences in levelsof job satisfaction, and potential reasons forthese differences.

Given the dearth of cross-cultural re-search in general, relatively few studies haveexamined cross-cultural differences in jobsatisfaction. For example, Griffeth andHom (1987) found that Latin Americanmanagers were more satisfied than Europeanmanagers. In a comparison of Dominicanand American employees working for thesame company, Marion-Landis (1993) foundthat the Dominicans were more satisfied thantheir American counterparts. Several studieshave also shown that Japanese employeestend to be less satisfied than American em-ployees (e.g., Smith & Misumi, 1989).

Direct cross-national comparisons in jobsatisfaction are interesting, but they often pro-vide little insight into why the differencesexist. To understand the basis of such differ-ences, it is useful to frame the issue in terms ofthe three approaches to job satisfaction dis-cussed at the beginning of this chapter. Whenviewed from the job characteristics perspec-tive, there are several plausible explanationsfor cross-cultural differences in job satisfac-

tion. For example, considerable evidenceexists for cross-cultural differences in values.Hofstede (1984) investigated differences invalues—including individualism/collectivism,masculinity, power distance, and uncertaintyavoidance—in 40 different countries. Theindividualism/collectivism dimension reflectsthe extent to which people are concernedwith their own interests and needs, rather thanthose of other people or of members of impor-tant collective units (e.g., family, work group).Masculinity reflects the degree to which thereis a focus on achievement and performance asopposed to the wellbeing and satisfaction ofothers. Power distance reflects the degree towhich those with high levels of authorityand status are distinct from those with lowerlevels. Finally, uncertainty avoidance reflectsthe extent to which people are comfortableworking in uncertain environments.

Hofstede’s (1984) findings have shownrather clearly the existence of cross-nationaldifferences on each of these four values. Forexample, the United States and countries inWestern Europe tend to place a very highvalue on individualism, while Hispanic andAsian countries tend to place a relatively highvalue on collectivism. With respect to mas-culinity, Scandinavian countries tend toplace a relatively high value on this dimen-sion compared to other countries. Powerdistance tends to have very high value inHispanic countries, but the opposite is truein countries such as Australia and Israel.Uncertainty avoidance was found to be high-est in countries such as Greece and Portugal,and lowest in Singapore and Denmark.

The primary implication of these cross-national differences in value preferences isthat cross-cultural differences in job satisfac-tion may be due to differences in what em-ployees desire from their jobs. Recall fromthe beginning of this chapter that job satis-faction has been purported to result from a


comparison between what people perceivetheir jobs provide and what they desire.Thus, when viewed from this perspective,cultural differences can be at least partiallyattributed to the fact that employees in differ-ent cultures seek different things from theirjobs, and may place different levels of impor-tance on different job facets.

While there is undoubtedly some meritto this argument, cross-cultural differencesin job satisfaction may also be influenced bycross-national differences in actual job con-ditions. Because of economic and politicaldifferences, employees in different countriesmay differ greatly in the quality of their on-the-job experiences. In the former SovietUnion, for example, it is unlikely that em-ployees in state-run organizations had muchdecision-making authority over many as-pects of their jobs. In contrast, employeesin countries that embrace free-market eco-nomics typically have greater participation indecision making and are more stronglyencouraged to engage in proactive behaviors.

Cross-cultural differences can also beviewed through the lens of the social infor-mation processing approach to job satisfac-tion. For example, it is possible that inaddition to value differences, cross-culturaldifferences may exist in the degree to whichsocial influence processes are salient to em-ployees. One might speculate that in an indi-vidualistic society such as the United States,social information may have a relatively min-imal impact, and job satisfaction may be onlyweakly related to prevailing cultural values.In contrast, in a more collectivist societysuch as Japan, social influence processesmay be much more important.

Compared to the job characteristics andsocial information processing approaches, thedispositional approach to job satisfactionwould appear to be less helpful in explainingcrosscultural differences in job satisfaction.

However, it is possible that the prevalence ofcertain dispositional traits that influencejob satisfaction may differ across cultures. Arecent study highlighting cultural influences onjob satisfaction can be found in Comment 5.3.

Outcomes of Job Satisfaction

Although job satisfaction is important for itsown sake, researchers and managers areinterested in job satisfaction primarilybecause of its possible relationship to othervariables of interest. Given the sheer volumeof research on job satisfaction that has beenconducted over the years, it would be nearlyimpossible to discuss all of the correlates ofjob satisfaction. Thus, this section describesrelations between job satisfaction and fourtypes of variables that have both theoreticaland practical importance: attitudinal varia-bles, absenteeism, employee turnover, andjob performance.

Attitudinal Variables: By far, job satis-faction has been found to correlate moststrongly with other attitudinal variables.These variables reflect some degree of lik-ing or disliking; that is, they are affective innature. Examples of common attitudinalvariables used in organizational researchinclude job involvement, organizationalcommitment (described later in the chap-ter), frustration, job tension, and feelings ofanxiety. Notice that all of these variables, toa large degree, reflect levels of affect. Forjob involvement and organizational com-mitment, this affect is positive. The othervariables reflect feelings of negative affect.

Considerable empirical research has sup-ported the relationship between job satisfac-tion and attitudinal variables. For example,in a comprehensive meta-analysis of 124published studies, Mathieu and Zajac (1990)found that the corrected correlation between


organizational commitment and job satisfac-tion was .53. It has also been found that jobsatisfaction is positively related to a multitudeof other measures that reflect positive affect,such as job involvement, positive mood, andorganization-based self-esteem, to name a few(e.g., Spector, 1997a). With respect to nega-tive attitudes, numerous occupational stressstudies have shown that job satisfaction is

strongly and negatively related to variablessuchas frustration, anxiety, and tension (Jack-son & Schuler, 1985; Jex & Spector, 1996;Spector & Jex, 1998).

Although there is little debate that jobsatisfaction is related to other attitudinalvariables, the precise mechanisms underlyingmany of these relations remain unclear be-cause much of the research on job satisfaction

COMMENT 5.3

NATIONAL CULTURE AND JOB SATISFACTION

HUANG AND VAN DE VLIERT (2004) recently con-

ducted a study of the determinants of job

satisfaction among individuals from 39 differ-

ent countries. Their total sample size was129,087, and the authors were able to deter-

mine how characteristics of the country, such

as the degree of collectivism versus individu-

alism, influenced the relationship between

other factors and job satisfaction. The extent

to which a country was collectivistic versus

individualistic was determined by three differ-

ent indexes, based on Diener, Diener, andDiener (1995Q7). The first was an index

reflecting individualism versus collectivism

provided by Hofstede (1991) in his study of

values in different countries. The second was

the ranking of a country’s individualism by

another leading researcher in the field, Harry

Triandis. Finally, each country’s divorce rate

was calculated between 1995 and 1998. Thereasoning behind this third indicator was that

countries with higher divorce rates are likely

more individualistic, as these nations recog-

nize the rights of individuals over collective

values of duty and obedience.

Huang and Van de Vliert argued that a

country’s level of individualism versus collec-

tivism may influence how factors such as joblevel and characteristics relate to job satisfac-

tion. For example, some researchers have

found that white-collar workers report higher

job satisfaction than blue-collar workers,

because the former employees have more

complex and interesting jobs. In the present

study the authors found that the tendency forwhite-collar workers to report higher job sat-

isfaction only occurred in more individualist

countries (e.g., the United States and the

United Kingdom), and not in more collectivist

countries (e.g., China and Indonesia). Fur-

thermore, job level and individualism/collec-

tivism interacted with job characteristics to

predict job satisfaction. The relationshipbetween job level and job satisfaction was

found primarily among jobs where indi-

viduals reported an opportunity to use their

skills and abilities at work, and this effect was

especially pronounced in individualistic cul-

tures. Having a white-collar job was actually

negatively related to job satisfaction when

individuals did not feel they had an opportu-nity to use their skills and abilities at work,

and this relaitonship was especially true in

collectivistic cultures.

The results of this study indicate the impor-

tance of examining how culture-level factors

can influence the prediction of job satisfaction.

In general, organizational researchers are

beginning to question how well results ob-tained with employees from a single culture

will generalize to employees in different cul-

tures.


has relied on self-report measures and cross-sectional designs. For example, a high level ofjob satisfaction may cause employees to haveother positive feelings toward their jobs, andmay lead to lower levels of negative feelings.Conversely, it is also possible that other pos-itive and negative attitudes cause high or lowlevels of job satisfaction. For example, a highlevel of job involvement, coupled with a lowlevel of frustration, may lead employees to feelsatisfied toward their jobs. It is also possiblethat such relations are the result of sharedcommon causes such as job conditions (Fried& Ferris, 1987; Jackson & Schuler, 1985;Mathieu & Zajac, 1990). Only the use of lon-gitudinal designs will allow researchers toinvestigate the causal direction between jobsatisfaction and other job attitudinal variables(Zapf, Dormann, & Frese, 1996).

Absenteeism: The study of absenteeism isimportant for both theoretical and practicalreasons. From a theoretical perspective,absenteeism represents a common wayin which employees may withdraw fromtheir jobs (Hulin, 1991). From a practicalperspective, absenteeism is a very costlyproblem to many organizations. When em-ployees are absent, work may not get doneor may be performed by less experiencedemployees.

It is certainly intuitively plausible that anemployee’s absence from work would be onereaction to a high level of job dissatisfaction.Despite its intuitive plausibility, empiricalresearch has provided only weak supportfor the relation between job satisfaction andabsenteeism. For example, Hackett andGuion (1985) conducted a meta-analysis of31 studies and found the corrected correla-tion between job satisfaction and absentee-ism to be only �.09. This suggests that jobsatisfaction may play some role in employeeabsences, but that role is marginal.

Hackett and Guion (1985) offer a numberof explanations for the weak relation betweenjob satisfaction and absenteeism. One reasonis the measurement of absenteeism itself.Although at first glance absenteeism wouldappear to be a rather simple variable, it isactually quite complex. For example, whenmeasuring absences, one can distinguishbetween excused and unexcused absences.Excused absences would be allowed forevents such as illnesses and funerals. In unex-cused absences, the employee simply doesnot show up at work. One could argue thatjob satisfaction would be more likely to play arole in unexcused than in excused absences.

Another reason for the weak relationbetween satisfaction and absenteeism is thatjob satisfaction represents a general attitude,whereas absenteeism is a specific form ofbehavior. For example, a person’s attitudetoward organized religion (a general atti-tude) would probably not be a good predic-tor of attendance at a worship service on oneparticular day. According to the Theory ofPlanned Behavior (AjzenQ6 1988, 2001), acomplex pathway links general attitudes(such as job satisfaction) to actual behavior.For example, variables such as subjectivenorms (the employee’s perception ofwhether important others expect him orher to go to work) and attitudes toward thebehavior of missing work on different occa-sions may be stronger predictors of absentee-ism than overall job satisfaction. Thus, jobsatisfaction may be weakly related to absen-teeism because of a failure to account forunmeasured variables such as normativestandards surrounding attendance, as wellas attitudes toward being absent from work.

Finally, an issue that absenteeism re-searchers typically confront is that absentee-ism is a behavior that has a low base rate (i.e.,it doesn’t occur frequently). Predicting a var-iable with a low base rate is problematic


because most of the statistical proceduresused by organizational psychologists, partic-ularly correlation and regression analysis, arebased on the assumption that variables arenormally distributed. In most instances inorganizational research, the variables exam-ined are not exactly normally distributed,but they do not deviate so far that conven-tional statistical procedures are seriouslybiased. However, in the case of absenteeism,distributions may be so skewed that the truerelationship between job satisfaction andabsenteeism is seriously underestimatedwhen commonly used statistical proceduresare used.

Employee Turnover: Another correlate ofjob satisfaction that is of considerable inter-est to both researchers and managers is em-ployee turnover. Some turnover inorganizations is inevitable and, in somecases, may even be desirable. However, veryhigh levels of turnover can be costly to orga-nizations since they must begin the processof recruiting, selecting, and socializing a newemployee. High levels of turnover may alsohave an adverse impact on the public imageof an organization, and hence increase thedifficulty of recruiting.

Given the importance of turnover, organi-zational psychologists have devoted consider-able attention to understanding itsantecedents. Although some of the work onturnover has been aimed at simply document-ing its relation with job satisfaction, muchmore has been aimed at modeling the rolejob satisfaction plays in employees’ turnoverdecisions. One of the earliest, and ultimatelymost influential, models of the turnover pro-cess was developed by Mobley (1977). As canbe seen in Figure 5.3, this model proposesthat employees’ decisions to leave a job arecomplex and consist of multiple stages. In thefirst stage, an employee evaluates his or her

FIGURE 5.3Mobley’s Model of the Turnover Process

Evaluation ofExisting Job

ExperiencedJob Satisfaction-Dissatisfaction

Intentionto Search

for Alternatives

Search forAlternatives

Evaluationof Alternatives

Comparison ofAlternatives versus

Present Job

Intent toQuit/Stay

Quit/Stay

Thinking ofQuitting

(a) Alternative forms of job withdrawal (e.g., absenteeism, passive job behavior).

(b) Nonjob related factors (e.g., transfer of spouse, may stimulate intention to search).

(e) Impulsive behavior.

(c) Unsolicited or highly visible alternatives may stimulate evaluation.

(d) One alternative may be withdrawal from labor market.

Evaluation of ExpectedUtility of Search and

Cost of Quitting

Source: W. H. Mobley. (1977). Intermediate linkages in the

relationship between job satisfaction and employee turnover.

Journal of Applied Psychology, 62, 237–240. Copyright # 1977

by the American Psychological Association. Reprinted with

permission.


existing job and, depending on this evalua-tion, experiences either satisfaction or dissat-isfaction. After this evaluation, if the employeeis satisfied, the process is unlikely to go fur-ther. If the employee is dissatisfied, however,this may lead to thoughts of quitting his or herjob. Notice, however, that the model allowsfor the possibility that employees may expressjob dissatisfaction through other forms ofwithdrawal, or by simply putting forth lesseffort.

Once a dissatisfied employee begins tothink about quitting his or her job, the nextstep in the model is some cognitive evaluationof whether a search will be successful, and thevarious costs associated with quitting thepresent job. Even if a person is extremelydissatisfied with a job, leaving entails certaincosts—moving to a new location and perhapsgiving up benefits accrued in the present job.If an employee decides either that a searchwould be unsuccessful, or that the cost ofleaving the job is too high, the process willend and the employee may simply find waysto adapt to the present situation.

On the other hand, if the employee be-lieves that a search will be successful, and thecosts associated with leaving are not prohib-itive, he or she will then progress to the nextstage in the model: an intention to search foralternatives. This is the point at which a per-son begins planning the job search, and, in alllikelihood, an intention to search will trans-late into actual search behavior. The employeemay post his or her resume on internet jobsites such as ‘‘Monster.com,’’ seek the servicesof an employment agency, or attend job fairs,hoping tofind alternative employment oppor-tunities. The model allows for the possibilitythat the intention to search will be motivatedby factors other than job dissatisfaction (e.g.,desire to live in another location).

After searching for alternatives, a personmay find that none is available. This outcome

obviously depends on a person’s level ofqualifications and the availability of jobs inhis or her profession. If a person finds noalternatives, he or she may have no choicebut to adapt to the present job. If alternativesare available, the next step is to evaluatethem. It is also possible that an individualmay be presented with employment alterna-tives suggested in unsolicited offers.

In evaluating different employment alter-natives, the model proposes two standards ofevaluation. Alternatives are evaluated againstthe employee’s internal standards for judgingthe acceptability of jobs, as well as his or herpresent job. Given these two evaluative stand-ards, it is possible that alternatives may exceeda person’s internal standards, yet still notmeasure up to his or her present job. If thisis the case, the job search may remind theemployee that the ‘‘Grass is not necessarilygreener on the other side,’’ and the presentjob may be viewed in a more favorable light.Another possibility, which is acknowledgedin the model, is that the individual may with-draw from the labor market completely. Forexample, a dissatisfied employee may decideto become a stay-at-home parent.

The job search may also result in one ormore alternative employment offers that areperceived to be more attractive than thepresent job. According to the model, if thisis the case, the person forms intentionsregarding whether to quit the present job.Why don’t people automatically quit theirpresent jobs if a better alternative is found?Based on the previously discussed Theory ofPlanned Behavior, a person may decline amore attractive job offer simply because heor she does not have a positive attitudetoward the act of changing jobs. Normativeinfluences may come into play as well. Forexample, a person may come from a family inwhich both parents worked for the sameorganization during their entire careers,


and thus may experience subtle (or not sosubtle) normative pressure to remain em-ployed with the same organization and notbe a job hopper.

Another reason that intentions may nottranslate into actual behavior is that the act ofquitting one’s job is much different from theidea of quitting. In Figure 5.3, Mobley’s(1977) model proposes that, relatively earlyin the process, an employee should evaluatethe costs associated with quitting. It isimportant to note, however, that early inthe process, quitting is an abstract conceptand not a concrete choice that a person isfaced with. Thus, although the model is notvery explicit, some reevaluation of the costsof quitting one’s job is likely to take placebetween the intention to quit and the actualquitting. A person may get cold feet whenfaced with a concrete job offer, and maydecide that the costs associated with leavingthe present job for a better one are not worthit after all. As a final note, the model allowsfor the possibility that the decision to quitmay be made impulsively. Perhaps somereaders have had the experience of makingan on-the-spot decision to quit a job.

Empirical research over the years hassupported Mobley’s model in two ways.First, studies that have tested the originalmodel, or variants of it, have generally pro-vided support (e.g., Hom, Caranikas Walker,Prussia, & Griffeth, 1992; Michaels &Spector, 1982; Mobley, Griffeth, Hand, &Meglino, 1979). In a recent examination ofa similar model, KammeyerMueller, Wan-berg, Glomb, and Ahlburg (2005) conducteda study where they examined employees overfive distinct points in time over a 2-yearperiod to assess those variables that pre-dicted who left versus remained with theorganization. Using Hierarchical LinearModeling, the authors found that employeeswho left the organization showed reductions

in work satisfaction and commitment to theorganization across the five data-collectionperiods. In addition, these employeesshowed increases in work withdrawal behav-iors and searching for alternative jobs acrossthe five time periods.

The model has also been supportedmore indirectly through studies examiningthe correlation between job satisfactionand turnover. Carsten and Spector (1987)conducted a meta-analysis of 42 studies andfound that the corrected correlation betweenjob satisfaction and turnover was �.24. Thecorrected correlation between behavioralintentions and actual turnover was .32. Tettand Meyer (1993) also conducted a meta-analysis and estimated the correlation to be.27. One would expect intentions to be morestrongly correlated with turnover than withjob satisfaction because intent is a more prox-imal cause of job satisfaction (see Figure 5.3).

These authors also examined, albeit indi-rectly, whether the availability of employ-ment alternatives would influence therelationship between job satisfaction andturnover. Specifically, the authors obtaineddata on the levels of unemployment thatexisted in the localities at the time whendata for each of the studies were collected.As expected, the corrected satisfaction–turnover and intentions–turnover correla-tions were both stronger during periods oflow (as opposed to high) unemployment.This is presumably due to the fact that alter-native employment opportunities are muchmore plentiful when unemployment is low.

These findings are consistent with therole that job satisfaction is proposed to playin the turnover process. In fact, when oneconsiders that job satisfaction is actually avery distal cause of turnover, and turnover isa low base-rate event, an overall correctedcorrelation of�.24 to .27 between these twovariables is actually quite remarkable. At a


more conceptual level, these findings suggestthat the desire to find more satisfying work isoften a driving force behind job changes.Therefore, organizations wishing to keepturnover to manageable levels cannot ignorejob and organizational conditions thatimpact job satisfaction.

Job Performance: The third correlate ofjob satisfaction is job performance. Of thenumerous variables that researchers have cor-related with job satisfaction over the years,job performance has perhaps the longesthistory. In fact, the attempt to link job satis-faction with job performance can actuallybe traced back as far as the HawthorneStudies. Based on their findings, the Haw-thorne researchers came to the relativelynaıve conclusion that one way to make em-ployees more productive was to make themmore satisfied. Stated differently, ‘‘A happyworker is a productive worker.’’ This notionthat job satisfaction influenced job perfor-mance became widely accepted and helpedto usher in what was described in Chapter 1as the Human Relations movement withinorganizational psychology.

Toward the end of the 1950s and in theearly 1960s, another trend in organizationalpsychology—reliance on cognitive process-ing models—would eventually change theprevailing views on the relationship betweenjob satisfaction and job performance.Vroom’s (1964) Expectancy Theory, forexample, proposed that employees wouldput forth more effort if they believed thateffort would translate into high levels ofperformance, and higher performancewould lead to valued outcomes. If perfor-mance is viewed from this perspective, thereis no reason to assume that job satisfactionshould play a causal role in determining jobperformance. On the other hand, if highlevels of job performance ultimately lead to

desirable outcomes, employees should bemost satisfied with their jobs when theyperform well and are rewarded for it. Ifviewed from this perspective, one wouldconclude that job performance causes jobsatisfaction. Thus, rather than trying to makeemployees happy, organizations would bemuch better off helping employees developthe skills they need to perform well, andlinking rewards to performance.

Unfortunately, much of the early debatesurrounding the relation between job satisfac-tion and job performance was based on opin-ion instead of empirical data. That basis beganto change in the 1970s and 1980s, when therewere more empirical investigations of therelation between job satisfaction and job per-formance. In the mid1980s, many of theseempirical studies were summarized in com-prehensive meta-analyses by Iaffaldano andMuchinsky (1985) and, later, by Podsakoffand Williams (1986). In the Iaffaldano andMuchinsky investigation, the corrected corre-lation between job satisfaction and job per-formance was found to be .17. Podsakoff andWilliams obtained very similar results.

Podsakoff and Williams (1986) alsofound that the satisfaction–performancerelation was moderated by the degree towhich rewards were linked to performance.In studies where rewards were closely tiedto performance, the corrected correlationbetween job satisfaction and performancewas .27. In contrast, in studies whererewards were not closely tied to perfor-mance, the corrected correlation was weaker(r ¼ .17). This moderator effect is importantbecause it suggests that when job satisfactionand performance are related, the most plau-sible causal sequence is from performance tojob satisfaction, rather than the reverse. Morespecifically, if rewards are tied closely toperformance, job satisfaction may be a nat-ural byproduct of receiving rewards.


Just when organizational researchers werebeginning to doubt whether job satisfactionwas a sufficiently strong predictor of job per-formance, Judge, Thoresen, Bono, and Patton(2001) reviewed past meta-analyses andargued there were limitations with the stud-ies. The authors conducted a new meta-analysis with over 54,000 participants from312 samples and found the true correlationbetween job satisfaction and performance was.30. The authors also pointed out that thecorrelations between job satisfaction andperformance varied widely between studies,suggesting there are other variables that deter-mine the strength of the job satisfaction–performance relationship. In their meta-analysis, Judge et al. found that the relation-ship was especially high when employees hadhighly complex jobs. Judge et al. proposedan integrative model of the satisfaction–performance relationship that emphasizedreciprocal relationships between the two var-iables as well as mediators and moderatorsof the satisfaction to performance and per-formance to satisfaction relationships. Theauthors argued that future researchers shouldfocus their attention on examining how jobsatisfaction is related to job performance andfactors that enhance versus reduce thestrength of the relationship.

Schleicher, Watt, and Greguras (2004)followed these suggestions and uncoveredan important moderator of the satisfaction–performance relationship: the affective-cognitive consistency (ACC) of job satisfaction.The authors noted that prior research in thefield of attitudes has shown that people whoreport discrepancies between the affective andcognitive components of an attitude objectevidence lower correlations between that atti-tude and behavior (Kraus, 1995). This pointemphasizes that individuals reporting the samelevel of job satisfaction may hold that attitudewith different levels of strength, and that we

should not expect job satisfaction to predictperformance if individuals do not exhibit con-sistency between the different components ofjob satisfaction.

Remember at the beginning of the chap-ter we emphasized that job satisfactionentails both positive affect or feelings aboutone’s job as well as cognitive beliefs aboutone’s job. Schleicher et al. (2004) argued thatjob satisfaction would not predict job perfor-mance well when there was inconsistencybetween these two components (e.g., whenemployees reported positive affect for theirjob but exhibited negative thoughts). In fact,the authors found a reasonably strong rela-tionship between job satisfaction and perfor-mance when ACC was high, but virtually norelationship when ACC was low. These find-ings highlight the importance of consideringthe nature (strength) of job attitudes whenexamining the ability of job attitudes to pre-dict outcomes like performance, absentee-ism, and turnover.

Based on the accumulated empiricalresearch, it is tempting to conclude that thejob satisfaction–job performance relation-ship is unpredictable and variable. However,according to Ostroff (1992), this conclusionmay be erroneous because the vast majorityof studies examining the relationshipbetween job satisfaction and job perfor-mance have been conducted at the individ-ual level of analysis. Ostroff points out thatalthough employees who are highly satisfiedwith their jobs may not necessarily performbetter than employees who are more dissat-isfied, this relation may be stronger at theorganizational level of analysis. Organiza-tions containing employees who are highlysatisfied will tend to perform better thanorganizations containing employees whoare highly dissatisfied. When employeesare highly satisfied, they may not be moreproductive as individuals, but they may


nevertheless engage in behaviors that facili-tate the effectiveness of the organization as awhole.

Ostroff (1992) tested this hypothesis byexamining relations between job satisfactionand several performance indexes in a nationalsample of 298 junior and senior high schools.As can be seen in Table 5.4, aggregate-leveljob satisfaction was significantly relatedto every performance indicator, and the mag-nitude ranged from �.11 to .44. Many ofthese correlations are considerably higherthan those found in individual-level studies(Iaffaldano & Muchinsky, 1985; Podsakoff &Williams, 1986). Interestingly, Ostroff arguedthat, at the organizational level, satisfactionlikely causes higher levels of performance,which is counter to individual-level studies(Podsakoff & Williams, 1986). Unfortu-nately, since this study was cross-sectional,the issue could not be addressed.

However, a recent study by Schneider,Hanges, Smith, and Salvaggio (2003) chal-lenges the argument that job satisfactiondetermines performance at the organiza-tional level. These authors studied 35 com-panies over 8 years and assessed jobsatisfaction and objective indexes of com-pany performance (i.e., financial and marketshare indexes). First of all, the authors foundthere were significant influences of the com-pany on individual reports of job satisfac-tion. Secondly, the pattern of relationshipsbetween job satisfaction and company per-formance over time suggested that high lev-els of company performance were morelikely to precede increases in job satisfactionthan job satisfaction preceding companyperformance. These results suggest thathighly performing companies end up withemployees who are more satisfied than em-ployees who perform poorly. Clearly, orga-nizational psychologists will need to conductadditional research to better understand thecausal direction between job satisfaction andjob performance at the individual and grouplevels, as well as variables that influence bothrelationships. A recent approach to job atti-tudes as a predictor of important behaviorsin organizations combines job satisfactionand organizational commitment into a singlemeasure of job attitude (Harrison, Newman,& Roth, 2006). This approach is consideredin the following section on organizationalcommitment.

ORGANIZATIONALCOMMITMENT

In addition to feelings of satisfaction or dis-satisfaction for one’s job, employees maydevelop feelings of attachment or commit-ment toward the organizations in which theyare employed. As with satisfaction or dissat-isfaction, a strong case can be made that the

TABLE 5.4Correlations between Organization-Level JobSatisfaction and Organization-Level PerformanceMeasures

Performance Measure Correlation

Reading achievement .30

Math achievement .31

Social science achievement .24

Percentage of students passing courses .20

Administrative performance .24

Percentage of students dropping out �.28

Percentage of students attending .24

Percentage of students with

discipline problems

�.27

Vandalism costs (in dollars) �.11

Student satisfaction with teachers .24

Overall student satisfaction .44

Note: All correlations are statistically significant beyond the

.05 level.

Source: C. Ostroff. (1992). The relationship between satisfac-

tion, attitudes, and performance: An organization-level ana-

lysis. Journal of Applied Psychology, 77, 963–974. Copyright #

1992 by the American Psychological Association. Adapted

with permission.


tendency to develop attachment or commit-ment ties extends far beyond the workplace.For example, people commit to each otherthrough marriage and other forms of kin-ship. Many people also faithfully committhemselves to activities such as exercising,institutions such as churches, and politicalideologies such as democracy. Given thesevast numbers of commitments, it is not sur-prising that employees also develop feelingsof commitment and attachment toward theorganizations in which they work.

Defining OrganizationalCommitment

At a very general level, organizational commit-ment can be thought of as the extent to whichemployees are dedicated to their employingorganizations and are willing to work ontheir behalf, and the likelihood that they willmaintain membership. Note that in this gen-eral definition, one can distinguish betweenaffective commitment and behavioral com-mitment (Mowday, Porter, & Steers, 1982).Commitment represents both the feelingsand the behavioral tendencies that employ-ees have toward the organization.

Meyer and Allen (1991) further refinedthe definition of organizational commitmentby pointing out there can be multiple basesof commitment—that is, employees may becommitted for different reasons, and thesereasons constitute unique forms of commit-ment. They proposed a three-componentmodel of commitment consisting of affective,continuance, and normative commitment.Affective commitment reflects the extent towhich employees identify with the organiza-tion and feel a genuine sense of loyaltytoward it. In contrast, continuance commit-ment is based on employees’ perceptions ofthe relative investments they have made inthe organization, and the relative costs asso-

ciated with seeking membership in anotherorganization. Normative commitment isbased on an employee’s feeling of obligationto the organization, wherein remaining amember is the morally right thing to do.

In addition to having multiple bases, em-ployee commitment may be focused at differ-ent levels within the organization and mayeven be directed to outside groups. Forexample, an employee may feel a sense ofcommitment toward his or her organizationas a whole, the primary work group to whichhe or she belongs, and perhaps the leader ofthe group. Many employees in organizationsalso feel a sense of commitment toward theprofession to which they belong. For exam-ple, physicians who work for Health Main-tenance Organizations (HMOs) are likely tohave some level of commitment to theiremploying organization, but are committedto the medical profession as well.

Meyer and Allen (1997) illustrate themultiple forms of commitment that canoccur in a matrix in which the three basesof commitment (affective, continuance, andnormative) are crossed with six distinct foci.As can be seen in Figure 5.4, an employeemay have feelings of affective, continuance,and/or normative commitment toward anynumber of foci within the organizationalenvironment. This reflects the fact that, foremployees in most organizations, commit-ment is a multidimensional, complex con-struct. Thus, if one were to come up to anemployee and ask, ‘‘How committed areyou?’’ the employee would most likely havea multipart answer.

Measurement of OrganizationalCommitment

As with most subjective attitudinal variables,organizational commitment is measuredwith self-report scales. Historically, the first


organizational commitment scale to gainwidespread use was the Organizational Com-mitment Questionnaire (OCQ; Mowday,Steers, & Porter, 1979). The original OCQprimarily reflected what Meyer and Allen(1991) described as affective commitmentand, to a lesser extent, normative commit-ment. The original OCQ also contained oneitem measuring an employee’s turnover inten-tions. The inclusion of this item promptedcriticism, particularly when the OCQ wasused to predict turnover. Most researcherswho have used the OCQ in recent years haveeliminated the turnover intent item. In manycases, researchers have also used shorter ver-sions of the original measure.

In general, there is evidence that theOCQ has desirable psychometric properties.Mathieu and Zajac (1990), in their meta-analysis of 124 organizational commitmentstudies, reported that the mean internal con-sistency reliabilities for various forms of theOCQ were all over .80. In this same study,the OCQ was found to correlate appropri-ately with conceptually related variables,thus providing some support for its constructvalidity. The major limitation of the OCQ isthat it measures primarily the affective com-ponent of organizational commitment, and

thus provides very little information on thecontinuance and normative components.This is an important limitation because thesedifferent forms of commitment are associatedwith different outcomes.

More recently, Allen and Meyer (1990)developed an organizational commitmentmeasure that contains three subscales thatcorrespond to the affective, continuance,and normative components of commitment.An example of an affective commitment itemis: ‘‘This organization has a great deal ofpersonal meaning to me.’’ An example of acontinuance commitment item is: ‘‘It wouldbe too costly for me to leave my organizationin the near future.’’ Finally, an example of anormative commitment item is: ‘‘I would feelguilty if I left my organization now.’’

Because the Allen and Meyer (1990) scalehas been developed more recently than theOCQ, comparatively less evidence has accu-mulated to support both its reliability and itsvalidity. However, the evidence accumulatedto date has been very encouraging. For exam-ple, Meyer and Allen (1997) reported thatthe median internal consistency reliabilitiesfor the affective, continuance, and normativecommitment scales are .85, .79, and .73,respectively. They also report that all three

FIGURE 5.4The Relationship between Bases and Foci of Commitment

Focu

s of

Com

mit

men

t

Affective

Organization

Top management

Unit

Unit manager

Team leader

Work team

Continuance

Bases of Commitment

Normative

Source: J. P. Meyer and N. J. Allen. (1997). Commitment in the workplace: Theory, research, and application. Thousand Oaks, CA:

Sage. Reprinted by permission of Sage Publications, Inc.


scales have exhibited reasonably high levelsof temporal stability.

In terms of construct validity, there is alsoimpressive supporting evidence. For exam-ple, several studies have supported the three-factor structure of the scale (summarized inMeyer & Allen, 1997). There is also evidencethat these forms of commitment are empiri-cally distinguishable from related constructssuch as job satisfaction, values, and occupa-tional commitment. The construct validity ofMeyer and Allen’s measure has also beensupported by the pattern of its relationshipswith other variables. (These will be describedin more detail in a later section.) The impor-tant point is that the three subscales corre-sponding to the three different forms ofcommitment appear to correlate with othervariables in an expected manner.

Other than the OCQ and the Allen andMeyer (1990) scales, a handful of other meas-ures of organizational commitment have sur-faced, but none has been used extensively.One recent measure worth noting was devel-oped by Becker (1992). In this study, organi-zational commitment was measured in termsof multiple bases (as per Meyer and Allen) andmultiple foci. Few other studies have donethis, so there is little empirical evidence on theviability of this approach to measuring com-mitment. However, in the future, it may beuseful to measure commitment in this fashionif, indeed, different outcomes are associatedwith different combinations of bases and fociof commitment.

Development and Predictors ofOrganizational Commitment

What determines employees’ levels of com-mitment toward their organizations? Giventhe complexity of the organizational commit-ment construct, this is not an easy questionto answer. Most researchers have ap-

proached this issue by examining the devel-opment of each of the three bases ofcommitment proposed by Meyer and Allen(1991). If one considers affective commit-ment, a prediction might be that employeeswill tend to develop this type of commitmentif they perceive that the organization is beingsupportive and/or treating them in a fairmanner (Meyer & Allen, 1991). In fact,research has shown that affective commit-ment is positively related to variables suchas perceived organizational support (POS)and procedural justice. POS simply repre-sents the extent to which the organizationis seen as helpful to the employee; in effect, itis ‘‘on the employee’s side.’’ Recall from theprevious chapter that procedural justicereflects the fairness of the procedures orga-nizations use in dealing with employees.Recent research with hospital nurses hasrevealed that a supportive organizational cul-ture is predictive of affective commitment,especially the culture of the specific ward inwhich the nurses work (Lok & Crawford,2001).

Another factor that may influence thedevelopment of affective commitment iswhether the organization is seen as a sourceof rewarding outcomes. Research has shown,for example, that a positive relationshipexists between affective commitment andvariables such as job scope, participativedecision making, job autonomy, and per-ceived competence (Meyer & Allen, 1997).One way to interpret such findings is basedon a belief that employees develop feelings ofaffective commitment if they see the organi-zation as a place where they feel they areimportant and competent.

Another way some researchers havesought to explain affective commitment isthrough behavioral commitment and retro-spective sense making. Put differently, onemight say that employees develop feelings of


affective commitment as a retrospectivemechanism to justify their tenure in the orga-nization and the level of effort they haveexpended on its behalf. This explanation ofcommitment is consistent with Salancik andPfeffer’s (1978) Social Information Processingtheory, described earlier in the chapter. Ingeneral, retrospective explanations of affec-tive commitment have met with very limitedsupport. However, as Meyer and Allen (1997)point out, this mechanism is quite subtle andthus may be very difficult to test empirically. Astudy examining this type of explanationwould have to be longitudinal, and employeeswould need to be assessed on multiple occa-sions (see Zapf et al., 1996).

Compared to affective commitment,explaining the development of continuancecommitment is much more straightforward.Most explanations of continuance commit-ment rely on Becker’s (1960) notion of sidebets as a mechanism committing one to acourse of action. If, for example, a personhas wagered a bet that he or she would lose20 pounds over the next 6 months, thiswould commit the person to that course ofaction. When this concept is applied to theworkplace, we can see very clearly that, overtime, employees accumulate a number of sidebets that commit them to their currentemployer. For example, the accrual of se-niority means that employees may be entitledto special benefits or privileges. If the em-ployee were to leave and work for anotheremployer, such benefits would be forfeited.Also, many employees develop numeroussocial relationships with their coworkers,and these bonds help to facilitate feelings ofbelonging and comfort. These feelings wouldbe forfeited in a switch to another employer.

Another proposed determinant of con-tinuance commitment is the extent to whichemployees perceive other viable alternativesto the present employer. The word perceive is

italicized because it really doesn’t matterwhether actual alternatives exist; the impor-tant thing is an employee’s perceptions. Per-ceptions of alternatives may be influenced byfactors in the environment, such as theunemployment rate, but may also be affectedby other more subjective factors. For exam-ple, an employee’s perception of his or heroverall competence, level of training, andmobility will all enter into the perception ofalternatives. As one might guess, continu-ance commitment will tend to be higheramong employees who perceive few alterna-tives to the present employer.

Compared to affective and continuanceforms of commitment, much less is knownabout the development of normative com-mitment. According to Meyer and Allen(1997), personal characteristics and thenature of an employee’s transactions withthe organization may influence the develop-ment of normative commitment. At a per-sonal level, individuals may differ in terms ofwhether their early socialization emphasizedthe development of strong loyalty and asense of moral obligation to their employers.Meyer and Allen also point out that theorganization may attempt to instill in em-ployees, during the initial socialization pro-cess, a strong sense of moral obligation to theorganization.

Perhaps the most powerful determinantof normative commitment is the manner inwhich an organization treats its employees.When employees enter an organization, animplicit agreement, or a psychological con-tract, exists between them and the organiza-tion (e.g., Schein, 1980). A psychologicalcontract essentially represents an employee’sperceptions of what he or she feels is reason-able treatment as a member of the organiza-tion. One would assume that normativecommitment is highest when an employeeperceives the organization as honoring its


end of the psychological contract. Moreresearch, however, is needed before moreconclusions can be drawn about the devel-opment of this form of commitment.

More recent research has emphasizedcross-cultural and ethnic influences on orga-nizational commitment. This type ofresearch has involved examining the levelof organizational commitment across coun-tries and assessing whether models of orga-nizational commitment developed withlargely United States and western Europeansamples generalize to other populations.Gelade, Dobson, and Gilbert (2006) exam-ined national differences in affective organi-zational commitment across 49 differentcountries. The authors also examinedcountry-level predictors of organizationalcommitment, such as personality attributes,socioeconomic indicators, and values. Theseauthors found that organizational commit-ment did vary across the 49 countries, withapproximately 11% of the variance in scoresbeing a function of nation.

The authors found the highest levels oforganizational commitment were shown incountries that were extraverted, low in neu-roticism, high in happiness, high in egalitar-ian values, and low in societal cynicism. Theauthors found no relationship between affec-tive commitment and gross national incomeper capita, indicating that making moremoney does not, by itself, account for orga-nizational commitment across nations.

An example of Meyer and Allen’s com-mitment model being applied to othercountries can be found in a study con-ducted by Lee, Allen, Meyer, and Rhee(2001). These authors examined whetherthe three-component model discussed pre-viously applied to employees in SouthKorea. In the first study the authors foundthat the original Meyer and Allen measureassessing the three forms of commitment

did not generalize well to South Korea, withthe Continuance Commitment scale show-ing especially low reliability. The authorsthen used items developed by Meyer andhis colleagues that avoided North Americanexpressions and were therefore more ame-nable to translations into other languages.Lee and his colleagues found that with thisset of new items the underlying structure(and therefore meaning) of organizationalcommitment was similar across the twocultures. These results illustrate the impor-tance of ensuring proper translation of orga-nizational constructs when examiningcross-cultural similarities and differences.

Consequences of OrganizationalCommitment

As with job satisfaction, researchers andmanagers are interested in organizationalcommitment largely because of its relation-ship with other variables. In this section, webriefly review evidence on the relationshipbetween organizational commitment andattitudinal variables, absenteeism, turnover,and performance.

Attitudinal Variables: Given Meyer andAllen’s (1991) distinction among affective,continuance, and normative commitment,the correlates of each of these forms of com-mitment are examined separately. Affectivecommitment has been shown to be stronglyrelated to other work-related attitudes. Asmentioned earlier in the chapter, Mathieuand Zajac (1990) found that the mean cor-rected correlation between affective organi-zational commitment and job satisfactionwas .53. Other consistent attitudinal corre-lates of affective commitment found in thismeta-analysis included job involvement(.36), occupational commitment (.27), unioncommitment (.24), and stress (.29).


Compared to affective commitment, lessempirical work has examined the relationbetween attitudinal correlates of either con-tinuance or normative commitment (Meyer& Allen, 1997). Based on the little evidencethat is available, however, it appears thatcontinuance commitment is correlated withmany of the same variables as affective com-mitment, yet there are some important differ-ences. Mathieu and Zajac (1990), forexample, found that affective commitmentwas more strongly related to job satisfactionand job involvement than was continuancecommitment. A more recent meta-analysisby Cooper-Hakim and Viswesvaran (2005)concluded that job satisfaction correlated .47with affective commitment, .09 with contin-uance commitment, and .29 with normativecommitment.

Absenteeism: Compared to attitudinalcorrelates, much less evidence exists on therelation between each form of organizationcommitment and absenteeism. Mathieu andZajac (1990) found that the corrected cor-relation between affective commitment andattendance was .12 and the correlation withlateness was .11. These findings suggestthat those with high levels of affective com-mitment tend to exhibit lower levels ofabsenteeism, but this trend is quite weak.Recall from the previous section that thecorrelation between absenteeism and jobsatisfaction is of a similar magnitude (e.g.,Hackett & Guion, 1985). As with job satis-faction, this weak relationship may be dueto variation in the measurement of absen-teeism, as well as more general issues inattitude–behavior consistency. Also, from aconceptual point of view, a high level ofaffective commitment indicates a desire tocontribute to an organization—a desire thatmay at times be negated by situational con-tingencies.

Again, compared to affective commit-ment, little evidence exists on the relationsbetween either continuance or normativecommitment and absenteeism. Studies thathave been done, however, have shown nei-ther of these forms of commitment to berelated to absenteeism (Meyer & Allen,1997). From a conceptual point of view,these findings are somewhat surprising. Forexample, if an employee’s commitment is ofthe continuance variety, it is in his or her bestinterest to attend work on a regular basis;failure to do so could jeopardize his or hermembership in the organization. This argu-ment of course is based on the assumptionthat organizational policy is such that fre-quent absenteeism would be met with nega-tive consequences. With respect tonormative commitment, frequent absentee-ism would seem to be inconsistent withcommitment based on a strong moral obli-gation toward one’s employing organization.Given the little research that is available,both of these possibilities await examinationin future research.

Employee Turnover: Given the nature oforganizational commitment, considerablymore evidence exists on the relation amongall three forms of commitment and turn-over. As might be expected, research hasgenerally shown a negative relation amongall three forms of commitment and turn-over (Allen & Meyer, 1996; CooperHakim& Viswesvaran, 2005; Mathieu & Zajac,1990). The fact that all forms of commit-ment are negatively associated with turn-over would appear to be a positive thingfor organizations. However, this may not betrue in some cases. For example, consideran employee who remains in an organiza-tion primarily because he or she has a highlevel of continuance commitment. Is thisnecessarily good for the organization, or


even for the employee? Such an individualmay adopt an attitude of doing the bareminimum and may be very unhappy in hisor her job. The same may be true for anemployee who remains in an organizationprimarily out of a sense of moral obligation(e.g., normative commitment).

A recent study on organizational com-mitment and turnover adopted a longitudi-nal approach, assessing employees at fivepoints over a 2-year period (Kammeyer-Mueller et al., 2005). At each point theauthors assessed perceived costs of leavingthe company, organizational commitment,critical events experienced by the employee(e.g., mistreatment), and other factors. Usingsurvival analysis, the authors found thatorganizational commitment (assessed byMowday & Steer’s measure) was a significantpredictor of turnover over time.

Job Performance: Much research over theyears has investigated the relation betweenorganizational commitment and job perfor-mance. In general, affective commitment hasbeen shown to be positively related to jobperformance, although the magnitude of thisrelation is not strong (CooperHakim & Vis-wesvaran, 2005; Mathieu & Zajac, 1990;Meyer & Allen, 1997). Determining themechanism(s) behind these relations is diffi-cult, however, because these studies haveused a wide variety of performance criterionmeasures. For example, some have usedsupervisors’ ratings of overall performance(e.g., Konovsky & Cropanzano, 1991),others have used objective indexes such ascost control (e.g., Shim & Steers, 1994), andothers have utilized self-ratings of perfor-mance (e.g., Baugh & Roberts, 1994). Onecommonality among these studies, however,is that the relation between affective com-mitment and performance is mediated byemployees’ effort. Employees who possess

high levels of affective commitment tend towork harder and exert more effort than em-ployees who possess lower levels of affectivecommitment. In some cases, this higher levelof effort will translate into higher levels ofperformance, although this is not always thecase (Campbell, 1990, 1994).

This link between affective commitmentand effort suggests that commitment is pos-itively related to performance when employ-ees possess adequate ability, whenperformance is primarily determined by moti-vation, and when employees have some levelof control over performance. This explainswhy researchers have generally found thataffective commitment predicts organizationalcitizenship behavior (OCB) better than inroleperformance (Meyer & Allen, 1997; Organ &Ryan, 1995). Recall from the previous chapterthat because OCB is largely motivationallybased, employees have greater control overit than they do over inrole performance.

Compared to affective organizationalcommitment, considerably less research hasexamined the performance-related implica-tions of either continuance or normative com-mitment. Meyer and Allen (1997) point out,however, that most of the available empiricalresearch has shown that neither of these formsof commitment is strongly related to either in-role performance or OCB (see also Cooper-Hakim & Viswesvaran, 2005). Furthermore,it is difficult to come up with a conceptualjustification for why they would be related toperformance. For example, there is no reasonwhy continuance commitment would promptan employee to exert high levels of effort or goappreciably beyond his or her required jobduties.

It is somewhat more plausible that highlevels of normative commitment wouldengender high levels of effort toward organi-zational goals. One can also make an equallyplausible counter argument that commitment


based on employees’ feelings of obligationwould not necessarily lead to greater levelsof effort on behalf of the organization. To thecontrary, one can even imagine that an em-ployee who feels compelled to remain in anorganization out of a sense of obligation mayeven grow to resent that organization andperhaps be compelled to engage in counter-productive behaviors (see Chapter 6).

Sinclair, Tucker, Cullen, and Wright(2005) have recently examined the interest-ing idea that profiles on affective and con-tinuance commitment may be associatedwith in-role and extra-role performance.These authors pointed out that no priorresearch has examined how combinationsof affective and continuance commitmentmight predict outcomes. The authors usedcluster analysis to identify four different em-ployee profiles relevant to commitment:allied (medium affective and continuancecommitment), free agents (medium contin-uance and low affective commitment),devoted (high affective and continuancecommitment), and complacent (mediumaffective and low continuance). The authorsfound relatively strong correlations betweenaffective and continuance commitment andin-role and extra-role performance. In addi-tion, the authors found that the devoted em-ployees were rated especially high by theirsupervisors on measures of in-role and extra-role performance compared to the othergroups.

Combining OrganizationalCommitment and Job Satisfactionto Predict Performance

So far in this chapter we have treated orga-nizational commitment and job satisfactionas two separate constructs, in keeping withthe vast majority of thinking and research inorganizational psychology. However, a recent

meta-analysis by Harrison et al. (2006)examined the effects of combining orga-nizational commitment and job satisfactioninto an overall job attitude. These authorsthen assessed the ability of this overall jobattitude to predict a combined measure ofwork-relevant behavior assessing such be-haviors as task performance, extra-role per-formance, and turnover. The logic behindthis approach is similar to that describedearlier in the chapter by Ajzen (1988), whoargued that general measures of attitudes willbe most related to general measures ofbehavior. Harrison et al. found that a meas-ure of overall job attitude correlated withoverall work-relevant behavior at .59, pro-viding strong evidence that job attitudes arelinked to broad measures of behaviors thatemployees engage in at work. There is nodoubt that this recent finding will rein-vigorate research on the importance of jobattitudes.

Practical Applications ofCommitment Research

One way to view the applications of organi-zational commitment research is to examinevarious ways in which organizations mayengender high levels of commitment amongtheir employees. Meyer and Allen (1997)describe several different human resourcespolicies that may influence employee com-mitment. For example, it has long been rec-ommended that during the selection andrecruitment processes, organizations pro-vide realistic information to potential em-ployees (Wanous, 1973). Retention hastypically been cited as the rationale for usingrealistic job previews, but Meyer and Allenpoint out that realistic job previews mayalso engender employee commitment. Em-ployees who are provided with candid infor-mation will presumably feel that the


organization has ‘‘laid its cards on the table,’’and the employees are able to make informedchoices about whether to join the organiza-tion. Such feelings of free choice mayenhance employees’ feelings of commitmentto the organization.

Providing realistic job previews may alsofacilitate commitment for more symbolic rea-sons. If an organization is honest, even aboutthe undesirable aspects of a job, recruits havea signal that the organization is going to treatthem in a fair and honest manner in thefuture. When these recruits become employ-ees, they will likely reciprocate such honestyand fairness with high levels of commitment.Conversely, if an employee feels that he orshe was provided with an overly positivepicture of the job prior to being hired, thissignals a lack of fairness and honesty in theorganization.

After employees enter an organization,their initial socialization and training expe-riences may have a strong impact on theirultimate level of commitment. Recall fromChapter 3 that organizations use a multitudeof strategies to socialize new employees (VanMaanen & Schein, 1979), and that new em-ployees may use a variety of strategies toobtain information (Miller & Jablin, 1991).Meyer and Allen (1997) point out that aninvestiture approach to socialization is likelyto lead to greater feelings of organizationalcommitment than a divestiture approach.Recall that when an investiture approach isused, the organization does not require thenewcomer to completely give up his or herold self. Rather, the organization allows thenewcomer to be a full-fledged member whilestill maintaining some individuality. Whatmessage does this convey? The newcomerreceives a message of affirmation and a will-ingness, on the part of the organization, torespect the rights of employees. Employeeswill often respond to this message with

greater feelings of commitment toward theorganization.

A divestiture approach to socialization,in contrast, requires the newcomer to essen-tially give up many aspects of his or her prioridentity and ‘‘fall in line’’ in order to assumefull membership in the organization. Thisform of socialization may suggest to the new-comer that the organization is elite, and thatachieving membership should be viewed as agreat privilege. On the other hand, it mayalso convey an unhealthy mistrust of out-siders and a condescending view of new-comers. Given these mixed messages, itwould seem possible that divestiture social-ization could lead to either very high or verylow levels of commitment.

In most recent empirical research on thesocialization process (e.g., Chao et al., 1994;Ostroff & Kozlowski, 1992), a consistentfinding is that the greatest initial concern ofnewcomers is to acquire task-related infor-mation. Before newcomers learn aboutthings such as the political climate of theorganization they want to be able to carryout their job-related tasks in a competentmanner. Given these initial concerns, orga-nizations must make sure that new employ-ees receive the training they need in order todo their jobs. Such training may require aformal training program or more informalon-the-job coaching activities.

Training may enhance organizationalcommitment because it conveys to new-comers that the organization is supportiveand has a vested interest in their success.Another reason is: If training ultimately facil-itates an employee’s success, this will likelyresult in positive outcomes for the employee(e.g., pay increases, promotions). If employ-ees recognize that the training they havereceived has contributed to their success,they are likely to be grateful to the organiza-tion. Such feelings of gratitude may very well


enhance the employee’s affective or norma-tive commitment.

Training may also contribute toenhanced feelings of continuance commit-ment. If an employee received training thatwas highly specific to a particular organiza-tion, this would greatly enhance the costassociated with leaving the organization. Asan example, much of the training thatmilitary personnel receive is so highly speci-alized that it does not transfer well to civilianjobs. For some individuals, this may enhancefeelings of continuance commitment andultimately contribute to a decision to pursuea longterm military career.

Given this apparent tradeoff betweenproviding general versus highly specifictraining, which course of action should anorganization choose? Training that is highlytransferable will likely engender feelings ofaffective commitment toward the organiza-tion. A potential drawback, however, is thatsuch training often does enhance employees’marketability. Highly specific training willenhance employees’ job performance, andthus may have the potential to enhance feel-ings of affective or normative commitment.This form of training may also heighten feel-ings of continuance commitment, and thusenhance employee retention. However, orga-nizations probably do not want employees toremain in the organization primarily becauseof feelings of continuance commitment. Ulti-mately, it is probably best for organizationsto have some balance between general andorganization-specific training programs.

The development of internal promotionpolicies is another area in which organiza-tions can make tangible use of organizationalcommitment research. As Meyer and Allen(1997) point out, promoting from within anorganization facilitates higher levels of com-mitment among employees. If an organiza-tion does pursue an internal promotion

policy, some important issues must beaddressed. Perhaps most important, anyinternal promotion initiatives should bepublicly made available to all employees. Ifemployees see internal promotion practicesas being unfair and secretive, the unintendedeffect may be a reduction of employees’ com-mitment. It also makes little sense to makepromotional opportunities available to em-ployees if the organization fails to help themacquire the skills necessary to be competitorsfor such opportunities. As was discussedpreviously, skill acquisition can certainly befacilitated by formal training programs.However, to provide employees with mean-ingful developmental experiences, organiza-tions may utilize other methods, such aslateral transfers or job rotation.

Organizations often utilize organiza-tional commitment research in the area ofcompensation and benefits. Although thereis some degree of variation, most organiza-tions tie their forms of compensation to em-ployee tenure. It is quite common, forexample, for organizations to require thatemployees accrue some minimum years ofservice before they can be vested in pensionprograms and receive matching contribu-tions in 401(k) savings plans. Such require-ments may enhance commitment, but it isprimarily of the continuance variety. Thus,having such requirements may induce em-ployees to remain in the organization but willnot necessarily motivate them to workharder on its behalf.

A more creative way that organizationsmay use compensation to enhance employ-ees’ commitment is through the use of profitsharing or employee stock ownership plans(ESOPS; Lawler & Jenkins, 1992). The ideabehind such plans is that employees benefitfrom the increased profitability of the orga-nization as a whole. This presumably helpsemployees to see the big picture and work for


the good of the entire organization. It is alsopossible that such compensation programsmay enhance all three forms of organiza-tional commitment. Having ownership inthe organization may evoke feelings of prideand identification that may ultimately facili-tate the employee’s affective commitment.Feelings of ownership may also evoke astrong sense of responsibility and moral obli-gation toward the organization; hence, nor-mative commitment may be heightened.Because employees in such compensationsystems stand to lose financially if they leavethe organization, continuance commitmentmay also be enhanced. These plans are dis-cussed in more detail in Chapter 9.

CHAPTER SUMMARY

In this chapter, we examined employee jobsatisfaction and the related topic of organi-zational commitment. Both topics are impor-tant, for theoretical and practical reasons.Job satisfaction is generally defined as em-ployees’ feelings of affect toward their jobs orjob situations, but may also contain cognitiveand behavioral components. Given the wayjob satisfaction is defined, the vast majorityof measures of job satisfaction have come inthe form of self-reports. Measures range fromthe very general Faces Scale to other meas-ures that allow researchers to assess employ-ees’ satisfaction with various facets of thework environment. As with any scales, meas-ures of job satisfaction must be evaluated onthe basis of construct validity. Two scales forwhich considerable evidence of constructvalidity has accumulated are the Job Descrip-tive Index (JDI) and the Minnesota Satisfac-tion Questionnaire (MSQ). More recently,the Job Satisfaction Survey (JSS) has shownconsiderable potential as a valid measure ofjob satisfaction.

Traditionally, the characteristics of jobs,and other aspects of the work environment,have explained differences in job satisfac-tion. In general, job satisfaction tends to behighest when the characteristics of a jobmatch the employees’ expectations in areasthat are deemed important. In recent years, ithas been proposed that job satisfaction is dueto cues from the social environment, as wellas stable dispositions. In reality, job satisfac-tion is likely the result of a complex inter-action among job characteristics, social cues,and dispositions, in addition to the culture inwhich employees are raised.

Research has shown that job satisfactionis consistently related to other measures ofpositive and negative affect. Evidence sug-gests, however, that job satisfaction is only avery weak predictor of absenteeism. Job sat-isfaction has been found to be related toturnover, albeit indirectly and only whenemployees perceive the existence of alterna-tive employment opportunities. In general,research suggests that the relationshipbetween job satisfaction and job perfor-mance is not strong. Under certain condi-tions, such as when rewards are directly tiedto performance, there is evidence that thetwo may be more strongly related. Evidencealso suggests that job satisfaction is a muchbetter predictor of organizational citizenshipbehavior than it is of inrole performance. Ithas also been shown that the satisfaction–performance relationship may be stronger atthe aggregate rather than the individual level.

Organizational commitment reflects em-ployees’ feelings of loyalty toward the orga-nization and their willingness to maintainmembership. Employees may be committedbecause they have positive feelings towardthe organization (affective), because theyrealize that the costs of leaving outweighthe benefits (continuance), or because theyfeel morally obligated to stay (normative).

Chapter Summary 163

Affective and normative commitment can beexplained largely on the basis of equitytheory. Feelings of commitment represent

employees’ desires to reciprocate what theyconsider fair and equitable treatment at thehands of the organization. Continuance


PAUL SPECTOR AND OUTCOMES OF JOB SATISFACTION

Soon after I completed graduate school in

1975, I read Work and Well-Being by Peter

Warr and Toby Wall that talked about the link

between stressful job conditions and employ-ee health and well-being. Inspired by that

book, I conducted a study that showed how

job satisfaction related to physical health

symptoms. Unfortunately, I did not have

much luck publishing that study as the field

was not very interested in the new area of job

stress. I then turned my attention to the link

between job satisfaction and turnover withbetter publication success.

From this work we learned that job dis-

satisfaction is an important antecedent to turn-

over; people who dislike their jobs are likely to

quit. However, the relationship between job

satisfaction and turnover is moderated by the

job market. In other words, unhappy people

can only quit when they can find other work.Thus job dissatisfaction can produce a desire

to quit, but people will only quit when they

find alternative employment.

In the middle 1980s I returned to study-

ing the role of job satisfaction in the job stressprocess. This has resulted in a research pro-

gram of more than 20 years linking job sat-

isfaction to other variables, including stressful

job conditions, counterproductive work

behavior (e.g., aggression at work, sabotage,

and theft), negative emotions, and physical

health symptoms. Much of the early work

was done with one of this book’s authors,Steve Jex. We looked at interpersonal conflict

and organizational constraints, two stressful

conditions of work that were not being

studied. We showed how they related, not

only to job satisfaction, but both behavior

and health as well. We also to research show-

ing that the personality variable of neuroticism

had a complex role in the process linkingstressful job conditions to job satisfaction.

For example, we learned that people high in

neuroticism tended to be found in jobs that

were of lower quality than people who were

low in neuroticism. Thus individuals high in

this personality trait have good reason to be

less satisfied with their jobs.

Although I have studied many variables inthe workplace, most of my studies have

included a measure of job satisfaction. This

is because job satisfaction is a contributor to

many important human phenomena in the

workplace.

Paul E. Spector


University of South Florida


commitment, on the other hand, is duelargely to employees’ perceptions of sunkcosts and the extent of alternatives.

Historically, the most popular measure oforganizational commitment has been theOrganizational Commitment Questionnaire(OCQ). A major limitation of the OCQ, how-ever, is that it measures only affective com-mitment. More recently, Allen and Meyer(1990) have developed a scale that measuresall three forms of commitment. Although thisscale is relatively new, evidence to date hasshown that it has excellent psychometricproperties. This scale will likely be the mostwidely used measure in future organizationalcommitment research.

Commitment has also been studied inorder to predict other variables. Affectivecommitment has been found to be consis-tently related to other attitudinal variables.Research, however, has not supported astrong link with absenteeism. This form ofcommitment has been found most stronglyrelated to turnover—a finding that is notsurprising, given the nature of this construct.Affective commitment appears to be relatedto performance only to the extent to which itincreases employee effort. Although consid-erably less research has been conducted oncontinuance and normative commitment,most studies have shown that these are pri-marily related to turnover. Recent researchsuggests that profiles of employees on differ-ent types of commitment may be an espe-cially strong predictor of performance. Inaddition, recent research highlights theimportance of potentially combining job sat-isfaction and organizational commitmentinto an overall measure of job attitude.

Commitment research also has a numberof practical applications. Organizations mayimpact employees’ feelings of commitmentduring the socialization process, as well asthrough other human resources manage-

ment policies. In general, human resourcesmanagement practices that convey a highlevel of organizational support tend to beassociated with high levels of affective andnormative commitment. Practices thatincrease employees’ sunk costs tend to engen-der feelings of continuance commitment.Organizations are typically best served byachieving some balance among affective,continuance, and normative commitmentamong their employees.


CooperHakim, A., & Viswesvaran, C.(2005). The construct of work commitment:Testing an integrative framework. Psycholog-ical Bulletin, 131, 241–259.

Harrison, D. A., Newman, D. A., & Roth, P.L. (2006). How important are job attitudes?Metaanalytic comparisons of integrativebehavioral outcomes and time sequences.Academy of Management Journal, 49, 305–325.

Ilies, R., & Judge, T. A. (2003). On theheritability of job satisfaction: The mediatingrole of personality. Journal of Applied Psychol-ogy, 88, 750–759.

Judge, T. A., Thoresen, C. J., Bono, J. E., &Patton, G. K. (2001). The job satisfaction-jobperformance relationship: A qualitative andquantitative review. Psychological Bulletin,127, 376–407.

Meyer, J. P., Allen, N. J., & Smith, C. A.(1993). Commitment to organizations andoccupations: Extension and test of a three-component conceptualization. Journal ofApplied Psychology, 78, 538–551.

Sanchez, J. I., Korbin, W. P., & Viscarra, D.M. (1995). Corporate support in the after-math of a natural disaster: Effects on employ-ee strains. Academy of Management Journal,38, 504–521.


Chapter Six

CounterproductiveBehavior inOrganizationsE

mployees typically behave in waysthat contribute positively to thegoals of their employing organiza-tions. That is, employees performtheir jobs to the best of their abil-

ities, occasionally go above and beyondthe call of duty, and may even come upwith innovative and creative ideas on theirown. Employees tend to engage in suchproductive behaviors because organizationsare selective in their hiring and, as will beshown in subsequent chapters, often setup motivational and leadership systems thatencourage such forms of behavior.

However, employees may also, at times,engage in behaviors that run counter to or-ganizational goals. Common forms of coun-terproductive behavior in organizationsinclude ineffective job performance, absen-teeism, turnover, and unsafe behavior. Otherforms of counterproductive behavior includeantisocial behaviors such as theft, violence,substance use, and sexual harassment.Although less common, these forms of be-havior may be quite destructive and ulti-mately costly to organizations.

This chapter examines counterproduc-tive behavior in organizations. In coveringthese forms of behavior, the emphasis willbe on understanding both the causes andthe consequences of such behaviors. A re-lated objective is to explore ways in which anorganization can eliminate these behaviorsor at least keep them at a level that isnot too destructive to the goals of the orga-nization.

DEFININGCOUNTERPRODUCTIVEBEHAVIOR

Most readers have probably received poorservice at a restaurant, or experienced theinconvenience of a long wait, brought aboutby poor scheduling or staffing shortages, ata doctor’s office. While obviously annoying,these experiences represent relatively mild

167

forms of counterproductive behavior inorganizations. More dramatic forms of coun-terproductive behavior, such as criminalactivity or violence, may have very negativeconsequences and become very newsworthyevents. For example, when a governmentemployee commits espionage, national secu-rity may be compromised, and media atten-tion surrounding such a crime is typicallyintense. Likewise, when a disgruntled emp-loyee enters an organization and fatallyassaults several coworkers, lives are perma-nently altered, and the event receives con-siderable media attention.

The specific examples in the precedingparagraph are all different, but each repre-sents a form of counterproductive behavior inorganizations. For the purposes of this chap-ter, counterproductive behavior will be definedas behavior that explicitly runs counter to thegoals of an organization. This definition isbased on a number of underlying assump-tions. For example, it is assumed that orga-nizations have multiple goals and objectives.A major goal of private organizations is prof-itability, but such organizations may havemany others as well. These may include ahigh level of customer service, a harmoniouswork environment, and the reputation ofbeing socially responsible. According to theprevious definition, if an employee engagesin behaviors that make it more difficult for anorganization to achieve any of its goals, theemployee is engaging in counterproductivebehavior.

The previous definition also makes noassumptions regarding the motives under-lying counterproductive behavior, althoughmuch of the research on counterproductivebehavior is focused on this very issue (Schat& Kelloway, 2005). A retail employee whosteals merchandise from his or heremployer is obviously doing it intentionallyand, most likely, for personal gain. On the

other hand, it is entirely possible for anemployee to engage in counterproductivebehavior without intending to. For exam-ple, an employee who is poorly trained orwho lacks job-related abilities may wantvery badly to perform well, but may notaccomplish that goal.

Finally, the previous definition makesno assumption as to the causes underlyingcounterproductive behavior. Recall fromChapter 4 that productive behaviors likelyresult from a complex interaction betweencharacteristics of individuals and character-istics of the environment. This same per-spective is adopted in the examination ofcounterproductive behavior. In fact, onecan make a strong argument for a person-by-environment interaction for literally allforms of counterproductive behavior (Fox& Spector, 1999). When an employee per-forms his or her job poorly, this may be dueto limited ability, but may also be partiallycaused by poor task design. Likewise, whenan employee engages in a violent act atwork, this may be due to deep-seated psy-chiatric problems, but may also be exacer-bated by an authoritarian organizationalclimate.

Based on the definition provided, thereare undoubtedly many forms of counter-productive behaviors in organizations.In organizational psychology, however,only a handful of these behaviors havereceived empirical scrutiny. The most com-monly studied counterproductive behaviorshave been ineffective job performance,absenteeism, turnover, and accidents. Morerecently, organizational researchers havebegun to examine several other forms ofcounterproductive behavior that are lesscommon, but are potentially more devastat-ing to organizations. These include theft,violence, substance use, and sexual harass-ment.

168 Counterproductive Behavior in Organizations

INEFFECTIVE JOBPERFORMANCE

Most people who go to work each day wantto do their jobs well, and for good reason.High levels of performance are often associ-ated with positive tangible outcomes such asmerit increases, cash bonuses, promotionalopportunities, and the like. Performingwell may also lead to intangible rewards suchas praise and admiration from others, and aheightened sense of personal accomplish-ment. Despite all the logical reasons for per-forming well, some employees do notperform up to par. Ineffective job perfor-mance is often a difficult issue for organiza-tions, for a number of reasons. For example,in many cases, it may be difficult foran organization to detect ineffective perfor-mance in the first place. Once detected, itis often challenging to diagnose the cause ofthe performance problem. Finally, organiza-tions often struggle with the issue of how torespond to, and prevent, instances of ineffec-tive performance. Each of these issues is dis-cussed in the following paragraphs.

Detection of IneffectivePerformance

Recall from Chapter 4 that models of job per-formance propose that behaviors constitutingjob performance may be categorized into anumber of different types, such as core tasksthat are specific to the job, and moregeneral or peripheral tasks. Ideally, all orga-nizations would have in place performancemeasurement systems that would allowassessment of the many behaviors that con-stitute the performance domain. If this werethe case, a routine performance appraisalwould be quite useful in the detection ofineffective performance. Unfortunately,most performance measurement systems

typically provide information about theimpact of employee behavior, but far lessinformation about the behaviors themselves.

Performance-related data that organiza-tions typically collect may be classified intofour different types: personnel data, produc-tion data, subjective evaluations, and morerecently electronic performance monitoring.Personnel data include items such as ab-sences, sick days, tardiness, disciplinaryactions, and safety violations. Some of these,as will be shown later in the chapter, arecounterproductive behaviors for which per-sonnel data provide a direct measure. Per-sonnel data may also, at times, provide usefulinformation in the diagnosis of the cause ofperformance problems. For example, an em-ployee who is absent or late may have asubstance abuse problem that ultimately neg-atively impacts his or her performance.

Production data provide an organizationwith useful information about tangible out-comes associated with job performance. Themost commonly used form of productiondata is probably sales commissions, althoughproduction indexes may be used in manyother settings. As a means of detecting inef-fective job performance, there are clearlyadvantages to using production data. Suchdata provide organizations with an objectiveperformance metric that an employee cannotdispute (i.e., numbers don’t lie). Such dataare also typically not costly to obtain becausethey are often collected for multiple pur-poses.

A potential drawback with productiondata is that they often provide an overlysimplistic view of employee performance. Asalesperson may exhibit reduced sales com-missions in a particular year, yet thesenumerical data provide an organization withlittle information about the source of theperformance problem. Also, in the firstauthor’s experience, reliance on production


data may lead supervisors to adopt a some-what callous attitude toward subordinateswho are experiencing a performance prob-lem. The response to reduced sales commis-sions may be: ‘‘Increase your sales, or else!’’

By far the most common form of emp-loyee performance data comes from subjec-tive appraisals. Most typically, an employee’simmediate supervisor(s) completes someperformance appraisal instrument on anannual or semiannual basis. In consideringsubjective appraisals, it is important to keepin mind that what is actually being measuredin most cases is the result of employees’behavior, or, more specifically, employee

effectiveness (Pritchard, 1992). To be sure,some organizations may invest the time andeffort required to develop and implementelegant behaviorally based performance ap-praisal systems. Most organizations, how-ever, still tend to rely on performance ap-praisal instruments that utilize rather generaldimensions of employee performance andengage in rather minimal efforts to trainraters (Cascio, 1998).

As a method of detecting ineffective per-formance, subjective appraisals have certainadvantages when compared to either person-nel data or production indexes. Comparedto production data, a supervisor’s thoughtful

COMMENT 6.1

THE IMPORTANCE OF PERFORMANCE FEEDBACK

WITHOUT A DOUBT, one of the least favorite tasks

of managers and supervisors is conducting

annual or semiannual performance reviews.

This is particularly true when an employee isperforming poorly. Ironically, though, perfor-

mance reviews have the potential to provide

the greatest benefit to those employees who

are not performing well—provided they are

done well.

Research on conducting performance

reviews has shown that there are several attrib-

utes of an effective performance review. One ofthe most important of these, particularly for a

poorly performing employee, is that the tone of

the review should be constructive rather than

punitive. An employee who is performing

poorly is likely to respond much more favor-

ably to a supervisor who says, ‘‘What can I do to

help you improve?’’ than to a supervisor who

lists all of the things that the employee is doingpoorly.

It is also very important that the feedback

provided to an employee is specific and is

focused on behavior. Telling a poorly perfor-

ming employee that he or she has a ‘‘bad

attitude’’ doesn’t provide that employee with

much diagnostic information. On the other

hand, telling the same individual that he or

she often does not thank customers after com-pleting a transaction is much more specific

and, more importantly, is something the indi-

vidual can change.

Finally, research has also shown that

performance reviews should be conducted

separately from salary reviews. When the per-

formance and salary reviews are conducted

together (which is common), most people tendto focus disproportionately on the size of their

salary increase. Unfortunately, what ends up

receiving much less attention in comparison is

performance feedback and, if necessary, sug-

gestions for improvement. This is particularly

unfortunate for employees performing poorly;

these individuals stand to benefit the most

from focusing on performance.

Source: H. H. Meyer, E. Kay, and J. R. P. French, Jr. (1965).

Split roles in performance appraisal. Harvard Business

Review, 43, 123–129.


consideration of an employee’s performancemay provide considerably greater insight intothe root causes of ineffective employee perfor-mance. Also, if appraisals are performed welland the information is regularly transmitted toemployees (e.g., Meyer, Kay, & French, 1965),they may prevent ineffective performancebefore it occurs (see Comment 6.1).

Despite these potential advantages, sub-jective appraisals are often of marginal valuein the detection of ineffective performance.Because many organizations still utilize per-formance appraisal instruments that assessvery global performance dimensions, suchratings may often fail to reveal performanceproblems. Also, despite the considerabletechnical advances in performance appraisalmethodology over the past 25 years (e.g.,Borman, 1991; Murphy & Cleveland,1990), many organizations still administerperformance appraisals very poorly or sim-ply ignore them.

A final method of detecting ineffectiveperformance, and in some cases other formsof counterproductive behavior, is electronicperformance monitoring (Zweig & Webster,2002). It is common, for example, for em-ployees in customer service call centers to bemonitored by supervisors when they interactwith customers on the telephone. Researchhas shown that electronic performance mon-itoring can be an effective method of detect-ing ineffective performance, particularlywhen employees know in advance that it istaking place (Hovorka-Mead, Ross, Whip-ple, & Renchin, 2002). However, it has alsobeen shown that many employees view it asan invasion of privacy (Aiello & Kolb, 1995).

Causes of Ineffective Performance

Let’s assume for the moment that an instanceof ineffective performance has been detected.A salesperson has failed to meet his or her

quota for 3 consecutive months; a clericalemployee repeatedly makes mistakes on hisword-processing assignments; a universityprofessor repeatedly receives negative assess-ments of his or her teaching performance. Ineach of these cases, all we know is that theemployee is not performing up to par. Whatis often not known is why the employee isperforming poorly.

In many organizational settings, the un-derlying causes of ineffective performanceare often unclear. As a result, the cause(s)of ineffective performance must be deter-mined by attributional processes; that is,after observing some instance of ineffectiveperformance, a supervisor must make somejudgment about the cause(s) of this behavior.Attribution theory suggests that people makeuse of several pieces of information whendetermining the causes of another person’sbehavior (Kelley, 1973). For instance, peo-ple examine the consistency of behaviorover time, between different settings or con-texts, and in comparison to others. Thus,if an instance of poor performance wereencountered, a supervisor would ask ques-tions such as: Is the poor performanceconsistent with this employee’s past perfor-mance? Does he/she perform poorly on allaspects of the job or just certain ones? Is thelevel of performance poor compared to otheremployees?

If a supervisor is able to find answers tothese three questions, he or she is likelyto make some determination as to the causeof the ineffective performance. Generallyspeaking, if the ineffective performance isconsistent over time and settings, and is seenas poor in relation to others, a logical con-clusion would be that the ineffective perfor-mance was due to a lack of ability ormotivation, both of which are internal tothe employee. In contrast, if the ineffectiveperformance is not a consistent pattern over


time and settings, and is not seen as beingpoor in relations to others, a supervisorwould likely conclude that the ineffectiveperformance was due to factors external tothe individual (e.g., poor task design, inter-ruptions from others).

Unfortunately the attribution process isnot always accurate. Furthermore, peoplemay in fact hold certain biases in assess-ing the causes of others’ behavior. The bestknown of these is termed the fundamentalattribution error (Ross, 1977) and refers tothe bias toward attributing the causes ofothers’ behavior to internal, as opposed toexternal, causes. Although the reasons forthis bias are complex, the basic issue is that,in most situations, people are more distinc-tive than the situations they are in. Thus,when any behavior occurs, there is a ten-dency to focus on personal (as opposed tosituational) factors being the cause.

There is evidence that the fundamentalattribution error may impact diagnoses ofineffective performance, although severalfactors may influence it. For example, in alaboratory study, Mitchell and Kalb (1982)found that supervisors who lacked ex-perience in the tasks their subordinates per-formed tended to attribute poor performanceto internal causes. In contrast, those withmore task experience made more externalattributions. In another laboratory study,Ilgen, Mitchell, and Frederickson (1981)found that supervisors who were highlyinterdependent with subordinates tendedto make more external attributions for inef-fective performance; supervisors who sawlittle interdependence tended to make moreinternal attributions.

Understanding the attributional proc-esses involved in determining the causes ofineffective performance is important becausesuch attributions may have a strong impacton supervisory responses to ineffective per-

formance. For example, if a supervisor seesthe cause of the ineffective performance asbeing poor task design, his or her responsemay be quite different than if it is seen as dueto a lack of effort. Ilgen et al. (1981) foundthat supervisors responded to ineffectiveperformance more favorably when theyattributed it to external (versus internal)causes. It has also been shown that super-visors react more favorably to ineffective per-formance when they perceive it as beingcaused by a lack of ability, as opposed to alack of motivation (Podsakoff, 1982). This ispresumably because employees have lesscontrol of the former than the latter.

For the moment, we’ll take as a given thatdetermining the cause(s) of ineffective per-formance often requires the use of imperfectattributional processes. What then are themost common causes of ineffective perfor-mance? To answer this question, it is usefulagain to think back to Chapter 4 and thediscussion of the causes of productive behav-iors, such as job performance. Based on thisvast literature, it can be concluded that inef-fective performance may be due to employ-ees’ inability to perform their jobs effectively(e.g., lack of ability, lack of skills, poortraining), lack of willingness to performeffectively (e.g., unwilling to put forth orsustain effort, putting efforts in the wrongdirection), or aspects of the environment thatprevent the employee from performingwell (e.g., poor task design, ineffective co-workers).

In examining each of these causes ofineffective performance, there are tangibleorganizational activities that may contributeto them. Despite the advances that havebeen made in employee selection (Guion &Highhouse, 2004), selection is still an imper-fect process. As a result selection errors mayresult in organizations hiring individualswho lack either the skills or the abilities


necessary to perform their jobs. Selectionerrors may also occur when employees pos-sess the requisite skills and abilities neces-sary to perform their jobs, but simply do notfit well into the culture of the organization(Kristof, 1996).

How can organizations avoid selectionerrors? At the risk of sounding overly sim-plistic, organizations simply need to put asystematic effort into employee hiring. Whilemany organizations clearly do this, manyothers do not. More to the point, many or-ganizations simply fail to gather and utilizedata that would help them make moreinformed hiring decisions. Although a com-plete exploration of the employee selection isclearly beyond the scope of this chapter (seeGuion & Highhouse, 2004, for completecoverage), selection errors may often beavoided by the systematic use of tests, per-sonal history information, and background/reference checks.

Another way in which organizations maycontribute to ineffective performance isthrough inadequate socialization and training.As was pointed out in Chapter 3, whenemployees first enter an organization, theytypically need to be trained on specific job-related skills, as well as more general infor-mation about the culture of the organization(Goldstein & Ford, 2002). Employees whoreceive either inadequate training or notraining at all may be set up for failure whenthey enter an organization. In such an envi-ronment, only those who have very highlevels of ability and self-confidence may sur-vive.

With respect to socialization, organi-zations may make a number of errors thatcould lead to poor performance among em-ployees. Specifically, failing to provide newemployees with information about impor-tant aspects of the culture of the organizationmay lead to failure. For example, if the

culture of an organization is such thattimely completion of work is highly valued,a new employee may inadvertently performpoorly by not completing work on time.Typically, this type of situation is resolvedwhen the new employee realizes the value oftimeliness.

A more problematic situation occurswhen new employees receive mixed signalsabout the culture of the organization andhow this relates to performance expecta-tions. Both authors have found that this is atypical problem for faculty at medium-sizeregional universities. Because such institu-tions do offer a limited number of doctoralprograms, some faculty and administratorsfeel that an organizational culture that placesa strong emphasis on research is appropriate.On the other hand, many institutions thissize have historically placed a strong empha-sis on undergraduate education, so manyothers feel that the culture should place astrong emphasis on teaching excellenceand availability to undergraduate students.Although such differences in philosophymay sometimes lead to insightful dialogue,they often prove to be very confusing to newfaculty members who must decide where tofocus their efforts.

Finally, in some cases, employees maywant to perform well but are preventedfrom doing so because of constraints in theenvironment. For example, an employee’sjob tasks may be designed in a way thatmakes it difficult to perform well, or in away that is incompatible with the organiza-tion’s reward systems (Campion & Berger,1990; Campion & Thayer, 1985). For exam-ple, if it is crucial for an employee to makeindependent judgments in order to performeffectively, it would not make sense to designthe job in a way that denies this employeedecision-making authority. Even if tasks aredesigned properly, other constraining forces


in the work environment may hinderperformance (Peters & O’Connor, 1980;Spector & Jex, 1998). For example, employ-ees may be unable to perform well because ofinterruptions from others, poor tools orequipment, and perhaps poor informationfrom others.

Management of IneffectivePerformance

Given the multitude of factors that may con-tribute to ineffective performance, managersneed to investigate its causes when it occurs.Thus, as a first step toward investigatingineffective performance, managers shouldtalk to the employee. Perhaps more impor-tantly, such discussions should involve con-siderable listening on the part of the manager(Meyer et al., 1965). Depending on the out-come of the conversation with the poorlyperforming employee, a number of correc-tive actions may be utilized by the managerto improve performance. In some cases, itmay be possible to improve an employee’sperformance through relatively straightfor-ward training interventions. For example, ifan employee is consistently producing poor-quality written reports, a logical way toimprove performance might involve someform of training aimed at improving his orher written communication skills.

In other cases, the underlying cause(s)of ineffective performance may not be asobvious. Let’s say, for example, that a realestate salesperson is failing to produce accep-table commissions. For a sales manager toaccurately diagnose the cause of this partic-ular performance problem, he or she mayneed to actually observe the employee tryingto close a sale. This type of activity maybe thought of as on-the-job coaching of theemployee. Coaching is a form of training, butit is much more extensive and time consum-

ing. The manager who provides coachingto employees is engaged in a form of activelearning that may involve examining allaspects of the employee’s performance-related behavior. Another option in dealingwith ineffective performance is the use ofcounseling and employee assistance programs(EAPs) (Cooper, Dewe, & O’Driscoll, 2003).Employees do not compartmentalize theirlives; thus, problems outside of work maymanifest themselves in the workplace. Mar-ital or financial problems may have a nega-tive impact on the performance of evenhighly competent employees. If this optionis considered, however, managers must bevery careful how they approach the em-ployee. Even if such efforts are well inten-tioned, the suggestion that employees needto seek such services may be met withconsiderable resistance on the part of anemployee. Despite these potential caveats,providing counseling or EAPs may be veryuseful ways of dealing with some instances ofineffective performance.

As is evident by now, organizations havea variety of ways of dealing with ineffectiveperformance when it occurs. Ideally, though,organizations usually prefer to prevent inef-fective performance before it occurs. A firststep toward preventing ineffective perfor-mance is the utilization of scientifically basedselection programs. There is considerableevidence that some variables—most notably,general cognitive ability, conscientiousness,and prior experience—predict performanceacross a variety of job types (Barrick &Mount, 1991; Schmitt, 2004; Schmidt &Hunter, 1998). Thus, organizations that em-ploy rigorous selection programs increasethe probability that employees enter withthe skills, abilities, and personality traits nec-essary to perform their jobs.

Once employees enter an organization,steps must also be taken to nurture employees’


skills and abilities so they are translated intoperformance (Colarelli et al., 1987). Asstated earlier, one way of addressing thisissue is through proper training and social-ization. The manner in which organizationsconduct initial training and socializationvaries widely (see Chapter 3). Organizationsthat take the time to properly socialize andtrain new employees clearly stand a goodchance of avoiding performance problemsin the future.

A final step toward the prevention ofperformance problems is having a systematicperformance measurement and feedback sys-tem. This helps to keep employees on trackwith respect to performance, and serves tocommunicate performance expectations. Inmany cases, ineffective performance maysimply be due to the fact that employees donot know what the organization (or theirimmediate supervisor) expects. Regular per-formance evaluations also signal to employ-ees that performance matters.

EMPLOYEE ABSENTEEISM

The second form of counterproductive be-havior examined in this chapter is employeeabsenteeism. Recall from the previous chap-ter that absenteeism was discussed, but onlyas a potential consequence of job dissatisfac-tion and low organizational commitment. Inthis section, we approach absenteeism from asomewhat broader perspective and examineother predictors, as well as various ways inwhich organizations can reduce the inci-dence of employee absenteeism.

Defining and Measuring Absenteeism

Absenteeism appears to be a relatively simplevariable to define and measure; that is, absen-teeism can simply be defined as not showingup for work. However, defining absenteeism

in such a general way is problematic whenthe goal is to predict and control absentee-ism. In the absenteeism literature, research-ers typically make some distinctions withrespect to the types of absences. The mostcommon distinction is between excusedand unexcused absences. Excused absenceswould be those due to reasons that the orga-nization deems as acceptable (e.g., illness,caring for a sick child). In contrast, unex-cused absences would be those that areeither due to unacceptable reasons (e.g.,decided to go shopping) or cases in whichan employee has not followed proper proce-dures (e.g., neglected to call in to one’ssupervisor). Making distinctions betweentypes of absences is important because differ-ent types may be caused by different varia-bles. To underscore this point, Kohler andMathieu (1993) examined a number of pre-dictors of seven different absence criterionmeasures among a sample of urban bus driv-ers and found different predictors for dif-ferent criteria. For example, they foundthat absences due to nonwork obligations(e.g., caring for children, transportationproblems) were most strongly related to var-iables such as dissatisfaction with extrinsicfeatures of the job, role conflict, role ambi-guity, and feelings of somatic tension. On theother hand, absences due to stress reactions(e.g., illnesses) were most strongly related todissatisfaction with both internal and ex-ternal features of the job, feelings of fatigue,and gender (women were absent more fre-quently).

To measure absenteeism, the most com-mon indexes are time lost measures and fre-quency measures (Hammer & Landau,1981). When a time lost measure is used,absenteeism is represented by the numberof days or hours that an employee is absentfor a given period of time. As an example, ifan employee is absent from work 3 days


over a 3-month period, that employee’slevel of absenteeism would be 3 days or24 hours (assuming that each workday is8 hours).

If a frequency metric is used, absentee-ism represents the number of absenceoccurrences for a given period of time. Anoccurrence can range from 1 day to severalweeks. In the previous example, if each ofthe 3 days that the employee is absentoccurs in a different month, the time lostand the frequency metrics would be iden-tical. However, if the employee was absentfor 3 consecutive days, the absence wouldbe recorded as only one occurrence if afrequency metric is used.

Although both time lost and frequencymeasures of absenteeism have been used instudies of absenteeism (e.g., Hackett &Guion, 1985; Steel & Rentsch, 1995), timelost measures are generally more desirablebecause they exhibit greater variability thanfrequency measures (Hammer & Landau,1981). Thus, it is generally more difficult topredict absenteeism when using frequency-based absenteeism measures.

Another important issue in the measure-ment of absenteeism is the time frame usedto aggregate absences. In terms of aggrega-tion periods, studies can be found in whichabsenteeism data are aggregated over periodsranging from as short as 1 month to as longas 4 years (Hammer & Landau, 1981; Steel& Rentsch, 1995). The primary advantage tousing longer aggregation periods is that thedistributions of such measures are not aslikely to be skewed as those from shorterperiods. Given that absenteeism is a lowbase-rate event (as are many other forms ofcounterproductive behavior) even for rela-tively long periods of time, aggregatingabsenteeism data over a very short periodof time may pose researchers with somevexing statistical problems.

Predictors of Absenteeism

For many years organizational psychologistshave focused on affective predictors ofabsenteeism, such as job satisfaction andorganizational commitment. As was shownin Chapter 5, however, meta-analytic reviewshave generally found the relationship be-tween affect and absenteeism to be ratherweak (e.g., Hackett, 1989; Hackett & Guion,1985; Mathieu & Zajac, 1990). These find-ings can be attributed at least in part to theissues discussed previously (e.g., choice ofabsenteeism measures, aggregation periods).They also may be due to the fact that theremay be other variables that predict absentee-ism better. In this section, we go beyondaffective variables and review other predic-tors that have been explored in the absentee-ism literature.

As a first step toward understanding ab-senteeism, it is useful to consider employeeattendance decisions in a general sense. Ac-cording to Steers and Rhodes (1978), twogeneral factors—the ability to attend and thedesire to attend—determine employees’attendance. Ability to attend is determinedlargely by an employee’s health but may alsobe due to factors such as nonwork respon-sibilities, reliability of transportation, andweather. The desire to attend work is deter-mined to a large extent by employees’ feel-ings about the organization or job, but mayalso be due to other factors. For example, anemployee may like his or her job but choosenot to attend because of some more attrac-tive nonwork alternative. For example, anemployee may choose to be absent on aparticular day in order to go holiday shop-ping.

Based on this view of absenteeism, threenonaffective variables have been shown tostand out as consistent predictors of absente-eism. For example, it has been consistently


found that women tend to be absent fromwork more frequently men (Farrell & Stamm,1988; Steel & Rentsch, 1995; VandenHeuvel& Wooden, 1995). Based on Steers andRhodes (1978), this is probably becausewomen are more likely than men to be insituations that constrain their ability to attendwork. For example, it has been shown that,even in dual-career situations, women tendto assume primary responsibility for child-care and household chores (Hochschild,1989).

Another important nonaffective predic-tor of absenteeism is the nature of an orga-nization’s absence-control policies. Someorganizations are quite lenient; they choosenot to even record employees’ absences. Atthe other extreme, some organizations re-quire extensive documentation for the rea-son for absences, and respond with strictdisciplinary actions when employees areabsent frequently. As one might expect, thefrequency of absenteeism tends to be lowerin organizations that have more strictabsence-control policies (Farrell & Stamm,1988; Kohler & Mathieu, 1993; Majchrzak,1987). It is important to note, however, thatsimply having a strict absence-control policyin place may not always reduce absenteeism.For example, Majchrzak (1987) found thatin Marine Corps units where the absence-control policy had been communicatedclearly and applied consistently, unauthor-ized absences were reduced significantlyover a 6-month period. In contrast, absencesremained constant in units where no policyexisted, or where the policy was not commu-nicated clearly.

Another important nonaffective predictorof absenteeism is absence culture. The termwas originally defined by Chadwick-Jones,Nicholson, and Brown (1982) as ‘‘the beliefsand practices influencing the totality of ab-sence frequency and duration—as they cur-

rently occur within an employee group ororganization’’ (p. 7). There are two things tonote about this definition. First, absence cul-ture is a group- or organization-level con-struct, and thus must be measured at theappropriate level (e.g., group or organiza-tion). Second, because organizations typi-cally consist of multiple groups, severalabsence cultures may in fact be operatingsimultaneously in the same organization.

Given that normative standards serve asan important guide for the members of anysocial unit (Hackman, 1992), one wouldexpect that group members’ absenteeismwould tend to be consistent with the prevail-ing absence culture. Unfortunately, to date,there has been relatively little empiricalinvestigation of the absence culture con-struct or of its effects on absenteeism. Oneexception is a study in which Mathieu andKohler (1990) examined the impact ofgroup-level absence rates on individual ab-sences. Using a sample of transit operatorsemployed by a large public transit authority,they found that the level of absences in thevarious garages in which these employeesworked predicted absenteeism using atime-lost measure.

A more direct test of the effect of absenceculture comes from a study conducted byMartocchio (1994). Unlike the methodin the Mathieu and Kohler (1990) study,Martocchio actually assessed absence culturewithin groups and investigated the impact ofthis variable on absenteeism. Based ona sample of clerical employees at a Fortune500 company, Martocchio found that group-level beliefs regarding absenteeism (e.g., ab-sence culture) were predictive of individualemployees’ absenteeism, measured in termsof the frequency of paid absences. Indi-viduals’ beliefs regarding absenteeism pre-dicted only the frequency of unpaidabsences. Figure 6.1 summarizes all of the


factors that have been shown to influenceemployee absenteeism.

Overall, the research on absenteeism pro-vides fairly clear guidance for organizationswishing to reduce absenteeism. More specif-ically, organizations need to have in placeabsence-control policies that are fair, yet atthe same time, discourage employees fromunnecessary absences. It is also important fororganizations to foster a positive absenceculture in the organization as a whole, aswell as within work groups. Finally, in somecases organizations can prevent absences byhelping employees overcome barriers toattendance. Benefits such as sick child care,flexible schedules, and telecommuting mayallow employees to reduce absences that aredue to nonwork demands.

Cross-Cultural Differencesin Absenteeism

Like most phenomena studied by organiza-tional psychologists, absenteeism has beenexamined largely in samples of either Amer-ican or Western European employees.Despite calls for cross-cultural absenteeismresearch (e.g., Martocchio & Harrison,

1993), few studies have examined cross-cultural differences in absenteeism. Onenotable exception is a study by Johns andXie (1998). Employees from the People’sRepublic of China and from Canada werecompared on a number of aspects of absen-teeism, such as perceptions of their ownabsence levels in comparison to those in theirwork groups; manager–subordinate agree-ment on absence norms; and legitimacy ofreasons for absenteeism.

The most notable cross-cultural differ-ence found in this study was that Chineseemployees were more likely than Canadiansto generate estimates of their own absentee-ism that favored their work group. This sug-gests that absence norms may be a morepowerful predictor of absenteeism amongthe Chinese. Along these same lines, it wasfound that Chinese managers were in greateragreement with their work groups on ab-sence norms than were Canadian managers.Finally, with respect to reasons for absence,the Canadians were less likely than theChinese to see domestic reasons as a legit-imate excuse for absences. In contrast, theChinese were less likely than the Canadiansto see illness, stress, and depression as legit-imate excuses.

Johns and Xie (1998) attributed theirfindings to well-documented differencesin values between Western and Easternsocieties. Most notably, in Eastern societies,the strong collectivist orientation suggeststhat social norms regarding such behaviormay have a more powerful effect than theydo in Western societies. This may alsoexplain why those in collectivist societiesmay see absences due to family reasons asmore legitimate than do those in more indi-vidualistic societies. In contrast, in Easternsocieties, norms surrounding the expressionof feelings may prohibit absenteeism basedon poor mental or physical health. The

FIGURE 6.1Summary of the Major Determinants of EmployeeAbsenteeism

EmployeeAffect

DemographicCharacteristics

Absenteeism

AbsenceCulture

AbsencePolicies


results of this study are provocative; theysuggest that cross-cultural absenteeismresearch may be a fruitful area of researchin the future.

EMPLOYEE TURNOVER

Like absenteeism, employee turnover wasdiscussed in the previous chapter as a corre-late of job satisfaction and organizationalcommitment. Furthermore, compared toabsenteeism, employee affect has been shownto be a stronger predictor of turnover deci-sions. Therefore, the focus in this section willbe to examine employee turnover from amore macro perspective (e.g., the impact ofturnover on organizations), explore nonaf-fective predictors of turnover, and, finally,explore a recent model that has appliedbehavioral decision theory to the study ofturnover.

The Impact of Turnoveron Organizations

Like many variables explored in this book,employee turnover has been studied from avery micro perspective; that is, researchershave sought to more fully understand theindividual-level decision-making processesthat characterize turnover decisions. Organi-zational researchers have generally paidmuch less attention to examining the impactof employee turnover on organizational effec-tiveness. Abelson and Baysinger (1984), dis-tinguished between optimal ternover and dys-functional turnover. Optimal turnover occurswhen poorly performing employees decide toleave an organization. These authors also sug-gest that, in some cases, turnover may beoptimal even if a high-performing employeeleaves because the cost of retaining that indi-vidual may be too high. This situation occursfrequently in professional sports when play-

ers from teams in relatively small marketsbecome free agents. These teams want toretain their star players, but they simply donot have the financial resources to do so.

Like optimal turnover, dysfunctionalturnover can be viewed in multiple ways. Ifan organization’s rate of turnover isextremely high, this can be very dysfunc-tional. High rates of turnover translate intoincreased costs associated with constantlyhaving to recruit and train new employees.A consistently high rate of turnover may alsoserve to tarnish the image of the organization(see Highhouse and Hoffman, 2001) andthus make it even harder to attract newemployees. In most industries, there areorganizations that have a reputation of‘‘chewing up and spitting out’’ employees.

Turnover is also dysfunctional if there isa consistent pattern whereby good employ-ees leave. As stated previously, in some casesthe cost of retaining high-performing em-ployees may be prohibitive; thus, some levelof turnover among high-performing employ-ees is inevitable. Unfortunately, if this is aconsistent pattern then an organization willbe losing valuable human capital.

Another way to view the impact ofturnover on organizations is to distinguishbetween what might be termed avoidableturnover and unavoidable turnover. Turnoveris avoidable when there are steps that anorganization could have taken to prevent it.As argued previously, this is somewhatsubjective and involves weighing the costsof losing employees versus the benefitsof retention. Unavoidable turnover, on theother hand, is illustrated by situations inwhich an organization clearly cannot preventan employee from leaving. This may occurwhen an employee’s spouse is transferred toanother location, or when there is simply noneed for the employee’s services. In othercases, turnover may be unavoidable simply


because an employee decides to withdrawfrom the labor force.

Nonaffective Predictors of Turnover

One nonaffective predictor that has actuallyreceived a fair amount of attention in theturnover literature is performance. Organiza-tions obviously prefer that turnover is highestamong lower-performing employees. Fur-thermore, empirical evidence has supportedsuch a negative relation between performanceand turnover (e.g., McEvoy & Cascio, 1987;Williams & Livingstone, 1994), although thisrelation is not strong. The relative weaknessof the performance-turnover relation may bedue to a number of factors. As Hulin (1991)has argued quite forcefully, turnover is a lowbase-rate event, and studies employing typi-cal parametric statistical procedures mayunderestimate the true relation betweenturnover and other variables. This becomeseven more problematic when performance isexamined as a predictor of turnover because,due to a variety of factors, the variability injob performance measures may be severelyrestricted (e.g., Jex, 1998; Johns, 1991).

A more substantive variable that mayimpact the performance-turnover relation isorganizational reward contingencies. One ofthe assumptions underlying the predictionthat turnover is negatively related to perfor-mance is that low performers will receivefewer organizational rewards than highperformers. Because of this, low-performingemployees are likely to become dissatisfiedand seek employment elsewhere. Given thatorganizations vary widely in the extent towhich they reward on the basis of perfor-mance, this would certainly account forthe weak performance-turnover relation.Furthermore, in one meta-analysis of theperformance-turnover relation (Williams &Livingstone, 1994), the average correlation

between performance and turnover wasstrongest in studies conducted in organiza-tions where rewards were tied to perfor-mance.

A third factor that may impact the per-formance-turnover relation is the form of thisrelationship. As in most studies in organiza-tional psychology, it has been assumed thatthe performance-turnover relation is linear.Jackofsky (1984), however, has argued thatthe performance-turnover relation may infact be curvilinear and best described by aU-shaped function. This means that turnovershould be highest among employees perfor-ming at very low and at very high levels.Jackofsky (1984) argued that, in most cases,very low performers are not going to berewarded very well and thus may becomedissatisfied. As performance moves towardmedium levels, employees are probablybeing rewarded at a level that keeps themfrom becoming extremely dissatisfied, andthus seeking alternative employment. As per-formance increases, however, there is agreater likelihood that employees will haveattractive alternative employment opportuni-ties and thus may be more likely to leave anorganization. This may even be true in orga-nizations that reward on the basis of perfor-mance. Employees who are extremelytalented may be receiving top salaries in aparticular organization, but organizationssimply may not be able to match whatanother organization is willing to pay inorder to lure the employee away.

To date, Jackofsky’s (1984) curvilinearhypothesis has not received a great deal ofempirical investigation, although it hasreceived some support. Schwab (1991)investigated the relation between perfor-mance and turnover among faculty at a largeuniversity, and found a negative relationbetween performance (measured by numberof publications) and turnover among


nontenured faculty. In contrast, among ten-ured faculty, there was a positive relationbetween performance and turnover.

The negative relation between perfor-mance and turnover among nontenuredfaculty is likely due to the fact that low-performing individuals, knowing they prob-ably will not receive tenure, leave beforethey are denied tenure. Among tenured fac-ulty, those performing at low levels are morelikely to remain with the organizationbecause their jobs are secure, and they arelikely to have relatively few alternatives.High-performing tenured faculty, in con-trast, often have very attractive employmentalternatives so they may be lured away byother universities.

In addition to Schwab’s (1991) study,more direct tests of the curvilinear hypoth-esis have supported this relationship (e.g.,Trevor, Gerhart, & Boudreau, 1997), as didthe previously mentioned meta-analysis con-ducted by Williams and Livingstone (1994).Trevor et al. also found that this curvilinearrelation is more pronounced if salary growthis low and rates of promotion are high. Whensalary growth is low, both low and highperformers have the most to gain by seekingother employment. When rates of promotionare high, low performers are likely to bedissatisfied and look elsewhere. High per-formers who are promoted rapidly are goingto be more marketable in the external labormarket than high performers who are pro-moted more slowly.

A second nonaffective variable that mayimpact turnover is the external labor market.Most people do not leave their present jobuntil they have secured other employment,so turnover should be highest when jobopportunities are plentiful. Steel and Griffeth(1989), for example, performed a meta-analysis and found the corrected correlationbetween perceived employment opportuni-

ties and turnover to be positive but relativelymodest (r ¼ .13). Gerhart (1990), however,found that a more objective index of employ-ment opportunities (regional unemploymentrates) predicted turnover better than percep-tions of employment opportunities. The factthat these findings are at odds suggests thatthe objective state of the external labormarket, and individuals’ perceptions of op-portunities, may operate independently toinfluence turnover decisions.

This issue was addressed in a somewhatdifferent way by Steel (1996) in a sample ofU.S. Air Force personnel. He examined theimpact of objective labor market indexes andperceptions of employment opportunities onreenlistment decisions. The results of thisstudy showed that reenlistment decisionscould be predicted with a combination ofperceptual and objective labor market varia-bles. Turnover was highest among indi-viduals who reported that they had strongregional living preferences and believedthere were a large number of employmentalternatives. The one objective labor marketmeasure that predicted reenlistment was thehistorical retention rate for each Air Forceoccupational specialty in the study. Those inoccupational specialties with high retentionrates were more likely to re-enlist.

Although Steel’s study is quite useful incombining perceptual and objective data, itsgeneralizability may be limited by its use of amilitary sample. In civilian organizations,employees are not bound to a certainnumber of years of service; thus, they mayleave the organization at any time. One mightsurmise that labor market conditions (bothobjective and perceptive) might be moresalient for military personnel because theyhave a window of opportunity; they canchoose between staying and leaving the orga-nization. As with any finding, generalizabil-ity is ultimately an empirical issue. Thus,


these findings must be replicated in a non-military setting.

A final variable—job tenure—maydirectly and indirectly impact turnover. Asdiscussed in the previous chapter, longer jobtenure is associated with higher levels ofcontinuance commitment and, hence, lowerlevels of turnover (Meyer & Allen, 1997). Jobtenure may also have an indirect effectbecause turnover may be influenced by dif-ferent variables at different points in an em-ployee’s job tenure. Dickter, Roznowski, andHarrison (1996) examined both job satisfac-tion and cognitive ability as predictors ofquit rates in a longitudinal study conductedover a period of approximately 4 years. Theirfindings indicated that the impact of jobsatisfaction on turnover is strongest whenemployees have been on the job about 1 year,and this effect gradually decreases overtime. It was also found that a high level ofcognitive ability was associated with de-creased risk of turnover. However, as withjob satisfaction, this relationship diminishedover time.

The results of Dickter et al. (1996) sug-gest that job satisfaction may drive turnoverdecisions early in an employee’s job tenure.However, as an employee builds up job ten-ure, the costs associated with leaving one’semployer become greater. Also, as job tenureincreases, it is likely that a greater number ofnonwork factors will come into play whenone is deciding whether to leave one’s presentemployer. For example, employees with chil-dren in school may not wish to change jobs ifdoing so involves a geographical move.

The fact that cognitive ability has lessimpact on turnover over time is also signifi-cant. Cognitive ability is associated withjob performance (e.g., Schmidt & Hunter,1998), and this supports the notion that therelation between performance and turnoveris nonlinear only among those who have

been employed a relatively short periodof time. Performance may not be related toturnover among longer-tenured employeesfor a number of reasons. For example, thelevel of performance among those who stayin an organization may be restricted, and thismay prevent performance from being relatedto turnover among this group. This essen-tially represents a self-selection effect.

It is also possible that true performancedifferences exist among longer-tenured em-ployees, but other factors are at work. Forexample, when employees have been em-ployed in an organization for several years,managers may be reluctant to highlight theirperformance differences. It is also possiblethat, over time, the experience employeesgain may compensate for what they may belacking in cognitive ability.

An Alternative Turnover Modelof the Turnover Process

As discussed in the previous chapter,Mobley’s (1977) model of the turnoverprocess, and variants of it, have dominatedthe turnover literature for the past 25 years.Although there are some differences amongthese models, they all basically have twothings in common. First, all propose thatemployee affect (mainly job satisfaction)plays a key role in the turnover process. Thatis, a lack of satisfaction or feelings of lowcommitment set in motion the cognitiveprocesses that may eventually lead anemployee to quit his or her job. Second,because of the emphasis on employee affect,an implicit assumption in most turnovermodels is that employee turnover is usuallydue to willingness to get away from thepresent job rather than attraction to otheralternatives.

According to Lee and Mitchell (1994),the dominant process models in the turnover


literature have been useful, but they havealso ignored some basic properties of humandecision-making processes. Based largely onbehavioral decision theory (Beach, 1993),they developed the Unfolding Model of theturnover process. A basic assumption of theUnfolding Model is that people generally donot evaluate their job or job situation unlessforced to do so. Lee and Mitchell refer toevents that force people to evaluate their jobsas ‘‘shocks to the system.’’ Shocks may benegative events (e.g., a major layoff), but theyare not necessarily. A shock is simply anyevent that forces an employee to take stockand review his or her job situation (Holtom,Mitchell, Lee, & Interrieden, 2005). Forexample, a promotion may also be a shockto the system, according to Lee and Mitch-ell’s definition.

Once an employee experiences a shock tothe system, a number of outcomes are pos-sible. One possibility is that the employeemay have a preprogrammed response to theshock, based on previous experience. Forexample, an employee may have previouslyworked in a company that was acquired by acompetitor, and decided it was best to leavethe company. If this same event happensin later years, the employee may not evenhave to think about what to do; he or shemay simply implement a preprogrammedresponse.

Where a preprogrammed response doesnot exist, an employee would engage in con-trolled cognitive processing and consciouslyevaluate whether the shock that has occurredcan be resolved by staying employed in thecurrent organization. To illustrate this point,Lee and Mitchell (1994) provide the exampleof a woman who becomes pregnant unex-pectedly (a shock to the system). Assumingthat this has not happened before, thiswoman would probably not have a pre-programmed response (quit or stay), and

most likely would not have a specific jobalternative. Rather, she would be forced toevaluate her attachments to both the organi-zation and her career. Such an evaluationmay also involve deciding whether continu-ing to work in the organization is consistentwith her image of motherhood.

A third type of situation involves a shockto the system without a preprogrammedresponse, but with the presence of specificjob alternatives. An example of this situationwould be where an employee receives anunsolicited job offer from another organiza-tion. This job offer may be considered ashock to the system because it forces theemployee to think consciously about his orher job situation and to compare it to theoutside job offer. Note that, in this type ofsituation, the employee may be reasonablyhappy in his or her job but may ultimatelyleave because another job is simply better.

A final alternative is where there is noshock to the system but turnover is affectinitiated—that is, over time, an employeemay simply become dissatisfied with his orher job for a variety of reasons. For example,the job may change in ways that are nolonger appealing to the employee. Alterna-tively, the employee may undergo a changein his or her values or preferences, and mayno longer see the job as satisfying. Accordingto Lee and Mitchell (1994), once a personis dissatisfied, this may lead to a sequence ofevents, including reduced organizationalcommitment, more job search activities,greater ease of movement, stronger inten-tions to quit, and a higher probability ofemployee turnover. This proposed sequenceof events is very consistent with dominantaffect-based models of the turnover process(e.g., Mobley, 1977).

Lee and Mitchell’s (1994) UnfoldingModel is relatively new, so it has not receivednearly the empirical scrutiny of more


traditional affect-based process models. How-ever, empirical tests of this model have metwith some success (Lee, Mitchell, Holtom,McDaniel, & Hill, 1999; Lee, Mitchell, Wise,& Fireman, 1996). As with any model, itis likely that further refinements will be madeas more empirical tests are conducted. Never-theless, the Unfolding Model does representan important development in turnoverresearch.

In recent years a major concept that hascome out of work on the Unfolding Model ofturnover is that of embeddedness. Mitchellet al. (2001) define embeddedness as thecombination of forces (in one’s both personaland professional life) that keep a person fromchanging his or her employer. A personwould be highly embedded in a professionalsense, for example, if he or she had a largesocial role in his or her current organization.A high level of embeddedness in one’spersonal life may result from one’s familysituation and perhaps a high level of com-munity involvement. Research has in factshown that embeddedness is a fairly goodpredictor of voluntary turnover, as well asbehaviors (e.g., Lee, Mitchell, Sablynski,Burton, & Holtom, 2004).

Accidents

Accidents represent a very serious and costlyform of counterproductive behavior in orga-nizations. For example, in the United Statesalone, the most recent estimate is that therewere 4.7 million recorded workplace inju-ries in the United States in 2002, and in thissame year 5,524 of these were fatal (U.S.Bureau of Labor Statistics, 2004). Further-more, it has been estimated that accidentscost organizations $145 billion per year (U.S.Department of Health and Human Services(2002)). Another indication of the impor-tance of safety in the workplace is that many

nations have enacted legislation dealing withsafety standards, and many have also createdgovernment agencies to oversee compliancewith these standards. In the United States,for example, the Occupational Safety andHealth Act provides employers with a set oflegal standards regarding safety in the work-place. This legislation also led to the creationof the Occupational Safety and HealthAdministration (OSHA) to enforce employ-ment safety and health standards.

Determinants of Accidents

As one might imagine, accidents are morecommon in certain types of work settingsthan in others. Employees who work in agri-culture and mining are particularly at risk,while those who work in sales or serviceindustries are least likely to be involved inaccidents (U.S. Bureau of Labor Statistics,2004). Work settings, however, do notcompletely explain variation in accidentrates. Because of this, considerable effortover the years has been devoted to gaininga better understanding of other causes ofaccidents, and, perhaps more importantly,how organizations can take proactive stepsto make the workplace safer.

Research on accidents has a long his-tory, although much of it has not beenconducted by organizational psychologists.For example, industrial engineers havefocused on the design of machinery andthe physical layout of the workplace aspossible causes of workplace accidents(Wickens & Hollands, 2000). Within psy-chology, early accident research focusedlargely on developing a profile of ‘‘theaccident-prone employee.’’ This researchidentified a number of characteristics thatwere occasionally correlated with accidents,but researchers were never able to consis-tently document a cluster of characteristics


that were consistently associated with acci-dent prevalence (see Hansen, 1988). This isnot surprising, given that much of thisresearch was largely devoid of any theoret-ical underpinnings.

Research examining personal character-istics associated with accidents has becomemore theoretically grounded over the years,and, in fact, has yielded some useful results.For example, Hansen (1989) studied predic-tors of accident frequency among employeesof a large petrochemical processing companyand found that accidents were most frequentamong employees who were younger, highlydistractible, and who scored high on a meas-ure of general social maladjustment. Morerecently, Frone (2003) found similar person-ality traits to predict accidents among high-school-age employees, so there does appearto be some consistency in characteristics thatlead to accidents.

In recent years, there has been a notice-able shift in accident research, from in-vestigating characteristics of individualemployees, to characteristics of group andorganizational climates. According to Griffinand Neal (2000), safety climate consists ofemployees’ perceptions of the policies, pro-cedures, and practices relating to safety in anorganization. Others (e.g., Zohar, 2003)view safety climate as more of an aspect ofan organization’s culture with regard tosafety. Regardless of the definition, two fun-damental questions are at the core of safetyclimate: Is employee safety considered ahigh-level organizational priority? And doesthis get communicated to employees throughformal organizational policies and manage-rial actions?

In the past 10 years a considerableamount of research has been done onthe relationship between safety climate andboth accident frequency as well as behaviorsassociated with safety. Clarke (2006) recent-

ly summarized much of this literature in ameta-analysis of 32 studies and found thata positive safety climate was associatedwith lower accident involvement, compli-ance with safety procedures, and participa-tion in more proactive safety behaviors. Thestrongest of these relationships was betweenpositive safety climate and participation inproactive safety behaviors.

This recent shift in emphasis towardsafety climate is important for a number ofreasons. At a general level, it offers a produc-tive departure from a long history of accidentresearch that has clung rather dogmaticallyto individual characteristics as predictors.This is not to say that individual character-istics have no bearing on accidents; forexample, the results of Hansen’s (1989)and Frone’s work would suggest otherwise.However, given the considerable effort thathas gone into the investigation of individualpredictors, the actual insight gained aboutaccidents and accident prevention has beenrather disappointing.

Safety climate research also represents arecognition that employees work in a socialcontext and that this impacts safety (Neal &Griffin, 2006). Thus, information communi-cated via the social environment may have apowerful impact on employees’ behaviors.Granted, research on safety climate is stillin its infancy and a number of issues are stillto be resolved (e.g., Does safety climate oper-ate equally at the group and organizationallevels? How does safety climate develop inthe first place? Do personal characteristics ofemployees interact with safety climate to im-pact actual safety behavior?). Despite theseunresolved issues, safety climate represents afruitful new approach that may yield consid-erable insight into safety and ultimately pro-vide organizations with concrete guidanceon reducing the incidence of workplaceaccidents.


Accident Prevention

Given the research reviewed, an organizationcan take one of four different approaches tothe prevention of accidents. First, based onhuman factors and industrial engineeringresearch, an organization may choose to focuson physical factors. For example, an effortmight be made to design equipment so it iseasier to use and make other features of thephysical environment safer for employees.This approach can be quite useful, given thatsome accidents can be prevented by betterequipment design. It may also be quite costly,depending on the modifications that may beneeded within the physical environment.

A second approach, and one that is usedfrequently, is behavior modification (see Chap-ter 9) that encourages employees to use safework practices and discourages employeesfrom being unsafe. This involves the use ofreinforcements for safe behaviors and the useof sanctions or punishment for unsafe behav-iors. An organization, for example, mightoffer cash bonuses to employees who havethe best safety records in a particular year.This is an approach that has been used suc-cessfully in a number of organizations (e.g.,Komaki, Barwick, & Scott, 1978; Komaki,Heinzmann, & Lawson, 1980). On the neg-ative side, organizations may take discipli-nary actions (e.g., written reprimands,suspensions) against employees who engagein unsafe work practices or who consistentlyhave poor safety records.

A third approach is to use selection as ameans of screening out employees who arelikely to be unsafe. If unsafe behavior isviewed as part of a general pattern of deviantantisocial behavior, then organizations mayhave a number of useful predictors at theirdisposal. For example, based on Hansen’s(1989) study, described earlier, general socialmaladjustment and distractibility would

appear to be two predictors that organizationscould use to screen out employees who mayhave poor safety records. On the positive side,organizations may consider the use of person-ality traits such as conscientiousness in selec-tion as a positive step toward improving safety(Barrick & Mount, 1991; Tett, Jackson, &Rothstein, 1991).

A final method of preventing accidents isby changing or improving the safety climateof the organization. Some possible ways ofdoing this might be publicizing the impor-tance of safety within the organization andmaking supervisors and managers account-able for the safety records within their units(Hofmann & Morgeson, 1999; Zohar, 2000).Over time, as more research on the safetyclimate construct is conducted, organiza-tions will likely be provided more guidancein their efforts to improve safety climate.

LESS COMMON FORMSOF COUNTERPRODUCTIVEBEHAVIOR

Up to this point in the chapter, we havecovered the most common forms of counter-productive behavior in organizations. Mostorganizations must deal frequently with inef-fective employee performance, absenteeism,turnover, and employee safety issues. These,however, are clearly not the only forms ofcounterproductive behavior in organiza-tions. In this concluding section, we exploreless common forms of counterproductivebehavior in organizations. These include:employee theft, workplace violence and mis-treatment, substance use, and sexual harass-ment.

Employee Theft

Employee theft may be defined simply as em-ployees taking things from the organization


that don’t belong to them. Based on thisdefinition, theft could range from relativelyminor acts, such as employees taking inex-pensive office supplies, to more seriousforms, such as a government employee’stheft of classified documents. Most of theliterature on employee theft has focused onwhat could be described as moderate forms ofemployee theft: retail store employees steal-ing merchandise, or convenience store em-ployees skimming money from the cashregister.

Research on the frequency of employeetheft has shown that it does occur withenough frequency to be problematic for manyorganizations. For example, it has been esti-mated that approximately 35% of employeessteal from their employers, and the financiallosses from theft are in the billions (Kuhn,1988). In a more recent estimate, it was foundthat 52% of people reported taking propertyfrom work without permission, and 25%reported falsifying receipts to get reimburse-ments for money they didn’t spend (Bennett& Robinson, 2000). Unfortunately, becausethe costs of employee theft are often passed onto consumers, the impact of this behaviorreaches far beyond the organizations in whichit occurs.

A review of the literature on the causes ofemployee theft reveals two clear themes. Thefirst, and by far the strongest, is that theft isdue largely to characteristics of the individ-ual (e.g., Collins & Schmidt, 1993; Jones &Boye, 1992; Ones et al., 1993). Furthermore,publishers of integrity tests have conductedmuch of this research. This is potentiallyproblematic because such organizationsmay lack the motivation to rigorously eval-uate the predictive capabilities of theirproducts. Despite these concerns, Ones etal.’s (1993) meta-analysis showed fairlyclearly that integrity tests do in fact predictemployees’ theft. Because most integrity tests

measure the personality trait of conscien-tiousness, this suggests that the employeesmost likely to steal have a low level of thistrait; in other words, they are unreliable,lack self-discipline, and have a disregardfor rules and authority. Other than consci-entiousness, it has also been shown thattheft tends to be higher among employeeswho have very tolerant attitudes toward theftand other forms of dishonesty (Jones &Boye, 1992).

A second theme in the literature is thattheft is impacted by conditions in the envi-ronment such as unfair or frustrating or-ganizational conditions. Greenberg (1990),for example, conducted a study in which apay-reduction policy was implemented intwo separate locations of a large manufactur-ing organization. In one of these locations,little explanation was provided as to why thepolicy was being implemented, and thisexplanation was given with little remorse orsensitivity. In the other location, however,management provided employees with amore extensive explanation as to why thepolicy had to be adopted, and did so withmuch greater sensitivity. This study showedthat the rate of theft in the plant where theinadequate explanation was provided wassignificantly higher compared to the plantgiven the adequate explanation and a thirdplant where no pay reduction had beenimplemented.

According to Spector (1997b), employeetheft is also caused by organizationalconditions that induce frustration amongemployees. Frustration is essentially the emo-tion evoked in people when things inthe environment are blocking their goals. Inorganizations, these barriers may includeenvironmental constraints such as poorequipment, unnecessary rules and regula-tions, and other policies that end up wastingemployees’ time. Thus, Spector has proposed


that employees may vent their frustrationstoward the organization through acts of theftand sabotage. As with many relations, thelink between frustration and theft may beinfluenced by other factors. For example,employees who are frustrated may feel likestealing but do not act on such impulseseither because they have no opportunity orthey are afraid of the consequences of suchbehavior. According to Spector, one variablethat may moderate the relation between frus-tration and theft is employees’ locus of control.Locus of control represents beliefs regardingthe control people have over reinforcementsin their lives (Rotter, 1966). A persondescribed as having an internal locus of con-trol generally believes that he or she hascontrol over reinforcements. In contrast, anexternal locus of control is associated withthe belief that one has little control overreinforcements.

The potential moderating effect of locusof control on the relation between frustrationand theft is depicted in Figure 6.2. As can beseen, this model proposes that frustration ismost likely to lead to destructive behaviorssuch as theft among employees who have an

external locus of control. Those with anexternal locus of control tend to respond tofrustration through theft and other forms ofdestructive behavior because they do notbelieve that frustrating organizational condi-tions can be changed through more con-structive means. In contrast, those with aninternal locus of control are more likely tobelieve that they are able to change frustrat-ing organizational conditions constructively.These individuals, for example, may chooseto exert their influence through participativemanagement practices or labor-managementcommittees. Spector’s (1997b) hypothesishas received some empirical scrutiny andin general has been supported (Chen &Spector, 1992; Spector & O’Connell, 1994;Storms & Spector, 1987).

Workplace Violenceand Mistreatment

Like employee theft, workplace violence is arelatively infrequent event. However, inrecent years, there has been an alarmingincrease in the number of violent incidentsin the workplace. For example, it hasrecently been estimated that in the UnitedStates nearly two million people may expe-rience physical attacks in the workplace eachyear (Barling, 1996). Even more alarming isthe fact that homicides are the second lead-ing cause of job-related deaths (U.S. Bureauof Labor Statistics, 1999). In addition toovert violence, many employees experienceother forms of mistreatment in the work-place such as rudeness, verbal abuse, andsocial exclusion. Both workplace violenceand more general mistreatment are coveredin the following.

Workplace Violence and GeneralizedMistreatment. Like other phenomenathat either have been or will be covered in

FIGURE 6.2Spector’s Model of the Impact of Locus of Controlon the Relationship between Frustration andCounterproductive Behavior

ConstraintsFeelings ofFrustration

DestructiveBehavior

LowControl

HighControl

ConstructiveBehavior

Source: P. E. Spector. (2000). Industrial and organizational

psychology: Research and practice (2nd ed.). New York: John

Wiley & Sons, Inc. Reprinted by permission.


this book, what is considered workplaceviolence is quite broad. For the purposes ofthis chapter, workplace violence is definedas physical acts of aggression by membersof an organization, carried out in organiza-tional settings. Notice that no attempt ismade to specify or restrict the target of theaggression. For example, a violent act couldbe directed at (or instigated by) a fellowemployee, one’s supervisor, or even a cus-tomer. Since most research on workplaceviolence and aggression has focused on theviolent acts of employees toward other em-ployees (see Schat & Kelloway, 2005 for areview), the focus in this section will beon violent acts committed by employeestoward other employees. However, it hasbeen shown that clients and customers alsocommit many of the violent and antisocialacts against employees in organizations(e.g., Bussing & Hoge, 2004).

To explain violent acts on the part ofemployees, there have generally been threefoci in the literature: (1) the physical environ-ment; (2) characteristics of the individual; and(3) the organizational environment. If thefocus is on the physical environment, we areable to draw on the social-psychological liter-ature that has linked aggression to violentcues in the environment, as well as factorsthat induce frustration (Worchel, Cooper,Goethals, & Olson, 2000). Considerableresearch has also linked stress-related symp-toms to monotonous machine-paced work(e.g., Broadbent, 1985). Unfortunately, nei-ther of these characteristics has been linkedto workplace violence.

Given that little empirical research hasexamined the link between the physical envi-ronment and workplace violence, we canonly speculate that environment may play arole. However, it is interesting to note thatsome of the most highly publicized acts ofviolence on the part of employees have taken

place in work environments that manywould consider somewhat noxious. In theU.S. Postal Service, for example, much of thework is highly monotonous and paced bythe speed of machines. Factories and othermanufacturing facilities are often noisy andhot. This link is obviously pure speculation,but over time, as more data are collectedabout violent incidents, it may be possibleto assess more clearly the contribution of thephysical environment.

A second focus in the workplace violenceliterature is identification of the character-istics of those who may be predisposed toviolent acts. Given that workplace violence isa low base-rate event, this is obviously not aneasy task. Nevertheless, research has showthat certain personal characteristics may indi-cate heightened risk of violence. Day andCatano (2006) summarized this literatureand concluded that the primary predictorsof violent behavior tend to be past history ofviolent behavior, alcohol and drug abuse,lack of conscientiousness, low agreeable-ness, and low emotional stability. There hasalso been evidence that individuals who arehigh on hostility may be predisposed toaggressive behavior in the workplace (Judge,Scott, & Ilies, 2006).

A third focus of workplace violenceresearch has viewed the organizational envi-ronment as a possible factor precipitatingviolent acts. Much of what can be said heremirrors the previous section on theft. Orga-nizations that treat employees unfairly andignore their frustrations may be at greaterrisk for violence than organizations thatemphasize fairness and support (Greenberg,1990; Spector, 1997b). It has also beenshown recently that leaders who are verypassive and unwilling to take action mayallow antisocial and aggressive behaviors tooccur in their work groups (Skogstad,Einarsen, Torsheim, Aasland, & Hetland,


2007). These could potentially escalate thelevel of physical violence.

In considering the research on the orga-nizational environment, it must be remem-bered that, even in the worst organizationalenvironments, very few employees engage inacts of violence. Thus, a negative organiza-tional environment will probably not have astrong main effect on the incidence of work-place violence. Rather, the best way to thinkof the contribution of the organizationalenvironment is in combination with per-sonal factors. For example, a person who isprone to aggression and violence may takeout his or her frustrations through violentacts if treated in an unfair or arbitrary man-ner by the organization.

Generalized Mistreatment. While work-place violence is clearly an important formof counterproductive behavior in organiza-tions, we also know that physical violencedoes not occur frequently in organizations.On the other hand, it is far more commonfor employees to experience rudeness, ver-bal attacks, invasions of their privacy, andin some cases even malicious attempts tosabotage their performance.

In organizational psychology suchbehaviors have received a variety of labels,including workplace incivility, bullying,mobbing, social undermining, and work-place deviance (see Bowling & Beehr,2006). While we do not view all of these asbeing the same, we also believe that these allhave a great deal in common and thereforediscuss them collectively under the label ofgeneralized mistreatment.

What do we know about generalizedmistreatment in organizations? Bowling andBeehr (2006) conducted a meta-analysis onconsequences of many forms of generalizedworkplace mistreatment that included90 samples. One clear finding from this

study is that people react to mistreatmentin a variety of negative ways, includingincreased depression and anxiety anddecreased job satisfaction. The findings alsoindicated that individuals who experiencehigh levels of mistreatment report higherlevels of physical symptoms and may beabsent from work more frequently.

While the effects of mistreatment arepretty clear, less research has addressed itscauses. Nevertheless, it has been shown thatleaders or supervisors may play a key rolein whether this type of behavior occurs(Skogstad et al., 2007). Individuals who arehostile, feel they are treated unfairly by theirorganization, are dissatisfied, and who feelemotionally exhausted may be more likely tolash out at others (Blau & Andersson, 2005;Judge et al., 2006). These findings, thoughlimited in scope, suggest that the way orga-nizations treat employees is clearly a majorcontributing factor. Of course it has alsobeen suggested (see Andersson & Pearson,1999) that mistreatment in the workplace ismerely a symptom of a general trend towardlower levels of civility and decorum on soci-ety in general (Comment 6.2).

Substance Use

According to Frone (2006), approximately14% of the U.S. work force (17.7 millionworkers) reports some use of illicit drugs,and approximately 3% (3.9 million) reportusing on the job. Use of alcohol, which isobviously legal, is undoubtedly muchhigher. These numbers represent a seriousconcern for organizations because employ-ees who use alcohol and illicit drugs may beabsent from work more frequently and maybe more likely than nonusers to engage in avariety of other counterproductive behaviors(Frone, 2004). We also know that substanceuse is related, either directly or indirectly, to


negative outcomes such as traffic fatalities,domestic abuse, and violent crimes.Research examining the impact of substanceuse in organizations has produced somefairly consistent findings. For example, ithas been shown that employees who areproblem drinkers and users of illicit drugsmay exhibit a number of negative outcomes,such as performance decrements, increasedabsenteeism, greater frequency of accidents,greater job withdrawal, and more antagonis-tic behavior toward others (Frone, 2004;Roman & Blum, 1995). Given these find-ings, the more pressing issues appear to be(1) identifying those who may have sub-stance use problems, and (2) deciding whatto do when employees show signs of sub-stance use problems.

The prediction of substance use has beenaddressed in two basic ways. As with theftand violence, substance use is seen by manyas part of a more general pattern of antisocialbehavior (e.g., Hogan & Hogan, 1989).Given this conceptualization, efforts havebeen made to predict substance use basedon personality traits more generally associ-ated with antisocial behavior. McMullen(1991), for example, found that the reliabil-ity scale from the Hogan Personality Inven-tory (HPI; Hogan & Hogan, 1989) wasnegatively related to self-reports of bothoff- and on-the-job substance use amongcollege students. Interestingly, in this samestudy, in an applicant sample, this scale dis-tinguished those who passed and those whofailed a urinalysis drug screening.

COMMENT 6.2

INCIVILITY ON TELEVISION

IN THEIR 1999 review, Andersson and Pearson

(1999) both defined what is meant by work-

place incivility and developed a model to

explain why it occurs and how it might esca-late into more serious forms of aggression.

Essentially, workplace incivility was defined as

‘‘low intensity deviant behavior with ambigu-

ous intent to harm the target, in violation of

workplace norms for mutual respect. Uncivil

behaviors are characteristically rude, discour-

teous, displaying a lack of respect for others.’’

(p. 1)While there are a multitude of factors

within the workplace that may lead to incivil-

ity, Andersson and Pearson also suggest that

this is a reflection of a more general societal

trend. Nowhere is this more apparent than on

television. In the highly popular show ‘‘Amer-

ican Idol,’’ for example, contestants are not

only eliminated from the competition, but theyare also frequently humiliated and made fun of

by the judges. Other reality TV shows often

highlight arguments and even physical con-

frontations among the contestants.

Why has there been such a rise in the

phenomenon of ‘‘incivility as entertainment’’?The most obvious answer is that people seem to

enjoy it. Television programming is driven

largely by viewer ratings, so as long as people

watch programs that highlight uncivil behavior

networks will continue to put them on. Per-

haps out of concern for the effect of such

programs on children (or simply because these

programs will get boring), people will commu-nicate to television networks that they want to

see examples of more civil behavior on tele-

vision. Hopefully this will translate into more

civil behavior in society in general, and the

workplace more specifically.

Source: Andersson, L., Pearson, C. (1999). Tit for tat? The

spiraling effect of incivility in the workplace. Academy of

Management Review, 24, 452–471.


Other than personality, research has alsoinvestigated personal characteristic predic-tors in the form of personal history. Lehman,Farabee, Holcom, and Simpson (1995)investigated a number of personal back-ground characteristics as predictors of sub-stance use among a sample of municipalworkers, and produced a number of mean-ingful findings. Those at the greatest risk forsubstance use were young males who repor-ted low self-esteem, had a previous arresthistory, came from a family with substanceuse problems, and tended to associate withsubstance-using peers.

Another line of inquiry has examinedenvironmental predictors of substance use.In this line of research, the variable thathas been examined most is stressful jobconditions. For the most part, this researchhas shown that although holding a stressfuljob may increase one’s risk of substanceuse, this effect does not appear to be large(e.g., Cooper, Russell, & Frone, 1990). Amore recent line of inquiry has examinedthe social norms surrounding substance usein organizations. Recall that this idea hasalso been explored, with some success, inthe study of both absenteeism and acci-dents. An example of this type of researchcan be seen in a study by Bennett andLehman (1998), in which the impact of aworkplace drinking climate was measured.It was found that in groups where a drink-ing climate was positive, individualsreported higher levels of both their ownand coworkers’ drinking activity. Thesefindings suggest that social factors withinwork groups, and perhaps even withinprofessions, may contribute to problemdrinking.

More recent findings seem to be consis-tent with the idea that the social environmentplays a key role in substance use. Frone(2006), for example, found that the preva-

lence of substance use differed considerablyby occupation, which suggests that somework environments or occupational culturesdiffer with regard to norms surroundingsubstance use; this could also reflect differ-ences in availability. This was also shownin other research on substance use amonghigh-school-aged employees as well (Frone,2003).

Based on the empirical research, whatcan organizations do to prevent substanceuse among employees? As with preventionof theft and violence, research on substanceuse suggests that organizations should focuson both screening out potential substanceusers and creating a social environment thatdoes not promote substance use. Whenscreening out potential employees, employersmust take into account the fact that recoveringalcoholics are protected under the AmericansWith Disabilities Act of 1990 and cannot bediscriminated against. Given the multitude ofcounterproductive behaviors that are associ-ated with low conscientiousness (e.g., Hogan& Hogan, 1989; Ones et al., 1993), assessingthis trait would appear to have some merit. Athorough preemployment background checkwould also seem to be a logical step towardpreventing substance use problems (Lehmanet al., 1995). As stated earlier, organizationsobviously must make sure that such checksdo not violate the rights of applicants.

Another method of preventing substanceuse has become increasingly popular: requir-ing applicants, and even current employees,to submit to drug screening, most typicallythrough urinalysis. Drug screening is bothexpensive and controversial (Rosen, 1987),so organizations must think very carefullyabout its use. Nevertheless, research hasshown that people are not strongly opposedto the use of preemployment drug testingfor jobs in which the safety of others couldbe put at risk by a drug-using employee


(Murphy, Thornton, & Reynolds, 1990), buthave less favorable attitudes in jobs withoutthese characteristics. Research has alsoshown that attitudes toward drug-screeningprograms are more positive when suchprograms are seen as procedurally fair(Konovsky & Cropanzano, 1991). Importantprocedural issues in drug testing include thebasis on which employees or applicants arerequired to submit to such tests, as well aswhether retesting is allowed.

Perhaps the most critical issue surround-ing drug testing is an organization’s responseto confirmed employee substance use. Anorganization essentially has two choices indeciding how to respond to such employees:punishment or treatment. Some organiza-tions have what could be described as zerotolerance policies with respect to drug use. Inthe military, for example, evidence of illicitdrug use will automatically disqualify arecruit and will result in immediate discipli-nary action against active duty personnel. Inother cases, when substance use problemsamong employees are discovered, organiza-

tions seek to provide these individuals withtreatment—typically, through EmployeeAssistance Programs (EAPs) and referrals.This is obviously a difficult issue. A recentexample of this is the difficulty major leaguebaseball has had in deciding how to respondwhen players are caught using anabolic ste-roids (see Comment 6.3).

Although cogent arguments can be madefor either approach, research suggests thatdrug testing is viewed more favorably if thoseidentified as having substance use problemsare provided with at least some form of treat-ment (Stone & Kotch, 1989). The provisionof treatment makes a drug-testing programappear to have a greater level of fairnesscompared to programs that have only punitiveoutcomes. A possible downside to treatmentis that an organization may run the risk ofconveying an overly tolerant attitude towardsubstance use. In dealing with substance use,an organization is best served by pursuing apolicy that combines clearly stated conse-quences with compassionate options thatassist with treatment and recovery.

COMMENT 6.3

JUICING THE GAME: BASEBALL AND STERIODS

WHEN MOST PEOPLE think about substance use

in the workplace, they think about alcohol

and illicit drugs such as marijuana. In major

league baseball, however, the primary sub-stance use issue in recent years has been play-

ers’ use of anabolic steroids. In his 2005 book

Juicing the Game: Drugs, Power, and the Fight for

the Soul of Major League Baseball, journalist

Howard Bryant provides an informative and

insightful account of this issue. As Bryant

explains, steroid use presented major league

baseball with a very difficult dilemma. Despitethe well-known dangers of steroid use, it is

also true that offensive production (e.g.,

home-runs) in baseball increased consider-

ably when steroid use was the heaviest.

Increases in offensive production, of course,

translate into greater attendance and reve-nues. Given the increased publicity surround-

ing anabolic steroids in the past 5 years, major

league baseball has instituted mandatory test-

ing for anabolic steroids, and increased the

penalties for those caught.

Source: Bryant, H. (2005). Juicing the game: Drugs, power,

and the fight for the soul of major league baseball. New York:

Viking.


Sexual Harassment

Sexual harassment has become a highly visi-ble issue in organizations ranging from cor-porations to universities. Sexual harassment isdefined as ‘‘unwelcome sexual advances,requests for sexual favors, and other verbalor physical contact when (a) submission tothe conduct is either explicitly or implicitly aterm or condition of an individual’s employ-ment, (b) submission to or rejection of suchconduct by an individual is used as a basisfor employment decisions affecting that indi-vidual, and/or (c) such conduct [that] has thepurpose or effect of unreasonably interferingwith work performance, or creating anintimidating, hostile, or offensive workingenvironment’’ (Equal Employment Opportu-nity Commission, 1980). The term quid proquo sexual harassment is often used to denotesituations in which an employee’s advance-ment or performance is adversely impactedby refusing the sexual advances of a super-visor or other employee who exerts powerover the employee. This form would applyprimarily to the first two parts of the defini-tion provided previously.

The second form of sexual harassment,often referred to as hostile work environment,refers primarily to the third part of the def-inition. In this form, there is no overt attemptto manipulate or threaten. Rather, the exis-tence of sexual harassment is based on thegeneral behavior of others in the workplace.Vulgar comments, telling ‘‘off-color’’ jokes,the display of pornographic images, andeven nonverbal gestures that elicit discom-fort may provide the basis for sexual harass-ment based on the hostile work environmentargument. This category is important becauseit highlights the fact that even behaviorintended to be for fun can be perceived asoffensive to others. Destructive intent is not aprerequisite for sexual harassment.

Given the variety of behaviors that mayconstitute sexual harassment, estimating itsprevalence is a challenge. In a recent reviewof the sexual harassment literature, Bates,Bowes-Sperry, and O’Leary-Kelly (2005)report that, in 2002, the number of formalsexual harassment charges handled by theU.S. Equal Opportunity Commission was14,396, which was a 37% increase from1992. Interestingly, these authors also reportthat some studies have shown that nearly75% of women surveyed report unwantedsexual attention, and nearly 50% have expe-rienced gender-related harassment. Thus, alogical conclusion one may draw is that veryblatant sexual harassment is not a commonoccurrence in organizations. On the otherhand, more subtle forms of sexual harass-ment occur quite frequently.

Organizational research on sexual harass-ment has examined a number of issues,including prevalence (Fitzgerald, Drasgow,Hulin, Gelfand, & Magley, 1997), causes(Gruber, 1998; Gutek, Cohen, & Konrad,1990), ways to respond to sexual harassmentallegations, and methods of prevention(Fitzgerald, 1993). Consistent with govern-ment estimates, this research suggests thatsexual harassment is quite prevalent, it ismuch more likely to be experienced bywomen than men, and that victims are oftenin positions of unequal power and heightenedvisibility in relation to the perpetrators (e.g.,women who work in largely male groups).

More recent research on sexual harass-ment suggests that sexual harassment is partof a more general pattern of mistreatmentand harassment in organizational settings.For example, Lim and Cortina (2005) foundthat sexual harassment and more generalincivility in the workplace tend to be posi-tively associated. It has also been shown thatminority women tend to experience higherlevels of sexual harassment than do white


women (Berdahl & Moore, 2006). Thesefindings suggest that organizations wishingto reduce sexual harassment need to focusnot only on sexual harassment, but also onthe more general social climate of the orga-nization.

Probably the most effective way fororganizations to prevent sexual harassmentis to have in place a clearly articulated sex-ual harassment policy (Bates, Bower-Sperry, & O’Leary-Kelly, 2005). Such apolicy serves the dual purposes of lettingemployees know what is considered sexualharassment, and the steps an organizationwill take if harassment occurs. Letting em-ployees know what is considered sexualharassment is often easier said than done.Given the wording of sexual harassmentstatutes, employees may often be confusedas to what is and what is not sexual harass-ment. However, based on the authors’ expe-rience, getting people to agree on what isappropriate and inappropriate behavior inmixed-gender company may not be nearlyas difficult as it may seem. Given commonsense and knowledge of the prevailing soci-etal codes of morality, the vast majority ofadults know what is and what is not properbehavior in mixed-gender company. Igno-rance is not a viable defense against chargesof sexual harassment.

Sexual harassment policies also need tocommunicate toemployees that sexualharass-ment is a serious matter, and that those whoengage in such behavior will encounter severeconsequences. Ultimately, however, the mostpowerful way to communicate an organiza-tional sexual harassment policy is through anorganization’s response to such behavior. Iforganizations respond to such behavior in amanner that is consistent with their policy,and do so regardless of the parties involved,this sends the powerful message that the orga-nization will not tolerate such behavior.

CHAPTER SUMMARY

This chapter examined counterproductivebehaviors, or those actions on the partof employees that explicitly run counter tothe goals of an organization. The most com-mon form of counterproductive behavioris ineffective job performance. Nevertheless,ineffective performance is often difficultto detect due to external constraints on per-formance and deficiencies in organizationalperformance measurement systems. Basedon models of job performance, ineffectiveperformance may be due to characteristicsof the employee as well as environmentalfactors. Organizations may respond to poorperformance in a number of ways, includingtraining, coaching, and, if all else fails, pun-ishment. A key issue in deciding the responseto poor performance is the underlying causesof performance difficulties.

Absenteeism and turnover are the othertwo most common forms of counterproduc-tive behavior in organizations. Absenteeismhas long been viewed by organizational psy-chologists as a behavioral response to nega-tive feelings about one’s job or job situation.Over time, however, this somewhat narrowview has given way to a broader view of thecauses of absenteeism. The most promisingof these appears to be group norms regard-ing absenteeism. This is due largely to therecognition that absenteeism is a complexphenomenon and thus may be impacted bya variety of factors.

Like absenteeism, turnover has beenviewed largely as a response to negativeaffect. Here too, more contemporary turn-over research has expanded and investigatedother nonaffective predictors of turnover.The external labor market, as well as employ-ees’ job performance, are two nonaffectivevariables that have been shown to have animportant impact on employee turnover.

Chapter Summary 195

Another important advance in this area is theuse of findings from behavioral decisiontheory to model the turnover process. Theconcept of embeddedness, which has grownout of this research, appears to be a promis-ing predictor of turnover.

Less common forms of counterproductivebehavior examined in the chapter includedaccidents, theft, violence and mistreatment,substance use, and sexual harassment. Manyyears of research have failed to uncover a clearprofile of the ‘‘accident-prone’’ employee, butmore recent research in this area has providedsome important insights. The ‘‘safety climate’’within an organization, in particular, appearsto be an important predictor of accident fre-quency. Attention to this climate, coupledwith a focus on the physical environmentand characteristics of employees, is likely tobe the best strategy for preventing accidents inorganizations.

Theft and violence, when considered to-gether, can be considered antisocial behaviorsin organizations. Although both are relativelylow-frequency events, they can neverthelessbe quite damaging to organizations. Lower-level forms of mistreatment, such as rudenessor verbal abuse, occur with much greater fre-quency. Like most forms of behavior, all ofthese forms of counterproductive behaviorcan be explained by characteristics of boththe employee and the environment. Withrespect to theft, considerable evidence hasaccumulated suggesting that employees witha combination of a low level of conscientious-ness and tolerant attitudes toward theft aremost likely to steal. Research has been muchless conclusive about personal characteristicsindicative of violence and mistreatment, al-though it is likely that violence is often indica-tive of underlying psychopathology.

With respect to environmental character-istics, there is some evidence that treating em-

ployees unfairly, and failing to addressfrustrations, may heighten the risk of all ofthese antisocial behaviors. This is particu-larly the case when employees believe theyhave no control over events that impactthem. Thus, organizations wishing to pre-vent antisocial behavior should combinethorough preemployment screening withefforts to treat employees fairly and removebarriers to performance.

Substance use is a form of counterproduc-tive behavior that may be quite damaging,particularly when employees perform dan-gerous work or are entrusted with the safetyof others. The causes of substance use arecomplex; however, it is interesting to notethat personality traits predictive of otherforms of antisocial behavior are also predic-tive of substance use. Prediction and preven-tion of substance use often pose a dilemma fororganizations because issues of employee pri-vacy and public relations are involved.

The final form of counterproductive be-havior examined in this chapter was sexualharassment. Sexual harassment may occur inthe form of direct acts, or more indirectlythrough behaviors that, in the aggregate, cre-ate a ‘‘hostile work environment.’’ Researchhas shown that women are typically the vic-tims of sexual harassment, and it is most likelyto occur in work situations in which womenare in the minority and fill positions of lowerpower than men. The best way to preventthis form of counterproductive behavior isto have in place a clearly articulated sexualharassment policy, and to heighten employ-ees’ awareness of the issue. More recentresearch also suggests that sexual harassmentmay be part of a more general pattern of mis-treatment; thus, organizations may also de-crease sexual harassment by promoting asocial climate of respect and civility towardsothers.



Berdahl, J. L., & Moore, C. (2006). Work-place harassment: Double jeopardy for

minority women. Journal of AppliedPsychology, 91, 426--436.

Bowling, N. A., & Beehr, T. A. (2006).Workplace harassment from the victim’sperspective: A theoretical model and meta-


MICHELLE DUFFY AND THE DARK SIDE OF WORKPLACE RELATIONS

I first became interested in social undermining

after conducting interviews (as part of my job

at the time) with a number of employees in

various occupations who were experiencing amultitude of physical health symptoms such

as chronic colds, flues, headaches, stomach

pain, and sleeplessness. One thing that struck

me was how frequently the direction of the

interview turned to a focus on interpersonal

workplace relationships. The power of these

workplace relationships and the meaning they

had in peoples’ lives stayed with me. Althoughpeople would talk about positive aspects of

their relationships and workplace interac-

tions, often they seemed to strictly focus on

the negative aspects of these interpersonal

interactions. Across different occupations

and ranks, I heard stories that were quite

similar regarding the power of negative work-

place interpersonal interactions in peoples’

work lives. People described interactions in

which they were ridiculed, insulted, ignored,delayed, hampered, and misled by their own

coworkers and supervisors. The power of

these undermining behaviors was destructive

and seemingly toxic.

At that time, muchof the research focuswas

on positive and supportive workplace interac-

tions. There was much less known about the

impact of negative interactions on employeehealth and well-being. I decided to study these

typesof interactions for mydissertation. I found

that undermining behaviors were indeed asso-

ciated with a variety of undesirable employee

outcomes such as lower physical and emotional

well-being, lower performance and poor job

attitudes. More interesting to me, however,

was the fact that undermining behaviors weremore predictive of employee outcomes than

were supportive behaviors. In terms of how

they affect peoples’ lives, a negative workplace

event carries much more weight than a positive

event. Beyond this, I also found people have the

most difficult time dealing with coworkers who

frequently engaged in both positive andnegative

behaviors (rather than just negative behavior).With these people, you never know where you

stand and the inconsistency in their behavior is

very damaging to a person’s well-being.

Michelle Duffy

Carlson School of Management

University of Minnesota


analysis. Journal of Applied Psychology, 91,998--1012.

Clarke, S. (2006). The relationship betweensafety climate and safety performance: Ameta-analytic review. Journal of Occu-pational Health Psychology, 11, 315--327.

Frone, M. R. (2006). Prevalence and distri-bution of illicit drug use in the workforceand in the workplace: Findings andimplications from a U.S. national survey.Journal of Applied Psychology, 91, 857--869.


Chapter Seven

OccupationalStress andEmployee HealthO

ccupational stress is a topic thathas generated a tremendous vol-ume of research (Beehr, 1995; Jex,1998), and a great deal of thatresearch has focused on the phys-

ical and mental health of employees (Hof-man & Tetrick, 2003). Stress is also a topicthat has been the focus of a great deal ofpopular media attention, and it comes upfrequently in everyday conversation. (Whodoesn’t have a stressful job!) Despite all thisattention, the scientific study of occupationalstress does not have a long history. Further-more, despite the considerable progress thathas been made over the years, we still havemuch to learn about the impact of stress onthe health and well-being of employees.

A question that is often asked aboutoccupational stress is: Does it really havean aversive effect on individuals and organi-zations, or are those who study occupationalstress ‘‘making mountains out of molehills’’?There is evidence that being consistentlyexposed to stressful work conditions isharmful to employees and may have a nega-tive impact on organizational effectiveness.Consider, for example, that the amount ofmoney extracted from the U.S. economydue to occupational stress has been esti-mated to be in the billions of dollars (e.g.,Aldred, 1994; Ivancevich & Matteson, 1980;Matteson & Ivancevich, 1987; Mulcahy,1991). Such estimates are based on theassumption that stress plays a role in nega-tive outcomes such as increased healthcarecosts, higher rates of absenteeism and turn-over, more on-the-job accidents, andreduced productivity.

Another indication of the harmful effectsof occupational stress is the increasing trendtoward stress-related workers’ compensationclaims (National Council on Compen-sation Insurance, 1988, 1991; DeFrank &Ivancevich, 1998). In the past, compensa-tion for work-related injuries was limited tophysical injuries caused by some physicalevent or stimulus. Increasingly, however,more and more states are recognizing thelegitimacy of physical and even psychologi-cal injuries that may be caused by somestressful aspect of the work environment that

199

is not physical in nature (e.g., an overlydemanding supervisor).

Occupational stress is also importantbecause of its impact on society as a whole.It is unlikely that a person experiencing con-stant stress on the job will function effec-tively in his or her other roles, such ashusband/wife, parent, neighbor, and com-munity member. Failure to perform theseroles effectively may not have direct eco-nomic costs but may, in the long run, havea tremendously negative impact on society.Thus, occupational stress is clearly not the‘‘cause of all societal ills,’’ but it does have animportant and real impact on individuals,organizations, and society.

A BRIEF HISTORY

The earliest scientific investigations relatedto the field of occupational stress were con-ducted by the well-known physiologist Wal-ter Cannon in the early part of the twentiethcentury (e.g., Cannon, 1914). Cannon’swork focused on the relationship betweenemotions and physiological responses, and isconsidered the earliest work in the field ofpsychosomatic medicine (the relationshipbetween psychological states and physicalillness). Cannon, however, is best knownfor having coined the term homeostasis.Homeostasis represents the body’s effort torestore normal physiological functioningwhen some deviation is required. When thebody is exposed to extreme cold, for exam-ple, the physiological changes that areevoked are designed to maintain a constantinternal body temperature. Stressful condi-tions on the job are typically perceived asaversive events that require some adaptiveresponse designed to return the employee tonormal functioning.

The first actual scientific investigations ofstress are attributed to Hans Selye (1956),

who is considered by many to be the ‘‘Fatherof Stress.’’ Selye, an endocrinologist, wasconducting research on reproductive hor-mones in animals. During the course of thisresearch, he was required to expose theseanimals to a number of aversive stimuli, suchas temperature extremes and radiation.While doing this, Selye observed a great dealof predictability in these animals’ efforts toadapt to the aversive stimuli. From this, Selyereasoned that humans do much the same intheir efforts to cope with the challenges ofeveryday life, and he developed the generaladaptation syndrome to describe this process.

The general adaptation syndrome con-sists of three distinct stages: alarm, resist-ance, and exhaustion. In the alarm stage,the physiological resources of the body aremobilized, in wholesale fashion, to deal withan impending threat. In the resistance stage,the body recognizes that not all of its re-sources may be needed, and thus continuesto mobilize only those that are necessary.Finally, in the exhaustion stage, the bodyrealizes that its physiological resources aredepleted and, as a result, makes anotherattempt to mobilize. If this second attemptat mobilizing physiological resources doesnot neutralize the threat, this may lead topermanent damage to the organism, or whatSelye termed diseases of adaptation. While thegeneral adaptation syndrome tends to beaccepted as a general framework by stressresearchers, it has not been submitted toextensive empirical test in humans. Further-more, it has recently been argued that thisresponse is more appropriate for males thanit is for females (Taylor et al., 2000).

Selye’s work was undoubtedly pioneer-ing, but it must be remembered that hefocused primarily on physiological reactionsto aversive physical stimuli. He was notfocusing specifically on stress in the work-place. The first large-scale program of

200 Occupational Stress and Employee Health

research focusing specifically on stress in theworkplace was undertaken at the Universityof Michigan’s Institute for Social Research inthe early 1960s. What is particularly note-worthy about this research effort is that thefocus was on psychosocial factors in the work-place that may be stressful to employees.Psychosocial factors represent aspects ofthe work environment having to do withinteractions with other people. In particular,the Michigan researchers focused much oftheir attention on what they termed role stress-ors (to be discussed in more detail later inthe chapter), which are aversive workingconditions associated with behaviorsexpected of each employee in an organization(e.g., Caplan, Cobb, French, Harrison, &Pinneau, 1975; Kahn, Wolfe, Quinn, Snoek,& Rosenthal, 1964).

Despite the contributions of the Univer-sity of Michigan research program, occupa-tional stress did not attract a lot of interestamong organizational psychologists in thelate 1960s and early 1970s. This changedin 1978, due in large part to a comprehensivereview and analysis of the occupational stressliterature. The authors of the review, whichappeared in the journal Personnel Psychology,were Terry Beehr and John Newman. TheBeehr and Newman (1978) compilation isgenerally regarded as an important scholarlywork and has been cited frequently, but itsgreatest contribution may have been to alertthose in the field of organizational psychol-ogy that occupational stress was an issueworthy of attention.

Evidence of the impact of Beehr andNewman’s review can be seen in the steepincrease in the volume of occupational stressresearch after its publication (Beehr, 1995,1998). Since then, several books, chapters,and comprehensive reviews have continuedto summarize this vast literature (e.g., Cart-wright & Cooper, 1997; Cooper & Dewe,

2004; Jex, 1998; Jex & Beehr, 1991; Kahn &Byosiere, 1992; Smith & Sulsky, 2005;Sullivan & Bhagat, 1992).

APPROACHES ANDTERMINOLOGY

Organizational psychologists have certainlycontributed much to the study of occu-pational stress, but a perusal of the occu-pational stress literature will show thatimportant contributions have also beenmade by physicians, clinical psychologists,engineering psychologists, labor economists,epidemiologists, and nurses, to name a few.To capture the interdisciplinary nature ofoccupational stress, Beehr and Franz (1987)proposed that occupational stress can beapproached from four different perspectives:(1) medical, (2) clinical/counseling, (3) engi-neering psychology, and (4) organizationalpsychology. The distinguishing feature of themedical approach to occupational stress is afocus on the contribution of stress in theworkplace to employee health and illness.When viewed from this perspective, stressfulaspects of the work environment may beconsidered pathogenic agents that contributeto disease conditions. Not surprisingly, manyresearchers who approach occupationalstress from this perspective are physiciansor have received their academic training insome other health-related field (e.g., healtheducation, nursing, public health).

The clinical/counseling approach to occu-pational stress emphasizes the impact ofstressful working conditions on mentalhealth outcomes (e.g., depression, anxiety).Beehr and Franz (1987) also point out that,compared to the others, this approach tendsto focus more on treatment than on research.That is, rather than focusing on why stressfulwork conditions lead to problems, adherentsof this approach tend to focus on developing

Approaches and Terminology 201

methods to relieve stress-related symptoma-tology (e.g., Beehr, Jex, & Ghosh, 2001). Asone would expect, the clinical/counselingapproach is dominated by those trained inclinical or counseling psychology.

The engineering psychology approach tooccupational stress focuses on sources ofstress that originate from the physical workenvironment. Examples of these mightinclude work schedules, pace of work, orperhaps the design of employees’ worksta-tions. This emphasis on the physical envi-ronment as a source of stress is notsurprising, given that the discipline of engi-neering psychology (also termed human fac-tors) focuses on the interface betweenemployees and the physical environment.Another distinctive feature of this approach,according to Beehr and Franz (1987), is thatit emphasizes the performance-related impli-cations of stress in the workplace. It is alsotrue, though not pointed out by Beehr andFranz, that much of the occupational stressresearch guided by this approach has exam-ined health-related outcomes such as phys-iological changes (Frankenhaeuser, 1979) orfatigue (Sparks, Cooper, Fried, & Shirom,1997).

The organizational psychology approach tooccupational stress is characterized by anumber of distinctive features. For onething, this approach tends to focus on whatwere previously defined as psychosocialsources of stress in the workplace. Thisimplies two things. First, this approach tendsto focus heavily on cognitive appraisal, or theprocess by which employees perceive thework environment and decide whether itis stressful. Second, as was pointed outearlier, this approach tends to focus onsources of stress that emanate from interac-tions with others (i.e., they are social innature). Another distinguishing feature ofthis approach, as compared to the others, is

that researchers tend to be interested in theimpact of occupational stress on employeeoutcomes that directly impact organizationaleffectiveness.

In recent years, these four approacheshave largely converged under one umbrellaknown as Occupational Health Psychology(OHP; Barling & Griffiths, 2003). OHP isessentially a new interdisciplinary field thatis concerned with using the theories andmethods of psychology (and closely relatedfields) to enhance the health, safety,and well-being of employees. It should bepointed out that, although occupationalstress is only one aspect of OHP, it is avery key one because of the impact thatstress has on employee health and safety(see Comment 7.1).

OCCUPATIONAL STRESSTERMINOLOGY

For many years occupational stress re-searchers have adopted (and, at times,struggled with) unique terminology. In thissection, the terminology used in occupa-tional stress research is briefly reviewed.No term has evoked more controversyand discussion than the term stress itself(see Comment 7.2). It can be defined in anumber of ways, but researchers havetended to adopt a stimulus, response, orstimulus–response definition. A stimulusdefinition implies that stress is some typeof force acting upon the individual. In every-day conversation, this definition might bereflected in the following sentence: ‘‘Bob hashad his share of stress at work during thepast year.’’ Notice in this statement that theterm stress is used to refer to negative aspectsof the work environment that may be trou-blesome to the individual.

A response definition implies that stressis synonymous with the way in which


employees react to stressful job conditions.Consider the following statement: ‘‘Barbarais feeling a lot of stress because of herupcoming performance review.’’ Stress isused here to represent the feelings that areevoked by something in the work environ-ment that the employee obviously considersto be aversive.

When a stimulus–response definition isused, the term stress is used merely to referto the overall process by which the workenvironment may negatively impact employ-ees. Rather than using the term stress to refersomething in the work environment or theresponse of an employee, the term stressor isused to represent aspects of the work envi-ronment that may require some adaptiveresponse on the part of employees. Forexample, one might note that a person

appears to be experiencing many stressorson his or her job. (Later in the chapter, anumber of common organizational stressorswill be described.)

The other term associated with thestimulus–response definition, strain, refersto a multitude of maladaptive ways employ-ees may react to stressors. For example, onemight observe that because of working longhours (a stressor), an employee appears to beshowing a great deal of strain. Occupationalstress researchers typically classify strains inthree categories: psychological, physical, andbehavioral. Psychological strains includeaffective or emotional responses to stressors.Common examples of these from the occu-pational stress literature include anxiety andfrustration (Spector, Dwyer, & Jex, 1988),hostility (Motowidlo, Packard, & Manning,

COMMENT 7.1

OCCUPTIONAL HEALTH PSYCHOLOGY: DEVELOPMENT OF A NEW FIELD

ONE OF THE most significant developments in

the study of occupational stress and health has

been the emergence of the field of Occupa-

tional Health Psychology (OHP). According toTetrick and Quick (2003), the purpose of

OHP is to ‘‘develop, maintain, and promote

the development of employees directly and

the health of their families’’ (p. 4). They go

on to state that many areas of psychology (e.g.,

clinical, industrial/organizational, health,

human factors), as well as other health-related

disciplines (e.g., occupational medicine, pub-lic health), have developed theories and meth-

ods that have the potential to positively impact

employee health and well-being, particularly

when members of these disciplines work to-

gether. Thus, the field of OHP can best be

described as an interdisciplinary approach to

the enhancement of employee health and

well-being.

While the field of OHP is still very new,

it has grown quickly and is international in

scope. Many universities offer concentrations

in OHP as part of the doctoral trainingin psychology, and there are now two major

professional organizations associated with

OHP (European Association of Occupational

Health Psychology; Society for Occupational

Health Psychology). There is also a scientific

journal devoted exclusively to OHP (Journal of

Occupational Health Psychology), and several

professional conferences have focused on thisfield.

Source: Tetrick, L.E., & Quick, J.C. (2003). Prevention at

work: Public health in occupational settings. In J.C. Quick

& L.E. Tetrick (Eds.), Handbook of occupational health

psychology (3–18). Washington, DC: American Psycholog-

ical Association.

Occupational Stress Terminology 203

1986), and depression (Heinisch & Jex,1997).

Physical strains include responses thatare related to employees’ physical healthand well-being. These have received consid-erable attention in recent years (e.g., Ganster& Schaubroeck, 1991) because of the esca-lating costs associated with healthcare. Themost common method of measuring physi-cal strain has been self-reported physicalsymptoms (e.g., Frese, 1985; Spector &Jex, 1998). Other methods that can be foundin the occupational stress literature includethe assessment of physiological indexes (e.g.blood pressure, heart rate; Fried, Rowland,& Ferris, 1984; Schaubroeck & Merritt,1997) and diagnosed disease conditions(Sales & House, 1971).

Behavioral strains have been explored theleast in occupational stress research. This islikely due to the difficulties associated withobtaining behavioral indexes, as well as a

lack of understanding of the many forms ofbehavior in organizations (e.g., Campbell,1990). Perhaps the most relevant form ofbehavioral strain in organizations is impairedjob performance. The majority of occupa-tional stress studies that have examined theimpact of stressors on job performance havemeasured it through the use of supervisoryratings (Jex, 1998). Other behavioral strainsthat have been examined, with varyingdegrees of success, include absenteeism,turnover, and substance abuse.

Despite the multitude of strains that havebeen studied over the years, it is important tokeep in mind that not all reactions to stres-sors are negative. A tight deadline may beseen as challenging, or perhaps a vaguelydefined role may provide an employee theopportunity to carve out her niche in anorganization. In fact, in recent years, psy-chologists have begun to systematicallystudy the more positive effects of many types

COMMENT 7.2

THE MEANING OF THE WORD STRESS

LIKE MANY OTHER researchers in various fields of

study, occupational stress researchers have

struggled considerably with terminology over

the years. Variance in terminology makes itmore difficult for researchers to communicate

and more generally decreases the rate of prog-

ress within any scientific endeavor. This can

also be a big problem if those who participate

in occupational stress studies have their own

definitions of important terms and concepts.

Several years ago, the first author con-

ducted a study, in collaboration with TerryBeehr and Cathy Roberts, in which we were

interested in a simple question: ‘‘How do sur-

vey respondents tend to interpret the word

‘stress’ when it is used on self-report question-

naires?’’ Our data seemed to indicate that

respondents had a tendency to use what is

termed a response definition. In other words,

if you ask people about their current ‘‘stress

level,’’ chances are they will answer in terms ofthe reactions they are having to the stressful

conditions they are experiencing in their job.

The practical implication of this study is

actually quite simple: Unless there is a com-

pelling reason to do so, it is probably not a good

idea to use the word stress in questionnaire

items. If it is used, researchers should be very

clear as to how they are defining it.

Source: S. M. Jex, T. A. Beehr, and C. K. Roberts. (1992).

The meaning of occupational ‘‘stress’’ items to survey

respondents. Journal of Applied Psychology, 77, 623–628.


of stressors (Britt, Adler, & Bartone, 2001;Nelson & Simmons, 2003).

OCCUPATIONAL STRESSMODELS

As was pointed out in Chapter 21, a theoret-ical model is simply a description of therelevant variables impacting some phenom-enon and the relationships among these var-iables. Theoretical models are quite useful inguiding behavioral science research and itsapplications. Over the years, a number oftheoretical models of occupational stresshave been developed to help guide bothresearch and organizational efforts to reducestress. Those that have had the greatestimpact are described in the following.

Institute for Social Research(ISR) Model

One of the earliest occupational stress mod-els came out of the previously mentioned

program of research at the University ofMichigan’s Institute for Social Research(French & Kahn, 1962; Katz & Kahn,1978). For this reason, it has come to beknown as the ISR model of occupationalstress. As can be seen in Figure 7.1, thismodel begins with the objective environ-ment. This includes anything in an employ-ee’s work environment: the number of hoursworked, the amount of responsibility, theextent to which interaction with others isrequired, and so on.

The next step in this model is labeledthe psychological environment. At thisstep, the employee perceives the objectiveenvironment. The employee at this point isappraising aspects of the work environ-ment and making some judgment as towhether or not they are threatening (Laza-rus, 1966). As was stated in the previousdiscussion of approaches to occupationalstress, the appraisal process is a key com-ponent of the organizational psychologyapproach.

FIGURE 7.1The ISR Model of Occupational Stress

ObjectiveEnvironment

PsychologicalEnvironment

Response(Physiological,

Behavioral,A e ctive)

Mental andPhysical

Health andDisease

Interpersonal Relations

Enduring Properties of the Person(Genetic, Demogr aphic, Personality)

1

A C DB

2

6

5

3 4

Source: Adapted from: D. Katz and R. L. Kahn. (1978). The social psychology of organizations (2nd ed.). New

York: Wiley. Reprinted by permission.


Once the environment is appraised, themodel proposes that the result may be imme-diate physiological, behavioral, and emo-tional responses on the part of theemployee. Physiological changes that arecommonly evoked by stressful stimuliinclude increases in heart rate, elevatedblood pressure, and increased respirationrate (Fried et al., 1984). Immediate behav-ioral responses may include decreased effort,or perhaps an inability to concentrate (Jex,1998). Emotional responses may includeincreases in both anxiety and depressivesymptoms, and a decrease in job satisfaction(Heinisch & Jex, 1997; Spector et al., 1988).

Depending on the severity and durationof the immediate responses, the result maybe adverse changes in mental and physicalhealth. For example, an employee whoseinitial response to a stressor (such as animpending deadline) is increased anxietymay end up feeling anxious all the time.On the physical side, an employee who expe-riences a short-term elevation in blood pres-sure in response to this same stressor mayeventually develop chronic hypertension orcoronary heart disease.

The next two components in the ISRmodel (5 and 6) are meant to illustrate theimpact of individual differences on all of theprocesses depicted in the model. For exam-ple, people obviously differ considerably interms of genetic makeup, demographic char-acteristics, personality traits, and the qualityof their interpersonal relations with others.Furthermore, any or all of these factors mayimpact the manner in which people perceivethe objective environment, their immediateresponses to perceived stressors, and, ulti-mately, whether stressors lead to adversemental and physical health.

While the ISR model has not been explic-itly tested, it has served as a conceptual guidefor a substantial portion of the occupational

stress research conducted over the years, andfor that reason it has been quite influential.The primary weakness of this model is itsgenerality, which is probably the main rea-son it hasn’t been empirically tested. Themodel also does not provide specifics abouteach of the steps in the process. One couldalso argue that important variables and proc-esses have been left out of the model. Forexample, the model does not explicitlyaccount for employees’ efforts to cope withstressors, or acknowledge that stressors mayimpinge on the employee from outside of theorganization.

Beehr and Newman’s Facet Model

In addition to providing a comprehensivereview of the occupational stress literature,Beehr and Newman (1978) proposed amodel of the occupational stress process.The primary reason for proposing thismodel, according to these authors, was toserve as a guide to categorizing the occupa-tional stress literature. Thus, like the ISRmodel, this model was meant to be a guidefor researchers rather than a model tobe empirically tested. As can be seen inFigure 7.2, this model proposes that theoccupational stress process can be brokendown into a number of ‘‘facets’’ that repre-sent categories of variables to be studied.Going from left to right in the model, thePersonal facet represents the stable character-istics that employees bring with them to theworkplace. Examples of variables includedhere would be demographic characteristics(e.g., age, gender, race) and personality.

The Environmental facet, in contrast,represents those stimuli in the work environ-ment that individual employees must con-front. Variables comprising this facet wouldinclude characteristics of the work per-formed (e.g., level of complexity) as well as


the nature of job-related interpersonal rela-tions. It is in the Process facet that character-istics of the person and the situation interact.This is the point at which employeesappraise the work environment and ulti-mately decide whether it is harmful.

If the employee perceives stressors to bepresent after the environment is appraised,there may be a variety of consequences forboth the individual employee and the orga-nization as a whole. The Human Conse-quences facet represents the multitude ofways in which employees may respond tostressors that primarily have implicationsfor each individual employee (e.g., healthproblems, substance abuse). In contrast,the Organizational Consequences facet rep-resents employee responses that primarilyhave implications for organizational func-tioning (e.g., higher rates of absenteeismand turnover, impaired employee job per-formance).

Depending on the consequences for theindividual and the organization, someresponse may be required. The Adaptive

Response represents efforts on the part ofindividuals and organizations to respondadaptively to stressors. An adaptive responsefor an individual may be to exercise when heor she feels tense or anxious. An organizationmay respond to increased absenteeism byinstituting flexible work hours.

The final facet in this model is Time. Asis evident from Figure 7.2, this facet has animpact on all other facets in the model. Itexhibits recognition of the fact that theprocess of employees’ appraising the envi-ronment, determining what aspect(s) arestressful, and ultimately responding tothose perceived stressors, is embedded in atemporal context. In some cases, this processmay be very short; at other times, it mayoccur over a period of several years. As Beehrand Newman (1978) aptly pointed out, this isprobably the least understood of all thefacets because of the reliance on cross-sectional research designs. (See McGrathand Beehr, 1990, for a more extensive dis-cussion of the role of time in occupationalstress research.)

FIGURE 7.2Beehr and Newman’s Facet Model of Occupational Stress

ProcessFacet

PersonalFacet

EnvironmentalFacet

AdaptiveResponses

Facet

HumanConsequences

Facet

OrganizationalConsequences

Facet

Time Facet

Source: Adapted from: T. A. Beehr and J. E. Newman. (1978). Job stress, employee health, and organizational

effectiveness: A facet analysis, model, and literature review. Personnel Psychology, 31, 665–699. Reprinted by

permission.


Demands---Control Model

A model of occupational stress that is muchmore specific than the others presented tothis point is known simply as the Demands–Control model. This model, which was pro-posed by Robert Karasek in the late 1970s,posits that the most stressful situations in theworkplace are those in which employees faceheavy job demands but, at the same time, aregiven little control over their work. (Karasek[1979] used the term Job Decision Latitude todenote control.) A good example would bethe situation of a typical factory worker in theScientific Management era. Recall fromChapter 1 that one of the major principlesof Scientific Management was to provide pro-duction employees with challenging goals,usually in the form of production standards.At the same time, proponents of ScientificManagement argued that these same employ-ees should have little control over things suchas the design of work methods and the sched-uling of rest breaks. Factory employees dur-ing this period also had little control over thereliability of machinery or the motivationlevels of their fellow employees.

Most research using the Demands-Control model as a theoretical frameworkhas examined health and physiologicaloutcomes (e.g., Fox, Dwyer, & Ganster,1993; Karasek, Baker, Marxer, Ahlbom, &Theorell, 1981; Perrewe & Ganster, 1989).This limits the scope of the model somewhat,although it is certainly possible that thescope of the Demands–Control model couldbe broadened. In fact, some research thattested the Demands–Control model hasinvestigated psychological outcomes (e.g.,Spector, 1987a).

It is alsoworthmentioning that recent testsof the Demands–Control model have shownthat the interaction between job demands andcontrol may be more complex than Karasek

originally proposed. Most notably, Schau-broeck and Merritt (1997) found, in predict-ing blood pressure, that the interactionbetween demands and control that is pre-dicted by Demands–Control theory wasobserved only for employees who reportedhigh self-efficacy. This suggests that havingcontrol over one’s work tasks is helpful to anemployeeonly ifheor she feels able toperformthose tasks (i.e., has high self-efficacy). Someresearch has also shown that the proposedDemands-Control interaction may be lesslikely to occur among employees who havehigh levels of social support (Jonsson,Rosengren, Dotevall, Lappas, & och Wilhelm-sen, 1999); in fact, some now refer to thismodel as the Demands-Control-Support model.

Person---Environment Fit Model

This model of occupational stress actually hasimplications for many organizational phe-nomena (e.g., selection, socialization). Thehistorical roots of the Person–Environment(P–E) Fit approach can be traced back to KurtLewin and his notion of interactional psychol-ogy. Lewin simply believed that humanbehavior is a function of the interactionbetween characteristics of the person andcharacteristics of the situation. One aspectof this interaction that is relevant to occupa-tional stress is the degree to which there is afit between the person and the situation.According to this approach, an employeeperceives the work environment as stressfulwhen there is a lack of fit (Caplan, 1987;French, Caplan, & Harrison, 1982).

The general notion of P–E Fit is rathersimple but there are many ways in which fit(and misfit) between an employee and thework environment can occur. According toKristof (1996), fit (and misfit) may indicatethe degree to which an employee’s skills andabilities match the requirements of the job he


or she is performing. An employee who lacksthe skills and abilities necessary to perform ajob may feel overwhelmed and inadequate.Conversely, when job requirements are wellbelow an employee’s capabilities, the resultsmay be boredom, frustration, and dissatisfac-tion. In either case, it is very likely that suchan employee will perceive the job as stressful.

The concept of P–E Fit (and misfit) mayalso occur at a more ‘‘macro’’ level of analysis.More specifically, one can speak of thedegree of fit between characteristics of theemployee and characteristics of the organi-zation. For example, suppose an employeewho places a very high value on individualaccomplishment goes to work for an organi-zation that places a very high value onteamwork. This would obviously be a poorfit and it is likely that such an individualwould ultimately find working in such anenvironment stressful.

The P–E Fit approach has proven to bequite useful to occupational stress research-ers. Over the years, considerable refinementshave been made both in conceptualizing fit(e.g., Edwards, 1994; Kristof, 1996) and inthe statistical analysis of P–E Fit data(Edwards & Parry, 1993). Perhaps the majorlimitation of the P–E Fit approach is that,compared to the Environment component,we seem to be much further ahead in themeasurement of the Person component ofthe model. That is, psychologists havedevoted considerable time and energy toconceptualizing and measuring individualcharacteristics such as abilities, skills, andpersonality. Far less attention has been givento conceptualizing and measuring uniquecharacteristics of organizations.

Effort-Reward Imbalance

This model has been developed fairlyrecently, and has become quite influential

in guiding occupational stress research inEurope. The model was developed byJohannes Siegrist in Germany, who pro-posed that the most stressful situations arethose in which an employee’s efforts are outof proportion with the rewards that his or herjob is providing him or her (Siegrist, 2002);put differently, an employee is putting moreinto the job than he or she is getting out of it.One can think of many occupations in oursociety that would likely fit this ‘‘high effort–low reward’’ profile (e.g., teaching, socialwork, nursing). Siegrist also proposed thatpeople who are extremely committed to workare most likely to experience this imbalancebecause it is unlikely that the rewards theyreceive will ever match their efforts.

Since the Effort-Reward Imbalancemodel has been developed fairly recently, ithas not been empirically tested as much asthe Demands-Control or P–E Fit models.However, it has been shown that those expe-riencing high Effort-Reward Imbalance aremost susceptible to symptoms of poor health(Peter, Geibler, & Siegrist, 1998; Siegrist,1996; Siegrist et al., 2004).

Comparison of OccupationalStress Models

Having described the four occupational stressmodels, some comparison of the relative mer-its of each is inorder. In termsofusefulness, allof the models presented have some merit.However, over the years, the ISR model andthe model proposed by Beehr and Newman(1978) have guided the bulk of occupationalstress research. This suggests thatbothmodelshave served as road maps to guide occupa-tional stress researchers’ efforts. It is relativelyeasy to use each of these models to guide aspecific research investigation or to clarifythe focus of a stress-related organizationalintervention.


Of all the occupational stress models, theDemands-Control model (Karasek, 1979)has clearly received the most empirical scru-tiny, and the results have been mixed (e.g.,Fox et al., 1993; Perrewe & Ganster, 1989;Spector, 1987a). As was stated earlier, thismay be due to the fact that the conditionsunder which demands and control interactare more complex than those Karasek origi-nally proposed (e.g., Schaubroeck & Merritt,1997). The P–E Fit approach has alsoreceived a fair amount of empirical testing,though not all of this testing has been in thecontext of occupational stress research. Inthis case, the results have generally beenfavorable (e.g., Kristof, 1996). Since theEffort-Reward Imbalance model has beendeveloped only recently, it has not beensubject to a great deal of empirical testingso more research is needed. However, tests ofthis model that have been conducted havesupported it (Siegrist, 2002).

WORKPLACE STRESSORS

A stressor represents anything in the job ororganizational environment that requiressome adaptive response on the part of theemployee. One of the difficulties in coveringstressors is simply deciding which ones todescribe when there are so many in theworkplace. The stressors covered in this sec-tion represent two general types: (1) thosethat have been commonly studied or havereceived considerable attention in the occu-pational stress literature, and (2) those thathave received less attention but have morerecently become the focus of attention.

Commonly Studied Stressors

Role Stressors: In the history of occupa-tional stress research, role stressors havebeen given far more attention than any

other type. A role is essentially a set ofbehaviors that are expected of an individ-ual. Most people have multiple roles (e.g.,parent, employee, student, spouse), so itstands to reason that people also have mul-tiple sets of role demands. In complexsocial systems such as organizations, rolesserve the important function of bringingorder and predictability to the behavior ofindividuals (Katz & Kahn, 1978). Rolesalso help individual employees gaugewhether they are doing what they are sup-posed to be doing.

Employees in organizations receive role-related information through both formal andinformal sources. In many organizations, themost common formal source of role-relatedinformation is a written job description.Other common sources are verbal and writ-ten communications with one’s immediatesupervisor. All of these formal sources mayprovide important information, but they mayalso be limited in defining an employee’srole. For example, written job descriptionsare often very general and become out-dated quickly. In addition, supervisors’ jobknowledge may be lacking, and communi-cation is often imprecise.

To compensate for limitations in formalsources of information, employees may lookto informal sources as they define theirorganizational roles. These may includeinformal interactions with coworkers at thesame level, as well as encounters with sub-ordinates and even with persons outside theboundaries of the organization (e.g., custom-ers, suppliers, regulatory agencies). The termrole set encompasses the various sources ofinformation, both formal and informal, thatemployees utilize in defining their roles inorganizations.

The communication of role-related infor-mation should be a smooth process in whichthe various members of a role set provide


clear and consistent information to employ-ees. However, we know that this does notalways happen. When role-related informa-tion is unclear, this may lead to a stressorknown as role ambiguity (Kahn et al., 1964;King & King, 1990). In the most generalsense, role ambiguity exists when an employ-ee is unsure of what he or she is supposed todo. This uncertainty can be manifested in avariety of ways: unclear performance stand-ards (Beehr, Walsh, & Taber, 1980), uncer-tainty regarding scheduling and workmethods (Breaugh & Colihan, 1994), andso on. A common example of role ambiguityexperienced by students is professors’ lack ofclarity regarding grading standards.

Another common problem that mayoccur is a lack of consistency in the role-related information provided by the variousmembers of an employee’s role set. Whenthis occurs, the stressor that may result isknown as role conflict (Kahn et al., 1964;King & King, 1990). Role conflict usuallyresults from inconsistent information orconflicting demands provided by differentmembers of an employee’s role set. It is alsopossible that the same individual within anemployee’s role set may communicateinconsistent information or requests overtime. For many college professors, theirteaching responsibilities and research activ-ities form a common source of role conflict.The more time they spend on teaching, theless time they have available for research,and vice versa.

A third role stressor that has been exam-ined, though not nearly as much as roleambiguity and conflict, is role overload. Thisstressor is defined by Jones, Flynn, and Kel-loway (1995) as occurring when ‘‘anemployer may demand more of an employeethan he or she can accomplish in a giventime, or simply, the employee may perceivethe demands of work as excessive’’ (p. 42).

Given the generality of this definition, it ispossible that an employee may feel over-loaded for two reasons. First, feelings of roleoverload may be due to the sheer volume ofthe demands emanating from an employee’srole set (this is referred to as quantitative roleoverload). During tax season, for example,many accountants experience this type ofrole overload. Second, role overload maybe due to the difficulty of the demands,relative to the skills and abilities of the em-ployee (this is referred to as qualitative roleoverload). This form of role overload isbecoming very common in all of the armedservices because the skill requirements ofnew technology often exceed those ofenlisted personnel.

A great deal of occupational stressresearch has been grounded in role theory,so it is not surprising that more empiricalresearch has been done on role stressors thanon any of the other stressors covered in thischapter. Several meta-analyses have beenconducted to summarize this vast literature(Abramis, 1994; Fisher & Gitelson, 1983;Jackson & Schuler, 1985; Tubre, Sifferman,& Collins, 1996). Overall, the results fromthese meta-analyses have been quite consis-tent in showing that role ambiguity and roleconflict are correlated with a variety ofstrains. Table 7.1, for example, showscorrected correlations from Jackson andSchuler’s study. As can be seen, high levelsof role ambiguity and role conflict are asso-ciated with low job satisfaction, high anxietyand tension, and a higher probability of turn-over. Correlations with behavioral outcomessuch as absenteeism and job performance,however, are very small.

Compared to role ambiguity and roleconflict, much less research has examinedthe effects of role overload. The few studiesthat have been done, however, have shownthis stressor to be related to higher levels of


both psychological and physical strain (e.g.,Caplan et al., 1975; Caplan & Jones, 1975;Jex, Adams, Elacqua, & Bachrach, 1998).Interestingly, some evidence suggests thatquantitative role overload may actually bepositively associated with job performance(Beehr, Jex, Stacy, & Murray, 2002). Em-ployees who perform their jobs well mayreceive a disproportionate share of workassignments. Also, in some jobs (e.g., sales),the volume of work one has to contend withis directly proportional to one’s success.Qualitative workload, however, is associatedwith lower performance ratings, at leastin a training context (Britt, Thomas, &Dawson, 2006).

Workload: Workload is typically definedas the amount of work an employee has todo in a given period of time. This defini-tion, however, is deceptively simple. Forexample, for many jobs it is necessary todistinguish between perceptions of work-load and objective workload. In a purelyobjective sense, two employees may haveexactly the same volume of work but per-ceive their respective workloads quite dif-ferently. Another complicating factor inattempting to understand the impact of

workload is that it is cyclical. Employees inretail stores, for example, experience asharp increase in workload as the end-of-year holiday season approaches, but thispeak is followed by a decline in January.Finally, as with role overload, it is neces-sary to distinguish between the sheer vol-ume of work one is required to perform(quantitative workload) and the difficultyof that work (qualitative workload).

In general, studies of the impact of work-load have focused heavily on physical out-comes. In one of the earliest studies ofworkload, Buell and Breslow (1960) foundthat working more than the typical 40 hoursper week doubled the chances of mortalityfrom coronary heart disease in men. Subse-quent research, over several years, hasshown that working long hours is associatedwith a variety of indicators of ill health,although the effects appear to be rather mod-est (Sparks et al., 1997). The negative phys-ical effects associated with long hours aremost likely due to physiological changes thatoccur during periods of overwork. Accord-ing to research conducted in Sweden byFrankenhaeuser (1979), the level of adrena-line and other catecholamines increases pre-dictably during periods of long work hours.If these adrenal hormones remain consis-tently elevated over an extended period oftime, the risk of a number of illnesses mayincrease. To date, however, longitudinalresearch linking cyclical elevations in adre-nal hormone levels to illness is lacking (Jex &Beehr, 1991).

This type of effect has also been shownwith indices of workload other than workhours. Rau (2004), for example, had jobanalysts independently assess the demandsof a sample of white-collar employees inGermany. Furthermore, demands mea-sured in this way were associated with ele-vations in blood pressure during the work

TABLE 7.1Corrected Correlations Between Role Stressorsand Both Affective and Behavioral Outcomes

Outcome Role Ambiguity Role Conflict

Job satisfaction �.46 �.48

Tension/Anxiety .47 .43

Turnover intent .29 .34

Absences .13 �.02

Performance ratings �.12 �.11

Source: S. E. Jackson and R. S. Schuler. (1985). A meta-

analysis and conceptual critique of research on role ambiguity

and role conflict in work settings. Organizational Behavior and

Human Decision Processes, 36, 16–78.


day. Interestingly, blood pressure returnedto normal levels after the work day only forthose who also had a great deal of controlover how they did their jobs.

In addition to physical strains, studieshave examined both the psychological andthe behavioral effects using a variety of work-load indexes. Spector and Jex (1998)recently summarized the results of severalstudies that examined perceived workload,and found it to be related to high anxiety andfrustration, reduced job satisfaction, andincreased turnover intentions. These authorsalso found that perceived workload was pos-itively related to job performance ratings.Recall that this was also true for quantitativerole overload (Beehr et al., 2000), suggestingthat not all stressors lead to negative out-comes. In the long run, however, this rela-tionship might be detrimental. Employeeswho perform well and thus have to shouldera disproportionate share of the workloadmay eventually tire of such conditions andleave the organization.

Interpersonal Conflict: Most jobs requireat least a minimal amount of interactionwith other people (e.g., coworkers, cus-tomers, contractors). Such social interac-tions are often a source of satisfaction andpersonal fulfillment (Nielsen, Jex, &Adams, 2000; Riordan & Griffeth, 1995).Interactions with others can also makework more stressful if interpersonal conflict(Keenan & Newton, 1985; Spector,1987), defined as negatively chargedinteractions with others in the workplace,develops. Negative interactions can rangefrom something as minor as a momentarydispute over a parking space to heatedarguments (see Comment 7.3). At extremelevels, interpersonal conflicts may evenescalate to physical violence (O’Leary-Kelly, Griffin, & Glew, 1996).

Research suggests that there may be sev-eral potential causes of interpersonal con-flict. Perhaps the most widely citedprecursor to conflict is competition amongpeople (Forsyth, 1999). In many organiza-tions, employees must compete for rewardssuch as pay raises, promotions, and compet-itive budget allocation processes (e.g., themore Department A receives, the less Depart-ment B receives). This policy of one person’s(or department’s) gain is another’s loss oftenfosters a high level of competition.

Another factor that may lead to interper-sonal conflict is rude or contentious behavioron the part of employees. This may occur, forexample, when one person tries to influenceanother through threats or coercion (Falbe &Yukl, 1992), although such behavior may attimes be directed at no one in particular(Andersson & Pearson, 1999). As Falbe andYukl point out, employees who are the targetof contentious influence tactics typically donot respond favorably; in fact, they may retal-iate. In either case, the odds that interpersonalconflict will occur are heightened.

Interpersonal conflict may also occur inresponse to behavior that is not intentionallydirected at another individual, but ultimatelyhas a negative effect. An example of this typeof behavior is ‘‘free riding’’ in groups (Alba-nese & Van Fleet, 1985; Roberts, 1995). Freeriding occurs when one or more members ofa work group do not ‘‘pull their weight’’ and,as a result, other group members must pickup the slack. Those who must pick up theslack may resent the free rider, and thisresentment may ultimately come out in theform of strained interpersonal relations.What’s important to note about this exampleis that the person who is perceived to be afree rider may have had no intention ofangering his or her fellow group members.This person, in fact, may not even realize thathe or she is free riding.


Compared to role stressors, consider-ably less research has been conducted onthe effects of interpersonal conflict in orga-nizations. Probably the most compre-hensive summary of the impact ofinterpersonal conflict was provided by thepreviously described meta-analysis bySpector and Jex (1998). This meta-analysisshowed that interpersonal conflict is corre-lated with a number of psychological,physical, and behavioral strains. The mostnotable finding from the meta-analysis wasthat interpersonal conflict was moststrongly related to feelings of anxiety atwork. This would appear to be a logicalfinding, given that anxiety is an emotionfelt in anticipation of future problems andchallenges (Spielberger, 1979). Those who

experience high levels of interpersonalconflict at work may spend time rumi-nating over the possible effects of pastconflicts, and may worry over future con-flicts before they even occur.

Organizational Constraints: Organiza-tions have a vested interest in facilitating thejob performance of their employees. To theextent that individual employees are effec-tive, the more effective the organization willultimately become. However, anyone whohas worked in any organization knows thatorganizational conditions do not alwaysfacilitate performance. In fact, organizationalconditions may even detract from or con-strain employee performance. For example,it is not unusual for employees to have

COMMENT 7.3

THE COMPLEXITY OF INTERPERSONAL CONFLICT

INTERPERSONAL CONFLICT IS a stressor that has not

been studied for a long period of time but has

been shown to negatively impact employees

(e.g., Spector & Jex, 1998). In fact, if you talkto people working in organizations, this is

often mentioned as a very important source

of stress. People strongly dislike coming to

work when they don’t get along with others,

or when fellow employees are embroiled in

conflicts.

Given the effects of interpersonal conflict, it

is important that occupational stress research-ers get a better handle on all of the forms in

which interpersonal conflict can manifest itself

in organizations. Most studies, to date, have

assessed relatively mild forms of interpersonal

conflict. However, we know that what consti-

tutes interpersonal conflict ranges widely from

minor arguments to physical violence.

Another aspect of interpersonal conflictthat has yet to be explored is its active versus

passive forms. Active conflict, which has been

the focus of most of the research, includes

arguments and saying rude things to others.

More passive forms of conflict might includenot returning a fellow employee’s phone calls,

or perhaps ‘‘forgetting’’ to invite a coworker to a

meeting.

In summary, interpersonal conflict is an

important variable that is much more complex

than current research would seem to indicate.

A great deal more conceptual work needs to be

done before we are able to get a clear picture ofthe impact of interpersonal conflict in organi-

zations.

Source: P. E. Spector and S. M. Jex. (1998). Development

of four self-report measures of job stressors and strain:

Interpersonal Conflict at Work Scale, Organizational Con-

straints Scale, Quantitative Workload Inventory, and

Physical Symptoms Inventory. Journal of Occupational

Health Psychology, 3, 356–367.


difficulty doing their jobs because of unneces-sary rules and procedures, a lack of resources,or interruptions from fellow employees.

Peters and O’Connor (1980) used theterm situational constraints to describe a vari-ety of organizational conditions that mayprohibit employees from performing up totheir capabilities. (In this section, the termorganizational constraints is used in recogni-tion of the fact that constraints are not alwaystied to specific situations.) To more fullydefine organizational constraints, Petersand O’Connor (1980) proposed a classifica-tion system consisting of 11 different cate-gories of organizational constraints. Theseinclude (1) job-related information, (2)budgetary support, (3) required support,(4) time and materials, (5) required servicesand help from others, (6) task preparation,(7) time availability, (8) the work environ-ment, (9) scheduling of activities, (10) trans-portation, and (11) job-related authority.

For any of these categories of constraints,the inhibiting effect on performance may bedue to unavailability, inadequacy, or poorquality (or some combination of these). Con-sider the category of job-related information.Employees, in some cases, may lack the in-formation needed to accomplish job-relatedtasks. In other cases, there may be informa-tion available, but not enough to accomplishrequired tasks. In still other cases, there maybe plenty of information available, but theinformation is of such poor quality that it isof limited value; hence, employee perfor-mance is constrained.

Since Peters and O’Connor (1982) firstintroduced the concept, many studies haveexamined relations between organizationalconstraints and a variety of stress-relatedoutcomes. In fact, so many studies havebeen conducted that two recent meta-anal-yses have summarized their findings (Spec-tor & Jex, 1998; Villanova & Roman, 1993).

The specific outcomes examined in thesemeta-analyses differ somewhat, but the majorconclusion from both is that organizationalconstraints are most strongly related to neg-ative emotional reactions on the part of em-ployees. These include things such as jobdissatisfaction, frustration, and anxiety.

One finding is common to both meta-analyses (which is somewhat puzzling): thelack of a relation between organizationalconstraints and job performance. Organiza-tional constraints are things in the environ-ment that inhibit performance, so one wouldexpect a much stronger relation. However,Peters and O’Connor (1988) point out thatin most organizations several factors workagainst such a relation. For instance, perfor-mance appraisals are often conducted poorlyand may ultimately restrict the variability insuch measures (Cascio, 1998). Also, in manyorganizations, performance standards arevery low, and employees are offered littleincentive to perform above these standards.

The lack of a relationship between orga-nizational constraints and performance mayalso be due to the fact that measures of in-roleperformance have been used. Jex, Adams,Bachrach, and Sorenson (2003) found thatthe presence of organizational constraintswas negatively related to altruistic or helpingbehaviors among employees. This suggeststhat when employees experience organiza-tional constraints they may do what theyhave to do (e.g., fulfill their basic job require-ments), but may not go beyond this.

Perceived Control: The human desire tomaintain control over the environment,even if this control is illusory, has been welldocumented in the behavioral sciences lit-erature (e.g., Averill, 1973; Friedland,Keinan, & Regev, 1992). Compared to theother stressors covered in this chapter, per-ceived control is much more general and


thus can be manifested in a variety of ways.According to Spector (1986), the two pri-mary ways that perceived control is mani-fested in organizations are through jobautonomy and participative decision making.A high level of job autonomy indicates thatan employee has discretion over how his orher job tasks are to be performed, and per-haps over things such as starting and end-ing times (Hackman & Oldham, 1980). Asan example, university professors have con-siderable autonomy (some would say, toomuch!) over many aspects of their jobs,while manual laborers and conveniencestore clerks typically have little autonomy.

Participative decision making is defined byLowin (1968) as an organizational form ofdecision making in which those responsiblefor implementing decisions have some inputin their formulation. This participation couldtake a variety of forms but is most typicallyidentified with labor-management commit-tees, quality circles, job enrichment, andother shared governance policies (Cotton,1995). As an illustration of how participativedecision making might be carried out, sev-eral years ago one of the authors and a col-league assisted a medium-size producer ofdairy products in conducting a company-wide employee opinion survey. All phasesof this project were conducted in collabora-tion with an ‘‘Employee Committee,’’ whichconsisted of approximately 12 individuals,from several divisions of the company,who were responsible for representing theinterests and views of their fellow employees.

Like other stressors covered in this sec-tion, meta-analyses have summarized theeffects of both job autonomy and participa-tive decision making. For example, Spector(1986) summarized the findings of 88 stud-ies conducted between 1980 and 1985. Asummary of the results of this meta-analysisis provided in Table 7.2. As can be seen, both

manifestations of perceived control are pos-itively correlated with job satisfaction andnegatively related to a number of strains.For example, employees who perceive a lackof control also tend to report being emotion-ally distressed and experiencing physicalsymptoms; exhibit lower levels of perfor-mance; and are more likely to quit their jobs.

Since Spector’s (1986) investigation,other meta-analyses have focused more spe-cifically on participative decision making(e.g., Wagner, 1994; Wagner & Gooding,1987). Most of the outcomes examined inthese meta-analyses, with the exception ofjob satisfaction and job performance, are notrelevant to occupational stress. However, inboth studies, participative decision makingwas strongly and positively related tojob satisfaction, suggesting that a lack ofparticipation may lead to negative attitudinalreactions.

Contemporary OrganizationalStressors

The stressors discussed in the previous sec-tion are those that have received the greatest

TABLE 7.2Corrected Correlations between Two Measures ofPerceived Control and Affective, Health, andBehavioral Outcomes

Outcome Job Autonomy Participation

Job satisfaction .37 .44

Emotional distress �.37 �.18

Physical symptoms �.33 �.34

Turnover intent �.26 �.20

Turnover �.25 �.38

Performance .26 .23

Source: P. E. Spector. (1986). Perceived control by employees:

A meta-analysis of studies concerning autonomy and parti-

cipation at work. Human Relations, 39, 1005–1016. Copyright

# The Tavistock Institute, 1986. Reprinted by permission of

Sage, Ltd.


attention in the occupational stress litera-ture. These stressors are also somewhat time-less in that they have been present in theworkplace for many years and will likely bethere for quite some time into the future. Thestressors covered in this section, in contrast,have received comparatively less attentionbecause their increased importance is relatedto more recent trends. These contemporarystressors include work–family conflict,mergers and acquisitions, layoffs and jobinsecurity, and emotional labor.

Work---Family Conflict: Conflict betweenwork and family is certainly not a new stres-sor. In recent years, however, there are rea-sons that work–family conflict has indeedincreased in importance as a stressor. Forexample, it has been reported that 85% ofemployees have some day-to-day familyresponsibilities that they must balance withwork (Bond, Galinsky, & Swanberg, 1998).

Also, extended families are becomingmore geographically dispersed and gen-erally are having less contact with eachother, compared to previous generations.Thus, it appears that, for employees today,the demands from work and family domainsare competing more than ever. At thesame time, sources of support that have tra-ditionally been available to help balancethose demands (i.e., extended family) areincreasingly unavailable.

In describing work–family conflict, re-searchers make the distinction between whatis termed work–family conflict, and what istermed family–work conflict. Work–familyconflict occurs when the demands of workinterfere with one’s family responsibilities.For example, an unexpected meeting late inthe day may prevent a parent from pickingup his or her child from school. In contrast,family–work conflict occurs when thedemands of family interfere with one’s work

responsibilities. A very common examplewould be a parent’s need to leave his or herwork in order to take care of a sick child.

Another distinction often made in thework–family conflict literature is betweentime-based conflict, strain-based conflict, andbehavior-based conflict. With time-based con-flict, the time demands in one domain makeit more difficult to attend to one’s responsi-bilities in the other. Since the typical work-week is now well over 40 hours per week(Sparks et al., 1997), this is often the reasonwhy work interferes with family. This wouldbe ‘‘time-based, work–family conflict.’’ Onthe other hand, family demands can be verytime-consuming and may interfere withwork. Any parent of infant or preschool chil-dren would certainly attest to the timedemands associated with this role. Thiswould be ‘‘time-based, family–work conflict.’’

Strain-based conflict occurs when thestrain due to stressors in one domain impairsa person’s functioning in the other. Liketime-based conflict, this can occur in twodirections. For example, if a person is anx-ious and tense because of an increase in hisor her workload, this response may have anegative impact on the quality of interactionswith family members. In contrast, an em-ployee who is emotionally distraught overhaving to care for an elderly parent may havedifficulty concentrating on work, and his orher performance may suffer.

Behavior-based conflict occurs when thebehaviors required in one domain conflictwith those required in the other. For exam-ple, the tough and demanding persona that afootball coach may need to project to hisplayers may conflict with the tendernessand patience needed as a parent. Like theother forms of conflict, this one could also goin the opposite direction as well. A summaryof the various forms of work–family conflictis presented in Table 7.3.


Although work–family conflict is a rela-tively new research domain, there has been afair amount of empirical research on theeffects of this stressor. Kossek and Ozeki(1998) summarized much of this researchin a meta-analysis and found that work–family and family–work conflict were bothnegatively associated with job and life satis-faction. Interestingly, though, work–familyconflict was more strongly correlated withboth job and life satisfaction than wasfamily–work conflict. The major implicationof this finding is that employees may find itmore stressful to have their work interferewith their family life than the reverse.Another interesting finding from this studywas that work–family conflict was morestrongly related to job and life satisfactionamong women than it was among men.Thus, when the demands of work interferewith family responsibilities, women evi-dently find it more stressful than do men.This may reflect the fact that despite recentsocietal changes, women still take on agreater share of family responsibilities thanmen do (Hochschild, 1989).

Recent research has also been conductedlooking more closely at the causes of em-ployee perceptions of work–family conflict.Judge and Colquitt (2004), for example,found that employees who perceived that theywere treated fairly by their employing organi-zations tended to perceive lower levels ofwork–family conflict. Along this same vein,Lapierre and Allen (2006) found that support

from supervisors tended to be associated withlower levels of work–family conflict. Bothfindings suggest that organizations may beable to do a great deal to reduce conflictbetween the work and family domains.

Mergers and Acquisitions: The trend fororganizations to engage in mergers andacquisitions started in the mid-1980s andhas continued since. Mergers occur whentwo separate organizations combine to formone. An acquisition, on the other hand,occurs when one company (which is typi-cally larger) obtains a controlling finan-cial interest in another company (which istypically smaller). The acquirer thenassumes a dominant role over the acquired.Hogan and Overmeyer-Day (1994) pointout that, in practice, it is often difficult toclearly distinguish between mergers andacquisitions. Acquiring organizations oftenwant to convey the impression that the twoorganizations are equal partners. Given thisfuzzy boundary, mergers and acquisitionsare discussed here as one stressor.

According to Hogan and Overmeyer-Day (1994), much of the research on merg-ers and acquisitions has focused on thefinancial and strategic implications of thesetransactions. A somewhat smaller body ofliterature has examined the stress-relatedimplications (e.g., Buono & Bowditch,1989; Rentsch & Schneider, 1991;Schweiger & DeNisi, 1991). Relatedresearch efforts have also been aimed at

TABLE 7.3A Taxonomy of Different Forms of Work-Family Conflict

Direction of Conflict

Time Time-Based, Work-Family Conflict Time-Based, Family-Work Conflict

Basis of ConflictStrain Strain-Based, Work-Family Conflict Strain-Based, Family-Work Conflict

Behavior Behavior-Based, Work-Family Conflict Behavior-Based, Family-Work Conflict


recommending strategies to help employeescope with mergers and acquisitions (e.g.,Ivancevich, Schweiger, & Power, 1987).

From the limited empirical literature, themost reliable stress-related correlates ofmergers and acquisitions are employees’ feel-ings of anxiety, uncertainty, and job insecur-ity. Considering that even rumors of mergersand acquisitions evoke considerable specu-lation among employees as to how themerger or acquisition will be handled, thisfinding is not surprising. Ivancevich et al.(1987) recommended that organizationsengaging in mergers and acquisitions shouldmake an effort to communicate to employeesas much information as possible. Given thatmergers and acquisitions will continue inthe future, more research is needed on thestress-related implications of this importantorganizational activity.

Layoffs and Job Insecurity: Like mergersand acquisitions, layoffs became a fact oflife in the 1980s. A survey by the AmericanManagement Association indicated that66% of U.S. firms with more than 5,000employees reported reducing their work-force through layoffs in the late 1980s(Henkoff, 1990). This trend continued intothe 1990s and is likely to continue for yearsto come (Kozlowski, Chao, Smith, & Hed-lund, 1993).

Layoffs are somewhat different from theother stressors discussed in this chapter. Likethe other stressors, layoffs occur in an orga-nizational context, but their most directimpact is felt outside of the organizationalcontext (Leana & Feldman, 1992). It isimportant to note, though, that layoffs oftenimpact those who do not lose their jobs.Employees who survive layoffs may have feel-ings of vulnerability, and even guilt(Brockner, Grover, Reed, & DeWitt, 1992;Brockner, Grover, Reed, DeWitt, & O’Malley,

1987), and may experience an increase inworkload because the amount of work typi-cally stays the same.

How do layoffs impact those who losetheir jobs? The evidence is rather unequiv-ocal: Job loss is bad for one’s mental andphysical health. For example, research overthe years has shown that job loss is stronglyrelated to decreases in both psychologicaland physical well-being. This was shownrather clearly in a recent comprehensivemeta-analysis of over 437 effects sizes(McKee-Ryan et al., 2005).

Research has also shown rather clearlythat the negative impact of job loss is miti-gated by reemployment (Eden & Aviram,1993; Vinokur, van Ryn, Gramlich, &Price, 1991), and the manner in whichindividuals cope with job loss. For exam-ple, Wanberg (1997) found that unem-ployed individuals who employed activecoping techniques (e.g., they activelysearched for employment) fared better thanthose who avoided looking for work. Anobvious reason is: Active coping is likelyto speed reemployment.

Compared to job loss, relatively less isknown about the impact of job insecurity. Arecent meta-analysis, however, showed thatjob insecurity has a negative impact onemployee emotions and well-being (Sverke,Hellgren, & Naswall, 2002). With respect toemotional reactions, it is likely those whosurvive a layoff may respond with reducedtrust and commitment toward their employ-ing organization (Buch & Aldrich, 1991).Seeing fellow employees laid off may signifya potential breach of the implicit psycholog-ical contract between employees and theorganization (Morrison & Robinson, 1997).Layoff survivors may also find that their jobduties have been expanded, leading to feel-ings of being overworked (Byrne, 1988;Tombaugh & White, 1990).


Another issue may confront layoff sur-vivors: To remain employed, they may beforced to accept job transfers that requirerelocation. This may be very difficult, par-ticularly for dual-career families (Gupta &Jenkins, 1985). Children may also perceivejob insecurity in parents, and such percep-tions may have a negative impact. Forexample, Barling, Dupre, and Hepburn(1998) found that college students per-ceived their parents’ job insecurity, andthese perceptions were associated with atti-tudes about the world of work. Studentswho perceived high levels of job insecurityamong their fathers reported lower levelsof both the Protestant work ethic (i.e., hardwork pleases God) and humanistic workbeliefs. Given the prevalence of layoffs inrecent years, the long-term implications ofthese findings are troubling (see Comment

7.4) and suggest that more research on jobinsecurity is needed.

Emotional Labor: During the past 50years, the structures of the U.S. economyand the economies of other countries havechanged dramatically. Once dependent onheavy manufacturing, the service sectornow dominates the economies of theUnited States and many other nations. Thisshift undoubtedly has enormous implica-tions for many organizational phenomena,but it has clearly changed the content ofpeople’s jobs. As a result, many employeesare faced with a very different set of stres-sors than were their forefathers whoworked in factories a half-century ago.

The term emotional labor, initially coinedby Hochschild (1979, 1983), refers to theemotional demands that employees face on

COMMENT 7.4

THE IMPACT OF OCCUPATIONAL STRESS ON CHILDREN

IN A RECENT study, Barling, Dupre, and Hep-

burn (1998) found that parental job insecurity

had a negative impact on children’s beliefs

about the value of hard work as well as beliefs

about humanistic values in the workplace.

Stated differently, parents’ job insecurity ledtheir children to question the value of hard

work and to believe that the workplace was

rather cold and unforgiving.

This study is interesting for two reasons.

First, the occupational stress literature offers

very little data on the impact of stressors

beyond the person experiencing them. It seems

logical, though, that if an employee is experi-encing stressors at work, the effects of these

stressors will be felt by his or her spouse and

children. People cannot simply block out work

when they leave. Second, this study suggests

that children whose parents worry about job

security may develop a rather cynical attitude

toward the workplace and may question

whether there is any value in hard work. This

finding is troubling, because hard work and

commitment are needed for societies to be

productive.Perhaps, as time goes by and people do not

expect to be with one employer for a long

period of time, job insecurity will become less

of an issue, and children may be less impacted

by it. However, in the meantime, the study by

Barling et al. (1998) reminds us that children

are keen observers of their parents’ work lives,

and they form many long-lasting attitudesbased on these observations.

Source: J. Barling, K. E. Dupre, and C. G. Hepburn.

(1998). Effects of parents’ job insecurity on children’s

work beliefs and attitudes. Journal of Applied Psychology,

83, 112–118.


the job. Emotional labor can take manyforms, but two stand out as being particu-larly relevant to the study of occupationalstress. In the first form, employees are forcedto confront negative emotions. Examples ofthis would occur when a grocery store clerkmust interact with a dissatisfied customer, orwhen a physician must interact with a griev-ing family. In another relevant form of emo-tional labor, an employee may be forced tosuppress his or her true emotional state inorder to further the goals of the organization.The term that has been used to describe sucha situation is surface acting (Grandey, 2000).Many occupations have ‘‘display rules’’ thattell the employee the appropriate emotion todisplay to customers or clients (Ekman,1973). Employees who work directly withthe public encounter this type of situationevery day. For example, a waiter at a restau-rant may be having a bad day yet must bepleasant to customers because his jobdemands it. In contrast, a bill collector maybe feeling happy yet must act very tough inorder to get people to pay their bills.

Research on emotional labor is still rela-tively new, but there is evidence linking it tostress-related outcomes. The most commonstress-related outcomes associated with emo-tional labor have been poor work attitudesand increased emotional exhaustion(Ashforth & Humphrey, 1993; Brotheridge& Grandey, 2002; Grandey, 2003). Higheremotional labor requirements are associatedwith heightened feelings of emotional exhaus-tion. Ashforth and Humphrey, however,point out that the relation between emotionallabor and emotional exhaustion (and possiblyother strains) may be quite complex. Forexample, if the emotional display rules of ajob are congruent with how an employee isactually feeling, this may not be harmful:Being cordial to a customer will not be diffi-cult for a salesperson who is in a good mood.

Recent research has shown other factorsthat may impact employees’ reactions toemotional labor. Grandey, Fisk, and Steiner(2005) found that employees with high lev-els of job autonomy tended to react lessnegatively to emotional labor compared tothose with less control. These authors alsocompared the reactions of French and Amer-ican employees and found that in generalAmericans reacted more negatively to emo-tional labor. The authors of this study rea-soned that this occurred because the Frenchculture allows more personal control overthe expression of emotions as compared toAmerican culture.

Emotional labor appears to be a veryfruitful area for further occupational stressresearch, considering the large number ofservice sector employees. It also may be use-ful to broaden the scope of emotional laborresearch to include jobs outside of the serv-ice sector, because even these jobs may havediffering emotional display rules that impactemployees.

REDUCING THE IMPACTOF WORKPLACE STRESSORS

The focus of the chapter so far has been onunderstanding the relationship betweenspecific stressors and strains (i.e., under-standing what is stressful on the job). At thispoint, we shift the focus to examining ways touse that knowledge to improve the healthand quality of life of employees in the work-place. Organizational efforts to reduce theimpact of job-related stressors generally takeone of five forms: (1) stress managementtraining, (2) reduction of stressors, (3) alter-native work schedules and telecommu-ting, (4) family-friendly benefits, and (5)health and fitness programs. Each of theseapproaches is discussed in the followingparagraphs.


Stress Management Training

Perhaps the most common method of com-bating the effects of workplace stressorsis referred to as stress management trainingor, more commonly, stress management(Murphy, 1984). Stress management train-ing is designed to help provide employeeswith the resources necessary to cope moreeffectively when they are faced with stres-sors. Note that the purpose of stress manage-ment training is not to eliminate or evenminimize the stressors themselves; their exis-tence is basically taken for granted.

The content of stress management train-ing programs varies widely from organiza-tion to organization (Beehr et al., 2001;Bellarosa & Chen, 1997), but there are somecommon program components. For exam-ple, most programs have some educationalcomponent; that is, employees are providedwith information regarding the nature ofstress and its effects. It is also very commonfor such programs to include some form oftraining that is designed to help employeesreduce the physical effects of stressors. Inmany cases, this involves some form of relax-ation training, in which employees learn torelease the muscular tension that oftenaccompanies stressful encounters atwork. Among other interventions, in biofeed-back training, employees learn to controlphysiological responses (e.g., heart rate andrespiration) to stressors with the aid of phys-iological monitoring equipment (Smith,1993).

Another common component of stressmanagement training programs involvesteaching techniques that are designed to helpemployees alter their appraisals of the workenvironment. As noted at the beginning ofthis chapter, the manner in which the workenvironment is cognitively appraised is a keyfactor in determining whether it is consid-

ered a stressor. One commonly used methodof accomplishing this is Meichenbaum’s(1977) Stress-Inoculation Training, whichconsists of three distinct phases (see alsoDriskell, Salas, & Johnston, 2006).

In the first phase, participants are pro-vided with information about stress, as wellas a conceptual framework for understand-ing the phases of the treatment that willfollow. In the second phase, participantslearn and rehearse various coping strategies,which are typically taught in the form of‘‘self-statements.’’ The idea underlying thisphase is that people often engage in dysfunc-tional self-statements when they en-counter stressors, and these may ultimatelyexacerbate the effect of the stressor. As anexample, before making an important salespresentation, a person may say to himself orherself: ‘‘I’m no good at speaking in front ofother people . . . ’’ Needless to say, negativeself-statements make the situation moreuncomfortable. According to Meichenbaum(1977), it is possible to replace such negativeself-talk with more functional statements.For example, when making a sales presenta-tion, the person could instead learn to say:‘‘One step at a time; you can handle thesituation’’ (p. 155). This type of self-statement is likely to be more functional andcan have a calming effect on the individual.

The final phase of Stress Inoculation isreferred to as Application Training. In thisphase, participants learn to apply and usepositive self-statements in everyday situa-tions. This involves developing an awarenessof situations that are perceived as stressfuland of the negative self-statements thataccompany such situations. Once this isdone, participants learn to substitute self-statements that facilitate coping. One thingto note about this process: Consistent use ofthese positive self-statements may requireconsiderable practice and rehearsal.


In recent years stress management pro-grams, particularly in European countries,have become more comprehensive and widerinscope.According toAust andDucki (2004),many German companies have begun usingan approach to stress management known ashealth circles. In a typical health circle, em-ployees meet periodically (e.g., once a month)to discuss ways to reduce stress and improveoverall employee health; these discussionstypically lead to recommendations to topmanagement. Some readers will recognizethe similarity of this approach to qualitycircles, which are used by some manufactur-ing firms to improve product quality.Unfortunately, little research has been con-ducted to evaluate health circles, but this is aninnovative approach to stress managementthat may grow in the future.

Reduction of Stressors

Another approach to dealing with occupa-tional stress is to attempt to reduce the levelsof the stressors themselves. This approach ismuch less popular than stress managementtraining, most likely because it’s more diffi-cult. However, if an organization is trulyinterested in reducing the effects of stressors,this approach has greater potential forimproving employees’ well-being and qual-ity of life, rather than simply treating theeffects of stressors (Hurrell, 1995).

There may be many interventions thatwould reduce stressors—some may evenhave a preventative effect. Examples mightinclude (but are not limited to) redesigninga job to increase autonomy (Griffin, 1991;Hackman & Oldham, 1980), providing em-ployees with opportunities for greater par-ticipation in organizational decisionmaking (Lowin, 1968; Wagner, 1994),training managers to communicate moreeffectively with subordinates, and training

employees to use more effective conflict-resolution techniques.

The interventions previously mentioned,when implemented in organizations, are typ-ically not labeled as stress reduction efforts. Inmany cases, these interventions are offered astraining programs, or as part of a compre-hensive organizational development strategy(see Chapter 15). The result of many orga-nizational development interventions, how-ever, is a decrease in stressors and,consequently, an increase in employees’well-being. This point is illustrated veryclearly in a study conducted by Schau-broeck, Ganster, Sime, and Ditman (1993),in which a Responsibility Charting interven-tion in a university was evaluated. Responsi-bility Charting is an activity that is typicallyconducted during team-building interven-tions. It helps a work group clarify who isresponsible for what, within the group(French & Bell, 1995). One of the effects ofResponsibility Charting, found in this study,was a reduction in role ambiguity.

This approach to stress management hasalso been used quite successfully to reducestressors associated with the physical envi-ronment, or what are often termed ergonomicstressors. This includes things such as repet-itive motions, poorly designed equipmentthat leads to back pain, and computer screensthat lead to eye strain. This approach hasbeen used to improve the physical conditionsof employees in the meat packing industry(May & Schwoerer, 1994), as well as officeworkers (May, Reed, Schwoerer, & Potter,2004).

Alternative Work Schedulesand Telecommuting

In many cases, stressors are the result oftime-based conflicts. For example, an em-ployee may be required to be at home when


his or her child arrives home from school inthe afternoon. Unfortunately, this may con-flict with regular work hours. To help em-ployees cope with this form of conflict, anincreasing number of organizations areimplementing policies allowing alternativework schedules. An alternative work scheduleis defined as any deviation from the typical5-day, 8:00–5:00 or 9:00–5:00 work sched-ule. Given this rather broad definition, thereare numerous forms alternative work sched-ules can take. The two most typical forms areflextime and the compressed workweek.

In a typical flextime arrangement, allemployees are required to be present dur-ing some portion of the day. This timeperiod is referred to as the core hours.Beyond the core hours, employees areallowed to choose their own hours, as longas they work an 8-hour day and their choiceof hours is relatively consistent. To illus-trate, an organization could mandate thatall employees must be present between10:00 A.M. and 3:00 P.M. Beyond these corehours, an employee may choose to startwork at 7:00 A.M. and leave work at 3:00P.M. An employee also has the option ofstarting his or her workday at 10:00 A.M.and leaving work at 6:00 P.M.

Because many organizations haveadopted flextime arrangements (Johnson &Packer, 1987), a number of empirical studieshave evaluated the impact of this form ofalternative scheduling (e.g., Pierce & News-trom, 1982; Pierce, Newstrom, Dunham, &Barber, 1989). Most of these empirical stud-ies have found that employees respond pos-itively to flextime arrangements in terms ofjob satisfaction, performance, and absentee-ism. Pierce and Dunham (1992), however,suggested that flextime arrangements thathave a small number of core hours and allowfor changes in the pattern of hours are likelyto be most appealing to employees.

In a typical compressed workweekarrangement, employees work four 10-hourdays instead of five 8-hour days. In such anarrangement, an employee could work 4consecutive days (e.g., Monday throughThursday) or arrange to have a day off inthe middle of the week. Compressed sched-uling has been found to have positive effectsbecause it allows employees greater schedul-ing flexibility (Latack & Foster, 1985; Pierce& Dunham, 1992). One potential problemwith compressed scheduling, however, isthat the fatigue associated with workinglonger days may offset the increased flexibil-ity (Goodale & Aagaard, 1975). According toPierce and Newstrom (1992), the best way toavoid this problem is to implement the com-pressed schedule so that employees work4 days and then have 4 days off. The key tomaking compressed scheduling work is toallow employees enough time to recoverfrom the longer days.

Because of advances in communicationtechnology, many organizations haveallowed employees to engage in telecommut-ing arrangements. In a typical telecommutingarrangement, an employee is allowed to do aportion of his or her work at home. It hasrecently been estimated that in the UnitedStates alone there are over 28 million tele-commuters, and this figure is likely to rise(International Telework Association andCouncil, 2001).

Given the relative recency of telecom-muting, there is not a great deal of empiricalresearch on its effects. Interestingly, though,research that has been done suggests thattelecommuting may have both positive andnegative effects. On the positive side, tele-commuters may be better able to preventwork demands from interfering with familydemands (LaPierre & Allen, 2006). Unfortu-nately, however, telecommuters may be moresubject to nonwork interruptions, and this is


particularly true for employees with largefamilies (Golden, Veiga, & Simsek, 2006).Future research on telecommuting will likelyfocus on ways to maximize its positiveeffects, while minimizing the negative.

Family-Friendly Benefits

To help employees cope with the often con-flicting demands of work and family, manyorganizations offer what have been termedfamily-friendly benefits. There is no standarddefinition of what constitutes a family-friendly benefit. Generally, these benefitsare specifically designed to help employeesbalance the demands of work and family.Typical family-friendly benefits include flex-ible spending accounts, child care and eldercare referrals, part-time work options, andon-site day care facilities. Both flexible workschedules and telecommuting, which werediscussed in the previous section, are alsoconsidered family-friendly benefits in manycases.

Unfortunately, a widely held misconcep-tion about family-friendly benefits is thatthey are exclusively for women. Such bene-fits may be more salient to women (Kossek &Ozeki, 1998), but they are designed to bene-fit both men and women. In fact, in a surveyof human resources executives, Milliken,Martins, and Morgan (1998) found that thepercentage of females in an organization wasnot indicative of whether an organizationoffered family-friendly benefits. This studyfound that the best predictor was whetherthe executives felt that work–family issueswould impact organizational effectiveness.

There is considerable testimonialevidence on the value of family-friendlybenefits, but methodologically rigorous eval-uations have been rare. Thomas and Ganster(1995), for example, conducted a study of thestress-related impact of family-friendly bene-

fits among hospital employees. They foundthat those who worked in organizations offer-ing family-friendly benefits reported higherlevels of job satisfaction and lower levels ofdepression and somatic complaints. Theywere also found to have lower cholesterolthan employees working in organizations thatdid not offer such benefits. This study alsoshowed that family-friendly benefits, partic-ularly flexible schedules, have a positiveimpact because they enhance employees’ per-ceptions of control and reduce feelings ofwork–family conflict.

One final point—one that complicatesevaluation of the impact of family-friendlybenefits—must be considered: The imple-mentation of such benefits is often at thediscretion of individual managers. Further-more, Casper, Fox, Sitzmann, and Landy(2004) found that supervisors were notequally aware of the existence of family-friendly benefits and, as a result, varied inthe number of employees referred to them.As a result, the mere existence of a family-friendly benefit does not guarantee that allemployees will have equal access. For exam-ple, an organization may institute a policythat makes it possible for employees toswitch from full-time to part-time status. Ifthis policy is at the discretion of individualmanagers, all employees may not have theopportunity to benefit. This raises the gen-eral point that perhaps the key to helpingemployees balance the demands of work andfamily does not lie in official solutions such asfamily-friendly benefits (Goff, Mount, &Jamison, 1990). Rather, the most importantfactor may be the flexibility, understanding,and compassion of individual managers(LaPierre & Allen, 2006). At a more generallevel, the overall culture of an organizationwith respect to work-family issues may alsobe more crucial than particular package offamily-friendly benefits (Cunningham, 2005).


Health and Fitness Programs

An increasing number of organizations areoffering a variety of programs designed toimprove employees’ physical health and fit-ness. Such programs can range from some-thing as simple as providing informationabout health-related topics to extensive on-site fitness facilities (O’Donnell, 1986). Formost organizations, the primary motivationfor offering health and fitness programs is toreduce employees’ healthcare costs (Falken-berg, 1987; Jex, 1991). Indeed, several stud-ies conducted over the years have shown thathealth and fitness programs do reducehealthcare costs (see Pelletier, 1991, for asummary). Another common reason is thatemployees who are healthy and physically fitare less likely to be absent due to illness. Aswith healthcare costs, there is empirical evi-dence showing that health and fitness pro-grams do indeed lead to reductions inemployee absenteeism (Cox, Shephard, &Corey, 1981; Kerr & Vos, 1993; Tucker,Aldana, & Friedman, 1990).

Some studies have also attempted to linkparticipation in health and fitness programsto outcomes such as psychological strain(e.g., anxiety, depression, job satisfaction).Compared to the studies examining health-care costs and absenteeism, evidence linkingfitness programs to psychological strain ismuch more equivocal. According to Jexand Heinisch (1996), this is largely due tomethodological flaws in the design of manystudies examining the impact of health andfitness programs. For example, many suchstudies do not employ control groups. Evenin studies that do employ control groups,fitness program participants often drop outprior to program completion, or participateat a very minimal level.

Perhaps the most accurate conclusion tobe drawn about health and fitness programs

is that they are useful for improving thephysical health of employees and, as a result,may lead to decreased absenteeism. Based onthe available empirical evidence, however, itis unclear whether participation in healthand fitness programs has a great deal ofimpact on other stress-related outcomes. Inthe future, methodologically sound evalua-tions of health and fitness programs will beneeded to provide more definitive evidenceon this issue.

CROSS-CULTURALOCCUPATIONAL STRESSRESEARCH

The majority of what we know about occu-pational stress comes from studies of em-ployees in the United States and, to a lesserextent, Great Britain, Germany, and Scandi-navia. As a result, very little evidence existsregarding basic issues such as whether occu-pational stress models generalize across cul-tures, whether cultural factors impact thestressors employees experience, and whetherthere are cultural differences in coping strat-egies. In this final section of the chapter, webriefly review research that has investigatedthese issues. Readers interested in a morecomprehensive coverage of cross-culturalissues in occupational stress should consultLiu and Spector (2005).

GENERALIZABILITY

Very little evidence exists regarding the gen-eralizability of occupational stress theories.This is not a trivial issue, given that mostoccupational stress theories have been devel-oped in the United States or other Westerncountries. Xie (1996) examined this issue bytesting Karasek’s (1979) Demands–Controlmodel in the People’s Republic of China(PRC). According to Xie, it would be tempting


to predict that Karasek’s model would notapply in a collectivist nation like the PRC,since this model focuses on personal control.However, as Xie points out, there are cleardifferences between blue- and white-collarworkers within the PRC. Blue-collar workers‘‘ . . . are generally less educated and lessexposed to Western influences. Therefore,they are more likely to maintain the tradi-tional values which impede the desire ofindividuals for personal control’’ (p. 1,600).White-collar workers, on the other hand,have had greater exposure to Western values,one of which is valuing personal control.These individuals, compared to blue-collarworkers, have also benefited much morefrom recent economic changes in the PRC.

Based on a sample of 1,200 respondents,the interaction between demands and con-trol that would be predicted by Karasek’smodel was found only for white-collar em-ployees for most of the outcomes in the study.This supports Xie’s hypothesis and suggests,more importantly, a potential limitation onthis very popular occupational stress model.Recall that Schaubroeck and Merritt (1997)also found that this model was supported onlyfor those with high self-efficacy.

One potential reason for this finding isthat the meaning of control may differbetween Chinese and Western cultures. Ac-cording to Spector, Sanchez, Siu, Selgado,and Ma (2004), personal control beliefs incollectivist cultures such as China do notcenter on the person exerting direct controlover the environment. Rather, in collectivistcultures such beliefs are focused on (a)adapting one’s self to the environment and(b) exercising control through one’s relation-ships with others. If this is true then many ofthe control measures used by occupationalstress researchers, which center on personalcontrol over the environment, may not beappropriate for collectivist cultures.

STRESSORS EXPERIENCED

Another important issue that has lent itself tocross-cultural research is whether there arecultural differences in the perceptions ofstressors. This issue was addressed in across-cultural study of role stressors amongmanagers from 21 nations, conducted byPeterson et al. (1995). These authors foundthat perceptions of role stressors (ambigu-ity, conflict, and overload) varied consider-ably across nations. They also found thatlevels of role stressors could be predictedfrom characteristics of different nationalcultures. Perceptions of role stressors dif-fered with respect to power distance (degreeof segregation by levels of power), level ofmasculinity, degree of individuality, anddegree to which individuals try to decreaseuncertainty.

For example, managers from nations lowin power distance (e.g., industrialized, West-ern countries) reported high levels of roleambiguity and low levels of role overload.The pattern of results was exactly the oppo-site in nations characterized as high onpower distance (e.g., Latin American andFar Eastern countries). This finding suggeststhat Western managers may not have prob-lems with the sheer volume of work but maybe uncertain about their responsibilities.Non-Western managers, in contrast, maybe clear about their responsibilities but seethe sheer volume of work as a stressor.

These findings, according to Petersonet al. (1995), suggest that ‘‘Role conflict,ambiguity, and overload contain a core ofmeaning wrapped up in the nature of formalrelationships within formal organizations’’(p. 447). The mistake often made in occupa-tional stress research is to assume that themeaning of organizational events is consis-tent across cultures. In the future, more ofthis type of cross-cultural comparative

Stressors Experienced 227

research may reveal other important differ-ences in stressors across cultures.

A second example of cross-cultural occu-pational stress research was provided in astudy by Van De Vliert and Van Yperen(1996) in which cross-national comparisonsin role overload were examined. Theseauthors contended that the cross-national dif-ferences in role overload reported by Petersonet al. (1995) could be explained, at least inpart, by cross-national differences in ambienttemperature. That is, nations characterized aslow in power distance by Peterson et al. werelocated in geographical areas in which theambient temperature tended to be compara-tively higher. They proposed that differencesin ambient temperature, rather than in powerdistance, could have led to differences inreports of role overload.

These authors reanalyzed the data fromPeterson et al. (1995), along with two othercross-cultural data sets. It was found, aspredicted, that controlling for ambient tem-perature eliminated the relationship betweenpower distance and role overload. Thus,these authors concluded that the relation-ship between power distance and role over-load might be due entirely to cross-nationaldifferences in ambient temperature. Thesefindings suggest that certain cultural charac-teristics may be determined, to a certaindegree, by climate, and that such character-istics may then impact organizations.

More recently, Liu (2003) examined dif-ferences in stressful incidents between em-ployees from China and the United States.She found that employees from both coun-tries reported many common stressors, butthere were also differences. For example,compared to the employees from the UnitedStates, Chinese reported mistakes at workand evaluations as greater sources of stress.In contrast, employees from the UnitedStates reported lack of personal control as a

source of stress more frequently than theChinese employees.

COPING WITH STRESS

Coping represents the manner in which em-ployees attempt to adapt to the stressors theyexperience. While people may cope withstressors in numerous ways (Latack & Hav-lovic, 1992), in general people either try todo something about the stressor they areexperiencing, or they simply try to adapt toit. Given the apparent differences in themeaning of personal control between indi-vidualistic and collectivist cultures (Spectoret al., 2004), one might expect that therewould be cultural differences in how directlyemployees confront stressors.

Research does support this, at least indi-rectly. For example, employees in the UnitedStates have been shown to confront conflictmore directly than Taiwanese (Trubisky,Ting-Toomey, & Lin, 1991) and Arab Mid-dle Eastern employees (Elsayed-Ekhouly &Buda, 1996). In a similar vein, Liu (2003)found that the American and Chinese em-ployees differed in the types of conflicts theyhad with others. Americans reported moredirect conflicts (e.g., being rude to others),whereas the conflicts reported by the Chi-nese were more indirect (e.g., doing thingsbehind people’s backs).

In summary, this section has provided avery brief sampling of recent cross-culturaloccupational stress research. While moreresearch obviously needs to be done, it is clearthat cultural factors may play an impor-tant role in the stress process. Many of thetheories, models, and variables that wereonce thought to be universal may not be.As organizations continue to expand glob-ally, the importance of further cross-culturaloccupational stress research cannot be over-stated.


CHAPTER SUMMARY

This chapter examined occupational stress, atopic that is becoming increasingly impor-tant both to organizations and to society as awhole. The roots of occupational stressresearch can be traced to the early twentiethcentury, although the first large-scaleresearch program did not begin until the1960s. Furthermore, a great deal of occupa-tional stress research has been conductedonly within the past 25 years.

Occupational stress researchers have, attimes, struggled with terminology. This islargely due to the fact that the study ofoccupational stress has always been interdis-ciplinary in nature. Organizational psychol-ogy has certainly contributed, but importantcontributions have also been made by thosein medicine, clinical psychology, and engi-neering psychology.

Several models of the occupational stressprocess have been proposed over the years,and four of these were discussed. The ISRmodel is probably the most influentialbecause it has guided a great deal of occupa-tional stress over the years. Other modelsdiscussed included Karasek’s Demands–Control Model, Beehr and Newman’s FacetModel, the P–E Fit Model, and finally theEffort-Reward Imbalance model.

Stressors represent things in the job ororganization that require some type of adap-tive response on the part of employees. Themost commonly studied stressors are thoseassociated with employee roles, although re-searchers have also examined workload,interpersonal conflict, organizational con-straints, and perceived control. Stressors thathave increased in importance in recent yearsinclude work–family conflict, mergers andacquisitions, job insecurity, and emotionallabor.

Organizations wishing to reduce theimpact of stressors have generally tried todo so in five different ways. The most com-mon method is the development of stressmanagement training programs that teachemployees how to cope more effectively withstressors. Other less common methodsinclude reducing stressors, offering alterna-tive work schedules, making family-friendlybenefits available, and offering employeeshealth and fitness programs.

Occupational stress has clearly laggedbehind in cross-cultural research. Recently,however, there appears to be some progressin this area. For example, research conductedin the People’s Republic of China hasshown that the Demands–Control Modelmay not apply to all cultures, most likelydue to differences in the meaning of personalcontrol. It has also been shown that culturemay have an impact on the types of stressorsthat are perceived, as well as the copingmethods used by employees. A great dealmore cross-cultural research is needed,however.


Aust, B., & Ducki, A. (2004). Comprehensivehealth promotion interventions at the work-place: Experiences with health circles inGermany. Journal of Occupational HealthPsychology, 9, 258–270.

Judge, T. A., & Colquitt, J. A. (2004). Orga-nizational justice and stress: The mediatingrole of work-family conflict. Journal ofApplied Psychology, 89, 395–404.

Lapierre, L. M., & Allen, T. D. (2006).Work-supportive family, family-supportivesupervision, use of organizational benefits,and problem-focused coping: Implications



GARY ADAMS AND THE CONSIDERATION OF WORK IN THE CONTEXT

OF LIFE

It would be nice to be able to say that since

deciding to become and I/O psychologist (way

back when I was a sophomore in college) I

always wanted to study topics related to occu-

pational stress and health but that justwouldn’t be true. I was originally interested

in personnel selection. I had learned about the

notion of person—environment fit and the

idea that when there is a good fit, good things

tend to happen for people and the organiza-

tions where they work. I thought I would

make my career as a consultant helping orga-

nizations put that principle into practice. Iearned a master’s degree and then took a posi-

tion in human resources. I enjoyed that work

and like to think it went well but my mentor

inspired me to pursue a Ph.D.

As I went back to school one of the main

ideas I took from the work I had been doing, as

well as my time in the Army and other posi-

tions, is that the impact of work on the indi-vidual (and vice versa) extends well beyond the

office or factory door. Thus, while keeping in

mind that idea of person environment fit, the

major direction of my work shifted to a focus

on the relationship between work and non-

work. This can be seen in some of my research

on the relationship between work and retire-

ment, which I had originally thought of as trade-

off between work and leisure. It can be seen in

the research I’ve done on social support and therelationship between work and family. It can also

be seen in my research regarding the after

effects of both traumatic events (post-traumatic

stress disorder) and chronic stress at work. The

other change that took place while I was pursu-

ing the Ph.D. was a shift from wanting to pursue

consulting as my full-time employment to a

greater interest in becoming a full-time academi-cian. Again, this was something inspired by a

mentor along the way.

The ‘‘what have I learned’’ question is a

tough one. It may be better to ask, ‘‘what I am

still learning,’’ and then a few things come to

mind. First, work is just one part of a larger

puzzle called life and it’s important to people

and organizations for those pieces of the puzzleto fit together well. I know that sounds cliche

but larger contexts do matter. I also learned

that people often talk about happy and pro-

ductive workers but forget that neither of these

two is likely to occur when the worker (and/or

the work) is unhealthy. Regarding advice to

those who may be just beginning a career in I/

O, I would suggest learning all you can, espe-cially in the area of methodology. It is impor-

tant to find good, relevant research questions,

use good methods to study them, and then be

able to communicate the answer you find to

others. On a more personal level, I have found

that I am fortunate to work in a field that

provides a variety of opportunities and the

flexibility to pursue them. I have also foundthat it’s nice to be in field that is populated by so

many good people. Again, larger contexts

really do matter.

Gary A. Adams


University of Wisconsin-Oshkosh


for work-family conflict and employee well-being. Journal of Occupational Health Psychol-ogy, 11, 169–181.

Liu, C., & Spector, P. E. (2005). Internationaland cross-cultural issues. In J. Barling, E. K.Kelloway, & M. R. Frone (Eds.), Handbook ofwork stress (pp. 487–516). Thousand Oaks,CA: Sage.


Bond, J. T., Galinsky, E., & Swanberg, J.(1998). The 1997 National Study of TheChanging Workforce. New York: Familiesand Work Institute.

Brotheridge, C., & Grandey, A. (2002).Emotional labor and burnout: Comparingtwo perspectives of ‘‘people work.’’ Journalof Vocational Behavior, 60, 17–39.

DeFrank, R. S., & Ivancevich, J. M. (1998).Stress on the job: An executive update. Acad-emy of Management Update (1998) 12(3):55–66.

Hofmann, D. A., & Tetrick, L. E. (2003).The etiology of the concept of health: Impli-cations for ‘‘organizing’’ individual and orga-nizational health. In D. A. Hofmann and L.E.Tetrick (Eds.), Health and safetyin organizations: A multilevel perspective(pp. 1–28). San Francisco: Jossey-Bass.

International Telework Association andCouncil. (2001). Telework America report.Retrieved November 20, 2007, from http://www.workingfromanywhere.org/telework/twa2001.htm.

Peter, R., Geissler, H., & Siegrist, J. (1998).Associations of effort-reward imbalance at

work and reported symptoms in differencegroups of male and female public transportworkers. Stress Medicine, 14, 175–182.

Siegrist, J. (1996). Adverse health effects ofhigh-effort/low-reward conditions. Journal ofOccupational Health Psychology, 1, 27–41.

Siegrist, J. (2002). Effort-reward imbalanceat work and health. In P. L. Perrewe &Ganster, D. C. (Eds.), Historical and currentperspectives on stress and health (pp. 261–291). Amsterdam: JAI Elsevier.

Siegrist, J., Starke, D., Chandola, T., Godin,I., Marmot, M., Niedhammer, I., & Peter, R.(2004). The measurement of Effort-RewardImbalance at work: European Comparisons.Social Science & Medicine, 58, 8, 1483–1499.

Sulsky, L., & Smith, C. (2005). Work Stress.Belmont, CA: Thomson Wadsworth.

Thomas, L. T., & Ganster, D. C. (1995).Impact of family-supportive work variableson on work-family conflict and strain: Acontrol perspective. Journal of Applied Psy-chology, 80, 6–15.

Chapter Eight

Theories ofMotivationM

otivation is concerned with aquestion: Why do people ‘‘dowhat they do?’’ Whether werealize it or not, all of us are‘‘naıve scientists’’ who often

attempt to figure out the motives behindthe behavior of others. We may read a news-paper and wonder why a person committeda violent crime, or perhaps why an athletemaintained a consistently high level of per-formance during his or her career. Withinorganizational psychology, the study of em-ployee motivation represents one of themost important topics in the discipline,and there are several reasons for this. First,motivation is a key to understanding manyforms of behavior in organizations. Under-standing what motivates employees helps usto understand the dynamics underlying suchimportant behaviors as job performance,absenteeism, turnover, and even counter-productive behaviors.

Second, an understanding of the dynam-ics underlying various forms of behaviorenhances our ability to predict these samebehaviors. For example, if an organization’sleaders understand the motivation underly-ing performance, they can predict theiremployees’ future performance. This isimportant when organizations are initiallyselecting new employees, but it may also behelpful when current employees are beingconsidered for promotional opportunities.

In addition, understanding an employ-ee’s motivation for performing at work canallow leaders to structure the work environ-ment to encourage productive work behav-ior and discourage counterproductive workbehavior. For example, if an organization

knows that employees are highly motivatedby financial incentives, this knowledgecan be used to influence performancethrough the strategic use of raises. How-ever, if leaders realize that having controlover job performance is especially impor-tant for motivating employees, then thework environment can be structured toallow employees sufficient autonomy overwhat they do. All organizations, in one wayor another, attempt to influence employees’behaviors. Organizations that are armedwith a clear understanding of motivationare in a better position to influence employ-ee behavior than are organizations that lackthis knowledge.

233

This chapter provides an overview of thetheories of motivation that have been mostinfluential in organizational psychology. InChapter 9, we will examine how motivationtheories are applied within organizations toinfluence employee behavior. Before examin-ing specific motivation theories, however, wewill briefly review the important issue of howmotivation is defined and the classificationof theories that have been applied to workmotivation. As will become evident, this isimportant because different theories concep-tualize motivation in somewhat differentways.

DEFINING MOTIVATION ANDTHEORETICAL APPROACHES

Definitional Issues in Motivation

According to Kanfer (1990), motivation is ahypothetical construct—we cannot see it orfeel it. However, we can observe the effectsor by-products that are indicative of differ-ing levels of motivation. To use an analogy,motivation is a bit like gravity. We cannot seeor feel gravity, but its effects would becomevery clear if one were to jump out of awindow of a five-story building.

Pinder (1998) argued that motivationdetermines the form, direction, intensity,and duration of work-related behavior. Theform of behavior refers to the types of activ-ities an employee would choose to engage inat work. For example, some employees maychoose to focus on task-based performance,whereas other employees choose to investtheir efforts on building interpersonal rela-tionships and facilitating group cohesion.The direction of motivation refers to the spe-cific path employees take in accomplishingthe goals they set for themselves. For exam-ple, a nurse may have a predetermined

schedule for the various tasks he or she needsto accomplish within a given time frame, andthis schedule serves to guide behavior duringthe course of the day. The intensity com-ponent of motivation addresses the vigorand amount of energy put into goal-directed work performance. A market ana-lyst can approach the tasks he or she faceswith intense energy or detached resignation.Finally, duration refers to how long an indi-vidual pursues a particular course of direc-tion at work, perhaps persisting even whenfacing obstacles. In a related definition,Steers, Mowday, and Shapiro (2004) re-cently defined motivation as ‘‘factors or eventsthat energize, channel, and sustain humanbehavior over time’’ (p. 379).

By observing these dimensions of be-havior, we can draw some conclusions aboutthe influence of motivation on employees’behaviors. Based on Pinder’s proposition, amajor question for organizational psycholo-gists studying employee motivation is: Whatdependent variable should be studied inempirical research? As readers will see, com-mon dependent measures for theories ofmotivation include employees’ effort, choice,or, in some cases, persistence. In addition,researchers have often examined motivationas a predictor of well-being and perfor-mance. Although factors other than motiva-tion (e.g., ability, lack of critical resources)can certainly influence well-being and per-formance, researchers have shown that mo-tivation has important implications for thesedependent variables.

A final issue to consider in definingand understanding employee motivation isdetermining what forms of behavior orga-nizations wish to influence. This willbecome particularly important when appli-cations of motivation theories are describedin Chapter 9, but it is also important in

234 Theories of Motivation

understanding different motivation theo-ries. As will be shown, motivation theoriescan be distinguished in terms of whetherthey ultimately predict outcomes most rel-evant to performance, employee citizen-ship, or simply the propensity to maintainorganizational membership.

Theoretical Approachesto Motivation

Given the importance of motivation in psy-chology, numerous theories of human moti-vation have been developed over the years.Many of these, however, either were notdeveloped to explain behavior in the work-place, or are simply difficult to apply in thework domain. The theories that are coveredin this chapter have been developed specif-ically to explain employee motivation orhave been applied successfully to the studyof work behavior. It is possible to place moti-vation theories into five general categories:

1. Need-based theories explain work moti-vation in terms of the extent to whichemployees satisfy important needs in theworkplace.

2. Cognitive process theories emphasize thedecisions and choices that employeesmake when they allocate their efforts.

3. The behavioral approach emphasizesapplying principles of learning to thework environment.

4. Self-determination theory emphasizesthe importance of the distinction bet-ween motivation being driven by extrin-sic factors (e.g., need for approval,money) versus intrinsic factors (e.g.,interest in the work itself).

5. Job-based theories place the source ofmotivation primarily in the content ofjobs that employees perform.

The first four categories described repre-sent theories of motivation that were devel-oped in the broader field of psychology andwere subsequently applied to motivationat work. The final category covers theoriesthat were developed within the specific con-text of the work environment. When takentogether, these separate theories provide acomprehensive explanation of employee mo-tivation at work.

NEED-BASED THEORIESOF MOTIVATION

By definition, a need indicates some deficientstate within an individual. We know, forexample, that humans need things such asoxygen and water in order to survive. Psy-chologists have also proposed that humanshave psychological needs that serve to drivemuch of human behavior. Murray (1938),for example, was one of the first psycholo-gists to propose a systematic taxonomy ofhuman needs (e.g., achievement, power,affiliation, intimacy). He proposed that theseneeds are evoked by different stimuli in theenvironment, and subsequently drive behav-ior (see Comment 8.1).

Maslow’s Need Hierarchy

Building on the work of Murray (1938),Maslow (1943) proposed his well-knownNeed Hierarchy as an explanation of the forcesdriving human behavior. It is important tonote that Maslow’s theory was not designedspecifically to explain behavior in the work-place. Rather, Maslow attempted to create auniversal theory that would explain the driv-ing forces behind all purposeful behaviors. Itis also important to consider that Maslow dev-eloped his Need Hierarchy based largely onclinical observations rather than systematic


empirical research. Despite these caveats,Maslow’s theory has become quite influentialin a variety of areas of psychology, includingorganizational psychology.

Figure 8.1 presents the five need levelsthat comprise Maslow’s Need Hierarchy. Atthe bottom of the hierarchy are physiologicalneeds. This level represents the needs forfood, oxygen, and water—attributes thatare physiologically necessary to sustain life.These needs are at the lowest level becausethey will motivate behavior only if they areunsatisfied. Thus, a person who lacks suchbasic necessities will be motivated primarilyto obtain them. The closest most of us havecome to being motivated by physiological

needs is a late-night excursion to a fast-foodrestaurant. In some parts of the world, how-ever, basic physiological sustenance is one ofthe major forces driving not only work be-havior, but many other behaviors as well.

When physiological needs are satisfied, aperson then ‘‘moves on’’ to the next level inthe hierarchy: safety needs, which includethings such as shelter from the elementsand protection from predators. As with allneeds, Maslow proposed that safety needswould motivate behavior only to the extentthat they are unmet. Compared to physio-logical needs, it is a bit easier to illustratehow safety needs may motivate work behav-ior. For example, work may allow a person to

COMMENT 8.1

DO HUMANS REALLY HAVE PSYCHOLOGICAL ‘ ‘NEEDS’ ’?

THEORIES OF MOTIVATION that emphasized need

satisfaction once strongly dominated the field

of organizational psychology. Over time, how-

ever, need theories have generally fallen out of

favor. At the present time, they are considered

more for their historical value than anything

else. One obvious reason for this demise isthat need theories have not stood up well

under empirical scrutiny. Another reason is

that the concept of psychological needs is

rather controversial.

Although it is rather easy to make a case

that humans have physiological needs, the idea

that psychological needs exist is more debat-

able. For example, it could be argued thatalthough things such as social belonging are

valued, people typically do not suffer dire

consequences if they have less social contact

than they desire. On the other hand, it has been

shown that social isolation may lead to certain

forms of psychopathology and contribute to

developmental disabilities (e.g., Bowlby,

1973). From this point of view, one could

mount a rather convincing argument that psy-

chological needs do exist.

Baumeister and Leary (2005) have recently

argued that all humans have a need to belong to

social groups, and that an individual’s self-

esteem is a subjective indicator of the extent

to which an individual is securely included invalued groups (Leary, Tamber, Terdel, &

Downs, 1995). These authors have argued that

when the need to belong is thwarted many

negative consequences occur. Haslam and his

colleagues (2004) have also argued that the

work group can be an important part of an

employee’s self-concept and therefore moti-

vate behavior at work so that the employeemaintains secure inclusion in the work group

or organization.

Source: J. Bowlby. (1973). Separation: Anxiety and anger.

New York: Basic Books. Baumeister, R.F., & Leary, M.R.

(1995). The need to belong: Desire for interpersonal

attachments as a fundamental human motivation. Psycho-

logical Bulletin, 117, 497–529.


provide his or her family with adequatehousing in a safe neighborhood, as well asthe security of having a guaranteed retire-ment income.

If safety needs are satisfied, the next levelthat becomes salient is love and belongingneeds. This level represents the need to formmeaningful social relationships with othersand the desire to feel a sense of belonging.Although love needs may be satisfied in avariety of ways, for most people work repre-sents an important context for satisfying thistype of need. People often develop closesocial ties with coworkers and derive con-siderable satisfaction from these affiliations.These social ties may also help employees tocope with many negative aspects of the workenvironment (Cohen & Wills, 1985).

After love needs are met, the next levelthat becomes important in motivating behav-ior is esteem needs. Esteem needs are linked toa desire to feel a sense of competence andmastery. As with social/belongingness needs,esteem needs may potentially be satisfied in

a variety of ways. For example, one may feel asense of esteem or competence by being agood parent, cultivating a productive garden,or having a neat and clean house. For manypeople, the workplace represents a primarysetting in which esteem and competenceneeds are satisfied (Baumeister, 1991). Forexample, an accountant may feel a sense ofpride and accomplishment when he or shecompletes a client’s tax return quickly andaccurately.

The highest need level that can bereached in Maslow’s hierarchy is self-actualization. According to Maslow (1943),to self-actualize is to realize one’s potentialand fulfill one’s capacities. Maslow pointedout that few people ever completely ‘‘satisfy’’the need for self-actualization. Compared tothe other levels of needs, self-actualization isa bit more difficult to describe becauseresearchers differ considerably in how theydefine the variable. Nevertheless, it is cer-tainly possible that work could provide theopportunity for self-actualization. A teacher,for example, may feel actualized by educat-ing future generations.

When viewed as a complete theory,Maslow’s Need Hierarchy is certainly intui-tively appealing and represents an insightfulstatement about human nature. The theory,however, initially fared very poorly as a pre-dictor of work behavior (Locke & Henne,1986). Empirical research did not supportthe number of levels in the theory or thenotion that lower levels in the hierarchy mustbe satisfied before higher-level needs willmotivate behavior (e.g., Hall & Nougaim,1968). In addition, reviews of motivationtheory in the 1990s included Maslow’stheory primarily for historical value andbecause it has served as the basis for moreelaborate theories of work motivation (e.g.,Ambrose & Kulik, 1999; Austin & Vancou-ver, 1996; Kanfer, 1990). However, Latham

FIGURE 8.1Maslow’s Need Hierarchy

Safety

Love

Esteem

Self-Actualization

Physiological


and Pinder 2001 recently argued thatMaslow’s theory has seen a resurgence inthe study of work motivation. Ronen(2001) studied employees from 15 differentcountries and found support for the discrim-ination among levels in Maslow’s hierarchy.Importantly, Maslow’s theory has alsoinspired other researchers to consider theroles that needs play in employee motivation.

ERG Theory

The most direct descendant of Maslow’sNeed Hierarchy was Alderfer’s (1969) ERGTheory of motivation. The acronym ERGstands for existence, relatedness, and growth.Essentially, Alderfer collapsed Maslow’s fiveneed levels into three. Existence encom-passes both the physiological and thesafety/security needs from Maslow’s theory.Relatedness corresponds to the social/belongingness level in Maslow’s theory.Growth represents the esteem and self-actualization levels from Maslow’s theory.

ERG theory also deviates from the NeedHierarchy in other important ways. UnlikeMaslow’s theory, ERG Theory allows for thepossibility that needs do not have to operate ina strict hierarchical fashion (Alderfer, 1969).For example, an artist may be trying toscratch out a living and, at the same time,to achieve his or her artistic potential. ERGalso allows for the possibility that people mayregress if their needs at one level are notsatisfied. Suppose an artist fails to achievehis or her potential. According to Alderfer,such a person may become more focused onsatisfying ‘‘lower level’’ needs. The artistmay become focused on making friends andconnecting with people socially. Recall thatMaslow’s theory is focused only on moving upthrough the hierarchy of needs. It says littleabout situations in which needs are thwarted.

When it was first proposed, ERG Theorywas seen as an improvement to Maslow’stheory, but it has fared only slightly betterin terms of empirical support. Alderfer’s(1969) original work supported the theory,but subsequent tests have offered only mixedsupport (e.g., Wanous & Zwany, 1977).However, like Maslow’s theory, ERG Theoryhas served as a foundation for future theoriesin which need satisfaction is proposed to be acentral component.

Need for Achievement Theory

A third need-based theory of motivation,Need for Achievement Theory (Atkinson,1964; McClelland, 1965), has proven to besomewhat more useful than the two previ-ously discussed. Need for Achievementdraws its historical roots from the early workof Murray (1938). However, rather thanfocusing on multiple needs, the emphasishas been primarily on the Need for Achieve-ment in explaining differences between peo-ple in goal-directed behavior.

The work of McClelland and others hasidentified some consistent distinguishingcharacteristics of those who have a high needfor achievement. For example, they tend tochoose moderate levels of risk, have a strongdesire for knowledge of results or feedback,and have a tendency to become veryabsorbed in their work. In the work environ-ment, this may be reflected in the tendencyof such individuals to set moderately difficultperformance goals, seek jobs that readilyprovide performance feedback, and perhapswork long hours.

McClelland also proposed that Need forAchievement has consequences for entiresocieties as well as for individuals. For exam-ple, in his book The Achieving Society,McClelland (1961) proposed that societies


differed in terms of their absolute levels ofNeed for Achievement, and that such differ-ences may explain differences in economicgrowth. Thus, one way to promote economicdevelopment in poor countries, according toMcClelland, is to promote higher levels ofNeed for Achievement among native orindigenous populations.

Compared to the other two need-basedtheories, Need for Achievement Theory isclearly narrower in focus. Rather than tryingto account for all needs and all forms ofbehavior, this theory focuses on only oneneed and a very specific form of behavior(e.g., achievement). This tighter focus makesNeed for Achievement Theory somewhatmore useful in organizations. For example,if a manager knows that one of his or hersubordinates has a high need for achieve-ment, this knowledge may be useful in deter-mining job assignments and the frequencywith which performance-related informationshould be communicated.

The narrow focus of Need for Achieve-ment Theory is also problematic in somerespects. McClelland acknowledged that fac-tors other than Need for Achievement influ-ence behavior, but then does not go on toindicate how Need for Achievement isrelated to these additional variables. How-ever, it is important to keep in mind McClel-land’s narrow focus and that he was notattempting to develop a full-blown theoryof human motivation.

COGNITIVE PROCESSTHEORIES OF MOTIVATION

Another way that we can view employeemotivation is in terms of the cognitive proc-esses underlying motivation. Cognition, ofcourse, means thought. What are some ofthe thought processes that go along with

employee motivation? As readers will see inthe theories described in this section, em-ployees make judgments about how fairlythey are being treated, choose where theywill direct their efforts, and are able to antici-pate future rewards associated with differentlevels of goal accomplishment. An under-standing of these cognitive processes pro-vides a great deal of insight into employeemotivation.

When one looks at the history of psy-chology, the closest the field has ever come tohaving what might be described as a domi-nant paradigm was during the 1960s and1970s, when behaviorism was at the heightof its popularity. Thought processes such asdecision making and choice were not con-sidered under the realm of psychology,because these could not be directly observed.

This view began to change in the late1970s, and the changes ushered in whatmany have referred to as the Cognitive Revo-lution in psychology. During this period,psychologists began to focus on the thoughtprocesses underlying phenomena such asproblem solving, choice, and even psychopa-thology. Another factor that facilitated thedevelopment of cognitive process theorieswas the growth in computer use. This isimportant because with this revolution camean increasing trend, particularly in memoryresearch, to equate human informationprocessing with computer information proc-essing. As readers will notice, the ‘‘mind ascomputer’’ metaphor is evident, particularlyin the more recent cognitive process theoriesof motivation.

Equity Theory

According to Homans (1958), humans tendto view social interactions as being much likeeconomic transactions. That is, we tend to


view relationships with others, as well astransactions with institutions (e.g., work,government), in terms of what we give andwhat we receive. Based on this notion, socialexchange theory was developed to explainhow we weigh and balance what we giveand receive from our interactions with others(Kelley & Thibaut, 1978).

Equity Theory is a type of social exchangetheory that focuses on how people determinethe fairness of social exchanges (Adams,1965). Although Equity Theory can reallybe applied to any form of social exchange,in describing this theory we focus on thework context. A basic assumption of EquityTheory is that employees bring to the work-place what they perceive to be a number ofinputs. Given that Equity Theory focuses oncognition, an input is essentially anything anemployee decides it is. Job-relevant inputswould include factors such as a person’sacademic credentials, years of prior experi-ence, and job-related skills, as well as thelevel of effort given to his or her employer.

The other important component ofEquity Theory is outcomes. Outcomes rep-resent those things that an employee feels heor she is receiving from the employmentrelationship. The most tangible of these ismonetary compensation, but outcomes mayalso include intangibles such as praise fromone’s supervisor, feelings of accomplish-ment, or even feelings of camaraderie amongone’s coworkers. Like inputs, outcomes arecognitive representations, and thus may dif-fer from employee to employee.

According to Adams (1965), employeescognitively compare their ratio of inputs-to-outcomes to the perceived ratio of somecomparative standard. A comparative stand-ard could be another employee employed inthe same job in the same organization, some-one performing a similar job in a differentorganization, or perhaps even the focal em-

ployee at a different point in time. If anemployee perceives that the ratio of his orher inputs to outcomes is equal to the ratio ofthe comparative other, a state of equity is saidto exist. This means that the employee is rea-sonably satisfied with the current exchangerelationship with his or her employer. Whenthese ratios are different, however, a state ofinequity is said to exist. In this case, the em-ployee is not satisfied with the current stateof the exchange relationship, and therefore ismotivated to bring it back into balance.

According to Equity Theory, the mostcommon form of inequity is referred to asunderpayment. This occurs when the ratio ofinputs to outcomes is perceived as less favor-able than the comparative standard. Forexample, if an employee perceives that heor she is working much harder than a fellowemployee who is paid the same salary, thismay engender feelings of underpayment. Ac-cording to Adams (1965), an employee canuse a number of strategies in an attempt torestore equity when feelings of underpay-ment exist. These strategies are summarizedin Table 8.1.

One way for an employee to restoreequity would be to attempt to increase hisor her outcomes. In the previous examplegiven, the employee could go to his or hersupervisor and ask for a raise in order tocompensate for his or her higher level ofwork effort. This may have the effect ofrestoring equity if the employee is successful,but it may also be risky. If the employee’srequest for a raise is denied, he or she mayfeel worse than before. This is especially trueif the employee regards the effort required torequest the raise as an additional input.

In a second strategy to restore equity, theemployee may reduce inputs so that the ratiobecomes equal to that of other workers, andthe underpayment is rectified. For example,an employee may reduce effort to a level that


is perceived as commensurate with out-comes. This strategy also carries some degreeof risk. Reduced effort on the part of anemployee may be perceived negatively by asupervisor or coworkers. This may, in turn,result in even fewer outcomes for the em-ployee.

A third strategy is to cognitively adjustone’s perceptions of inputs and outcomes ina way that restores equity. For example,employees may cognitively reevaluate theiroutcomes and decide that they are morefavorable than was first thought. For exam-ple, an employee who perceives he or she isbeing paid poorly may consider that his orher cost of living is lower than employeesin other cities. The employee could alsoreevaluate his or her inputs and decide thatthey do not have as much value as firstthought, or perhaps decide that there areadditional outcomes that were not consid-ered initially. The inputs and outcomes ofthe comparison standard may also be cogni-tively adjusted in order to bring the tworatios into balance.

Of all the strategies listed in Table 8.1,cognitive adjustment of the perceptions ofinputs and outcomes clearly requires theleast amount of effort on the part of anemployee, and it is the least risky. For exam-ple, the employee does not have to make aneffort to increase his or her outcomes, and

does not incur the risk that goes along withreducing his or her inputs. A potential draw-back with this strategy is that it may result inan employee being taken advantage of. Thereare situations in organizations where peopleare treated unfairly, and cognitively adjust-ing one’s perceptions does not change unfairtreatment.

The fourth possibility listed in Table 8.1is that an employee who perceives under-payment may change his or her comparativestandard so that the ratio is perceived morefavorably. For example, if the authors wereto use a professional baseball player as acomparative standard in making equityjudgments, this would undoubtedly leadto strong feelings of inequity, at least withregard to salary. On the other hand, chang-ing the comparative standard to ‘‘associateprofessors in psychology departments’’would provide a greater chance of restoringequity. Keep in mind, though, that evenwithin the same occupation or profession,multiple comparisons may be possible. Forexample, among academic I/O psychologists,distinctions can be made between those inpsychology departments and those in busi-ness schools. Even within psychologydepartments, a distinction can be madebetween those teaching in doctoral programsand those employed at the master’s level (seeComment 8.2).

TABLE 8.1A Summary of the Mechanisms That Can Be Used to Restore Equity

Mechanism Example

Increasing outcomes Asking one’s supervisor for an increase in salary

Reducing inputs Decreasing the level of effort devoted to work tasks

Cognitive adjustments Changing the perception of the value of one’s inputs or outcomes, to restore equity

Changing the

‘‘comparative standard’’

Choosing a different person to compare the ratio of inputs to outcomes

Leaving the field Obtaining a job that provides a more favorable ratio of inputs to outcomes


A final way that an employee may re-spond to underpayment inequity wasdescribed by Adams (1965) as exiting thefield of play, or withdrawing from theinequitable exchange. In an employment set-ting, this would typically take the formof employee turnover, although it couldtake more subtle forms. For example, anemployee who is feeling inequitably treatedmay psychologically withdraw from theorganization. This may simply involve veryminimal participation or reduced feelings oforganizational commitment (see Chapter 5).As Adams pointed out, leaving the field is astep that is typically taken after other meth-ods of resolving inequity are exhausted. Incertain cases, however, this may be an em-ployee’s best option. For example, if there islittle chance that equity can be restored, itmay be best for an employee to seek otheremployment.

Recall that Equity Theory also proposesthat feelings of inequity will arise when theratio of a person’s inputs to outcomes ismore favorable when held to the compara-tive standard. This is referred to as overpay-ment. Given that feelings of underpaymentlargely represent feelings of unfairness orinjustice, how then can we describe the qual-ity of feelings of overpayment? According toAdams (1965), feelings of overpayment areuncomfortable, as are feelings of underpay-ment. Qualitatively, feelings of overpaymentare probably best described as guilt ratherthan unfairness.

According to Adams (1965), an employ-ee experiencing overpayment may use thesame basic strategies that can be used torestore equity when feelings of underpay-ment exist. For example, one could increaseone’s inputs to make them proportional toone’s outcomes, attempt in some way to

COMMENT 8.2

SALARY EQUITY IN PROFESSIONAL SPORTS

ONE AREA IN which the effects of Equity Theory

can be observed quite readily is professional

sports. It’s almost comical, for example, to see

a professional athlete who is being paid $5

million per year complain bitterly that he is

being treated unfairly because he is not being

paid $10 million. Most of us would be ecstaticto be paid even a fraction of either of those

amounts. However, if one keeps in mind the

comparative standard used by a professional

athlete, such feelings of inequity become

much easier to understand. More specifically,

highly paid professional athletes compare

their earnings to other highly paid profession-

al athletes of the same stature. When thesetypes of comparisons are made, the fact that

one is a multimillionaire is really irrelevant.

What’s important is how one’s salary com-

pares to these other players.

A related issue that Equity Theory can help

to explain is how highly paid professional

athletes reconcile the fact that they are paid

a great deal more than physicians, teachers,

scientists, and others who perform work that isextremely important to society. Here’s one

somewhat speculative answer: A major league

baseball player making $10 million may reason

that he is deserving of this because of all the

years he spent developing his skills, the years

he spent playing in the minor leagues, and the

fact that his career could be ended at any time

by an injury. In Equity Theory terms, what thisplayer is doing is cognitively adjusting his

inputs relative to his outcomes.


decrease one’s outcomes, cognitively adjustone’s inputs or outputs, change the compar-ative standard, or even leave the exchange.Of all these strategies, the most common iscognitive adjustment, most likely because itis easier and more feasible than most of theothers.

In general, research has supportedEquity Theory very well, particularly withrespect to the underpayment condition. Ithas been shown, for example, that percep-tions of underpayment inequity are un-pleasant and will motivate employees to dosomething about the inequity (e.g., Green-berg, 1990; Lord & Hohenfeld, 1979). Inrecent years, equity theorists have distin-guished between equity with respect to theoutcomes employees receive, and the proce-dures used to determine those outcomes.Perception of the equity of one’s outcomesis referred to as distributive justice. The termprocedural justice is used to denote percep-tions of equity with respect to the proceduresused to determine outcomes (Folger &Cropanzano, 1998). Some researchers havealso argued for a separate category of interac-tional justice, which refers to the perceptionthat an employee has been treated with dig-nity and respect in regard to a particulartransaction (Colquitt, Conlon, Wesson, Por-ter, & Ng, 2001). Research has supportedthe distinctions between these three forms ofjustice (Cohen-Charash & Spector, 2000).

Collectively, these three forms of justiceare often referred to as fairness perceptions,and a great deal of recent research has beenconducted on the outcomes of these percep-tions (e.g., Sweeney & McFarlin, 1997).Fairness perceptions have been especiallyrelated to engaging in counterproductivework behaviors, including theft (Greenberg,2002), workplace revenge (Trip, Biese, &Aquino, 2002), and sabotage (Ambrose,Seabright, & Schminke, 2002). Clearly,

organizations should not want employeesto be motivated by a lack of perceived fair-ness at work.

The weakest support for Equity Theoryhas typically come from studies that haveexamined the overpayment condition(Pritchard, 1969). While it has been shownin a laboratory setting that feelings of over-payment can be induced (e.g., Lawler,Koplin, Young, & Fadem, 1968), there isvery little evidence of this effect in organiza-tional settings. This may be due to the factthat the whole notion of overpayment israther questionable. At least with respect tosalary, it simply may be unlikely that manypeople see themselves as overpaid. It is alsopossible that people may be able to cogni-tively adjust their perceptions very quickly toalleviate feelings of overpayment. For exam-ple, a person who is being paid what he orshe considers too much may rationalize thisby adjusting his or her perceptions of inputs(e.g., ‘‘My experience is a little better than Ithought’’) or outputs (‘‘With today’s prices,that salary is not as great as it seems’’).

Expectancy Theory

One of the things that is unique abouthumans, at least with respect to cognition,is their ability to anticipate the future andadjust their behavior accordingly. ExpectancyTheory is based on this uniquely humancharacteristic, and is focused on the cogni-tive processes that drive employees’ deci-sions regarding where they will direct theirefforts (Vroom, 1964, 1995). The basicpremise of Expectancy Theory is that em-ployees will generally direct their effortstoward behaviors or courses of action when:

1. There is a high probability that they willbe able to perform the behavior if theytry.


2. There is a high probability that the be-havior or course of action will lead tosome outcome.

3. The outcome that will result from thebehavior or course of action has value tothe person.

If any of these three conditions is lacking,a person is unlikely to direct his or her effortstoward that particular course of action.

According to Vroom (1964, 1995), thebelief that one’s efforts will allow one toperform a given behavior is referred to asexpectancy and is typically denoted aseffort-to-performance (E 1 P). Becauseexpectancy is a belief about the future,Vroom proposed that this is a probabilityfunction and, as such, may range from 0 to1. An expectancy of zero essentially meansthere is no way that a person’s efforts willresult in a given level of performance. Incontrast, an expectancy of close to 1 indi-cates that an employee has considerable con-fidence that, if he or she puts forth effort, agiven level of performance can be achieved.Expectancy beliefs may be based on a num-ber of factors: a person’s innate ability, his orher level of training, or the existence or lackof significant performance constraints.

The belief that a given behavior or level ofperformance will be associated with a givenoutcome is referred to as instrumentalityand is typically denoted as performance-to-outcome (P 1 O). Like expectancy, instru-mentality is a probability function. Forexample, an employee may perceive theinstrumentality for the relationship betweena given level of performance and a payincrease to be zero if salary raises are acrossthe board or are determined by collectivebargaining. On the other hand, a high instru-mentality would indicate a strong possibilitythat a given level of performance would berewarded with a given pay increase. Instru-

mentality beliefs are based, to a large extent,on stated organizational reward policies (i.e.,the existence of merit pay), but are also basedon the manner in which such policies arecarried out.

The value of the outcomes that an em-ployee may obtain is referred to as valence.According to Vroom, because of a number offactors, people differ on the value they attachto outcomes that can be obtained for differ-ent levels of performance. One person, forexample, may place a high value on mone-tary compensation; thus, a high raise mayhave considerable valence. Another person,in contrast, may place greater value on feel-ings of mastery and praise from others. Oneinteresting thing about valence is that it cantake on negative values, and this has impli-cations for predicting the direction of effort.Consider, for example, all of the things thatmay occur if an employee performs his or herjob very well. Pay raises, praise from one’ssupervisor, recognition from others, andfeelings of accomplishment are outcomesthat most people would find at least moder-ately desirable. In contrast, those who per-form their jobs well often end up having toperform a greater proportion of the work,and their higher salaries may encounterresentment from fellow employees. Theseoutcomes would be considered by most peo-ple to be at least moderately undesirable.

Vroom proposed that Expectancy,Instrumentality, and Valence can be com-bined, in equation form, to explain employeemotivation. This equation is presented inTable 8.2. The variable that this equationpredicts is labeled force, which representsthe level of effort that an employee will directtoward a given level of performance. Readersshould be clear that force is not the sameas performance. A person may direct hisor her efforts in a way that is consistentwith Expectancy Theory, yet not perform


well because of a lack of innate ability orperformance-related constraints.

As is shown in Table 8.2, for each possi-ble outcome that can result from a given levelof performance, instrumentality is multi-plied by the valence. These values are thensummed, and this sum is then multiplied byexpectancy. Given this equation, force willbe highest when employees believe thateffort will lead to a given level of perfor-mance, and that the level of performance willlead to valued outcomes. Conversely, if anyof these values are near zero, the motiva-tional force will be considerably lower. Forexample, let’s say an employee believes thereis a high probability that effort will lead to agiven level of performance, and that the out-comes that are possible are highly valued. Ifthis employee does not believe that theseoutcomes are contingent on performance(e.g., instrumentality is low), then force willbe low.

As another example, consider an em-ployee who believes that effort will lead toa given level of performance, and that per-formance will lead to a number of outcomes.In this case, force may still be low if theoutcomes have little value to the employee.The possibility of a promotion, or perhaps ofpraise, does not mean much to the employee.

Finally, an employee could believe thatperformance leads to highly valued outcomes,but he or she does not believe that the effort

will lead to performance (e.g., expectancy islow). For example, many marathon runnersbelieve that setting a world record wouldlead to a number of highly valued outcomes(e.g., money, fame, feelings of accomplish-ment), yet do not believe they can achieve thislevel of performance, even with considerableeffort.

Since its development in 1964 by Vroom,Expectancy Theory has become one of thedominant motivational theories in organiza-tional psychology. As a result, considerableresearch has examined expectancy theorypredictions. Van Eerde and Thierry (1996)performed a meta-analysis of 77 studies thathave tested Expectancy Theory predictions,and examined the correlations betweenexpectancy theory components and out-comes such as performance, effort, inten-tion, preference, and choice.

The results of this study showed mixedsupport for Expectancy Theory. For exam-ple, although individual components suchas expectancy and instrumentality were cor-related with a number of outcomes, multi-plying terms together, as suggested byExpectancy Theory, did not result in greaterprediction. Another important finding fromthis meta-analysis was that correlationsbased on studies employing within-subjectsdesigns were stronger than correlations fromstudies employing between-subjects designs. Ina within-subjects design, Expectancy Theorywould be used to predict a particular indi-vidual’s choice among different levels of per-formance or different courses of action. In abetween-subjects design, Expectancy Theorywould be used to predict performance oreffort from a large number of individuals.This finding supports the contention thatthe theory is useful in predicting how peoplewill direct their efforts when faced with anumber of different choices (e.g., Mitchell,1974; Muchinsky, 1977).

TABLE 8.2The Equation Representing How the Componentsof Expectancy Theory Interact to DetermineMotivational Force

F ¼ E (S I � V)

F ¼ Motivational force

E ¼ Expectancy (E ! P)

S ¼ Summing over all possible outcomes

I ¼ Instrumentality (P ! O)

V ¼ Valence


In addition to direct empirical tests,Expectancy Theory has received indirectsupport from studies that have examinedthe impact of financial incentives (Jenkinset al., 1998; Lawler, 1990; Lawler & Jenkins,1992). Although financial compensation willbe discussed in greater depth in Chapter 9,suffice it to say that considerable evidencehas shown that financial incentives can be apowerful motivator. Although this in itselfdoes not constitute direct support forExpectancy Theory, it is certainly consistentwith many of its propositions.

Goal-Setting Theory

The idea that human behavior is motivatedand regulated by goals and aspirations haslong been recognized by psychologists(Austin & Vancouver, 1996). Thus, likeExpectancy Theory, the conceptual under-pinnings of Goal-Setting Theory can be tracedback many years. Organizational psycholo-gists, most notably Edwin Locke, have elab-orated on the basic notion of goal-setting andhave described how this drives behavior inorganizations.

Before describing the specifics of Goal-Setting Theory, it is important to considerwhy goals motivate employees’ behaviors.According to Locke (1968), goals have moti-vational value for three reasons:

1. Goals serve to direct our attention andfocus our efforts in a particular direction.A student who has a goal of obtaining anA in a course is likely to direct much ofhis or her attention toward that course.

2. Goals help us to maintain task persis-tence. This is important because, inmany cases, people will fail or get side-tracked when they are trying to accom-plish something.

3. The existence of goals tends to facilitatethe development of task strategies. Forexample, the student in item 1 maydevise very innovative methods of study-ing his or her course material in order toenhance retention.

Having described the functions servedby goals, we now examine the attributes ofgoals that make them motivating. One attrib-ute that has been supported very stronglyover the years is goal difficulty (e.g., Locke& Latham, 1990a). Generally speaking,goals that are difficult are more motivatingthan easier goals. For example, a salespersonis going to be more motivated if he or she hasa goal of making $100,000 in commission,rather than a goal of $50,000.

The second attribute that must bepresent for a goal to have motivational valueis goal acceptance. To a large extent, goalacceptance hinges on a person’s belief thata goal is attainable. If a person does notbelieve he or she can attain a particular goal,this goal will probably not be accepted. Overthe years, it has been suggested that employ-ee participation in goal setting is a necessarycondition for goal acceptance. Latham andLocke (1991), however, point out that evi-dence has shown that assigned goals can bejust as motivating as jointly set goals, as longas they are accepted.

The third condition necessary for goals tobe motivating is goal specificity. Goals aremuch more motivating when they specify aparticular level of performance (e.g., ‘‘Sell 20cars in the next month’’), as opposed to beingvague (e.g., ‘‘Be a good salesperson’’).Because of the importance of goal specificity,many goal setting studies have what isreferred to as a ‘‘do your best’’ condition inwhich participants are given no concreteperformance goals.


Fourth, it has generally been recognizedthat employees must receive feedback inorder for goals to motivate performance.Attaining a goal is often an incremental pro-cess; thus, it is important that employeesreceive feedback regarding their progress.Furthermore, according to Latham andLocke (1991), the relation between goalsand feedback is actually reciprocal; that is,feedback helps employees to keep on trackwith respect to goal attainment. Conversely,the existence of goals helps to put feedbackinto a meaningful context.

In terms of research support, goal settingis one of the most well-validated theories inall of organizational psychology. Over 30years of research has supported the motiva-tional value of goals in both laboratory andfield settings (e.g., Locke & Latham, 1990a,1990b). Because of this wide support, thefocus in the past 15 years has largely been onexplaining the mechanisms underlying goalsetting, as well as identifying boundary con-ditions of the theory and moderators of someof the key relationships predicted by thetheory.

With respect to mechanisms underlyinggoal setting, considerable research has beenconducted on goal commitment, goal accep-tance, feedback, and self-efficacy (Ambrose& Kulik, 1999). It has been found, for exam-ple, that monetary incentives can be used toenhance goal commitment and acceptance(Wright, 1992), and that both feedback andself-efficacy are necessary conditions forgoal setting to be effective (Latham & Locke,1991; Locke & Latham, 2006). It makessense that employees will not be able totranslate their goals into high levels of per-formance if they are not given adequatefeedback regarding the extent to whichtheir behaviors are contributing to perfor-mance. A great deal of recent research has

revealed the importance of both employeesseeking feedback regarding their perfor-mance (Ashford & Black, 1996; Bernichon,Cook, & Brown, 2003) and managers pro-viding feedback to employees (Ilgen &Davis, 2000).

In terms of boundary conditions, severalstudies have indicated that goal setting maynot work in all situations. Because goals tendto narrow one’s focus, they may actually becounterproductive in situations in whichan employee may need to alter a poorlydesigned task (Staw & Boettger, 1990). Also,those who are assigned specific goals may beless likely to spontaneously help coworkers(Wright, George, Farnsworth, & McMahan,1993). This tendency for goals to lead totunnel vision may be counterproductive inorganizations of the future, since it has beenpredicted that role boundaries will be muchless well defined (Bridges, 1994).

Another boundary condition of goal set-ting is that there may be a law of diminishingreturns with respect to the number of goalsthat an employee can use to guide his or herbehavior: As the number of goals begins toincrease, the probability of conflict betweengoals increases (Gilliland & Landis, 1992).Furthermore, when an employee has a largenumber of goals, the probability increasesthat he or she may not even be able to keeptrack of them all. Given that goal specificity isa key element of the theory, it is unlikely thatan employee will be able to retain the specif-ics of an excessive number of performance-related goals.

A final boundary condition that has beenexamined in recent years is task complexity.Research has shown that goal setting may bemore effective for simple (as opposed to com-plex) tasks (Mone & Shalley, 1995). Themost frequently cited reason for this is thatif they are going to motivate performance,


goals require a portion of a person’s cognitiveresources (e.g., Kanfer, Ackerman, Murtha,Dugdale, & Nelson, 1994). Devoting cogni-tive resources to goals will have a detrimentaleffect when one is performing a complex task(e.g., preparing an annual budget). Also,when goal setting is used for complex tasks,goals are often set at inappropriate levels.Setting very distal goals will probably notbe very helpful when one is performing acomplex task. For example, if a researchscientist were to set a very distal goal (e.g.,‘‘I want to obtain three scientific break-throughs in the next 10 years’’), this mayhave very little impact on performance. Onthe other hand, if a person performing thisjob were to set more proximal goals (e.g.,‘‘Read three important research articles thisweek’’), this could potentially facilitate taskperformance. Given the increasing complex-ity of future jobs, this is an issue that clearlywarrants more attention in goal-settingresearch.

Self-Regulation

In many areas of psychology, a trend inrecent years has been to explain behaviorin terms of self-regulation mechanisms.Some of these theories will be discussed inthe following paragraphs.

Control Theory

Control theory represents a very generaltheory that attempts to explain self-regula-tion processes underlying motivation(Carver & Scheier, 1981; Powers, 1973a,1973b, 1978b). Control theory is typicallydiscussed within the organizational psycho-logical literature in the context of goal set-ting, where individuals attempt to reducethe discrepancy between a desired end state(i.e., goal) and their current level of progress

toward that end state (e.g., Klein, 1989;Latham & Pinder, 2004).

According to Powers (1973a), any con-trol system consists of four distinct parts:

1. A sensor is a component that gathersimportant information about the controlsystem. For humans, the sensor repre-sents one’s observations and perceptions.

2. A standard represents some state that asystem attempts to maintain or achieve.In terms of motivation, this would mostlikely be some type of goal, such as alevel of performance or perhaps a moregeneral aspiration (e.g., wanting tobecome a doctor).

3. A comparator or discriminator representsthe mechanism by which informationthat is obtained by the sensor is comparedto the standard. For example, a personmay cognitively compare his or her rateof progression toward a desired goal.

4. An effector represents the mechanism bywhich the system can interact with itsenvironment. With humans, the effectormechanism makes it possible, for exam-ple, to adjust one’s effort if it is deter-mined that progression toward a givengoal is too slow.

Control theory conceptualizes motiva-tion as an ongoing process by which peoplecognitively monitor their progress towardsome goal or standard, and may make adjust-ments based on whether they are makingprogress toward that goal or standard(Scheier & Carver, 1988; see also Nelson,1993). This is a negative discrepancy model,where individuals are motivated to minimizethe discrepancy between a current anddesired end state. Furthermore, individualsobviously have many goals they are pursuingat any point in time, and these goals arehierarchically organized. According to


Carver and Scheier, the highest level of feed-back loop is the idealized self-image, whichrepresents the ideal images individuals haveset for themselves. For example, an individ-ual may have an idealized self-image of beinga good and decent person. The next highestlevel of feedback involves principles, whichwhen achieved contribute to the idealizedself-image. Principles can be seen as similarto values, in that they provide guidelines forbehavior. Examples of principles may bekindness, altruism, courage, or achievement.The lowest level of feedback involves pro-grams, which are the specific behaviorsthat are carried out in order to achieve theprinciples an individual has adopted. Forexample, an individual who endorses theprinciple of achievement would constantlyenact programs where achievement was pos-sible (e.g., excelling on particular tasks atwork).

Given its generality, control theorycould be used to explain essentially anyform of purposeful behavior (e.g., weightloss, progress in psychotherapy, accumu-lation of wealth). As stated earlier, organi-zational psychologists have used controltheory primarily as a means of explainingthe mechanisms underlying goal setting(e.g., Klein, 1989; Lord & Hanges, 1987),although it has also been used in otherareas, such as occupational stress (e.g.,Edwards, 1992).

According to Klein (1989), controltheory augments goal setting in many ways.For example, control theory provides a moreelegant description of the process by whichfeedback influences goal-setting processes.In control theory terms, feedback representsa sensor that facilitates the process by whichan individual compares his or her perfor-mance to a given goal, and makes adjust-ments as necessary. Control theory alsoprovides a plausible explanation for why a

person may revise his or her goals in the faceof repeated failure.

Beyond the implications for goal setting,control theory also serves as a useful frame-work for examining many other types of be-havior in organizations. For example, anemployee who suddenly begins to put forthmore effort may be doing so because he or shefeels that the current level of effort matcheswhat is considered to be the effort of a ‘‘goodemployee.’’ An employee who decides tochange jobs may feel that the current jobdoes not match his or her perception of whata job should be providing. Finally, an em-ployee may decide to decrease his or herhours at work and spend more time at homebecause this is more congruent with his orher perception of being a ‘‘good parent.’’

Carver and Scheier have accumulatedevidence for control theory within the con-text of their theory of self-regulation. How-ever, less research has been conducted usingcontrol theory separate from goal settingwithin an organizational setting. The over-whelming support for goal-setting theoryover the years certainly bodes well for theviability of control theory. In the future,there is likely to be more research on goalsetting performed from a control theory per-spective. It is also quite possible that organi-zational researchers will examine otheremployee behaviors from a control theoryperspective.

Social Cognitive Theory (SCT)

Albert Bandura was one of the main pioneersof the cognitive revolution in psychology,and his SCT emphasized that cognitive var-iables mediate the relationships betweenenvironmental events and behavioral conse-quences (Bandura, 2001). A central com-ponent of Bandura’s model is the constructof self-efficacy, which he initially defined as


‘‘beliefs in one’s capabilities to organize andexecute the courses of action required to pro-duce given attainments’’ (1997, p. 3). Thisdefinition emphasizes the confidence theindividual has to complete his or her task.Stajkovic and Luthans (1998) expanded onBandura’s original definition and definedself-efficacy as, ‘‘an individual’s confidenceabout his or her abilities to mobilize themotivation, cognitive resources, and coursesof action needed to successfully executea specific task within a given context’’(p. 63). These authors argue that self-efficacyaffects several aspects of motivation, includ-ing behavior initiation, effort, persistence,and success. In measuring self-efficacy, it isimportant to remember that the construct isdomain-specific, and therefore should beassessed with reference to performance ona specific task (Latham & Pinder, 2005).

In addition to the concept of self-efficacy,Bandura (2001) has also contributed to the-oretical work on self-regulation through hisemphasis on dual control systems that deter-mine goal-directed behavior. Bandura arguesthat goal-directed behavior involves twocontrol systems. The first system involvesthe creation of a positive discrepancybetween an individual’s current state and adesired future goal. The second is the nega-tive discrepancy model discussed earlier,where individuals attempt to reduce the dis-crepancy between the desired goal and thesteps required to reach the goal. Bandura’smodel can be integrated with goal-settingtheory by noting that setting difficultgoals involves creating a negative discrep-ancy control system where a high goal is tobe mastered.

THE BEHAVIORAL APPROACHTO MOTIVATION

The underlying assumption of the behavioralapproach to motivation is that behavior is

largely a function of its consequences. Forexample, when working with laboratory ani-mals, the frequency with which a rat pressesa bar is largely a function of the conse-quences of performing that behavior. If theconsequence is positive for the rat (e.g., afood pellet), this will increase the probabilityof the behavior occurring in the future. Onthe other hand, if the consequence is eithernegative (e.g., an electric shock) or neutral(e.g., nothing happens), this will decreasethe probability of the behavior occurring inthe future.

The behavior of people in work settings ismuch more complicated than the behavior oflaboratory rats. However, at a very basic level,the general principle described also governsbehavior in organizations; that is, people inorganizations generally try to behave in waysthat produce positive outcomes, and avoidbehaving in ways that produce negative orneutral outcomes. In the remainder of thissection, we will examine some of the mecha-nisms underlying behavioral explanations ofmotivation in organizations.

One of the major principles that can beused to influence behavior in organizationsis reinforcement. Reinforcement can bedefined as any stimulus that increases theprobability of a given behavior. If an em-ployee writes a good report and receivesverbal praise from his or her supervisor,the verbal praise could be considered rein-forcement. Readers will notice that this def-inition is not conceptual; rather, we definereinforcement primarily in terms of its func-tion or consequences.

One of the key issues in the use of rein-forcement to influence behavior is how it isadministered. According to Luthans andKreitner (1985), schedules of reinforcementdescribe various strategies that can be usedto administer reinforcement. A general dis-tinction that can be made about reinforce-ments is between those that are continuous


and those that are intermittent. If reinforce-ment is provided continuously, this simplymeans that a person is constantly receivingreinforcement for his or her actions. Thistype of reinforcement schedule is rarely usedin organizations, but may have some usewhen new employees are initially learningtheir jobs. For example, a supervisor mayinitially reinforce a new employee every timehe or she successfully completes a workassignment.

An obvious problem with continuousreinforcement is that it is inefficient for theorganization. Also, if reinforcement is pro-vided continuously, it may eventually losevalue to the employee. Thus, in most cases,reinforcement in organizations is providedaccording to intermittent schedules. Onecommon form of intermittent reinforcementis a fixed-interval schedule—the administra-tion of reinforcement according to predict-able time periods. Paying employees once amonth is an example of such a schedule in anorganization. A key decision to be madewhen using a fixed-interval schedule is thelength of time between administrations ofreinforcement. For example, when employ-ees are first learning a task, it is common forintervals between reinforcement to be verysmall. Gradually, however, the intervalsbetween administrations of reinforcementbecome larger. That is, an employee mayreceive a compliment or other reward per-haps once every few days. One problemwith a fixed-interval schedule is the well-established tendency for organisms to slackoff performing immediately following thereinforcement.

A variable-interval reinforcement sched-ule is also the administration of reinforce-ment over time. However, unlike thefixed-interval schedule, when a variable-interval schedule is used, the time intervalbetween administrations of reinforcementvaries. For example, an employee may

receive compliments from his or her super-visor twice in the same week, but may notreceive another compliment during thenext 3 weeks. The power of variable rein-forcement lies in the fact that the employeedoes not know exactly when it is coming.Some rewards cannot be administered thisway (e.g., salary), but variable schedules canbe a powerful way to motivate behaviorusing other more intangible reinforcers.

Intermittent reinforcement can also beadministered after a given number of timesa desired behavior is performed; suchschedules are referred to as ratio schedules.For example, in a laboratory setting, a ratmay receive a food pellet for pressing a bara certain number of times. In an organiza-tional setting, an employee may receive areward based on the performance of a givenbehavior (e.g., selling a car, or a certainnumber of cars). If a fixed-ratio scheduleis used, reinforcement is administered aftera behavior has been performed a givennumber of times. For example, employeesworking at a manufacturing plant maybe given a bonus for every 50 productsdelivered.

As with the fixed-interval schedule, a keydecision is the number of behaviors thatmust be performed before the employeemay receive the reward. In many cases, thenumber of behaviors required to obtain rein-forcement has to do with the skill level ofemployees. For example, when employeesare first learning a task, the number of behav-iors required to obtain reinforcement willgenerally not be very high. Over time, asthe employee becomes more skilled, morebehaviors are typically required in order toobtain reinforcement.

When a variable-ratio reinforcementschedule is used, reinforcement is alsoadministered based on the behavior per-formed. However, unlike the fixed-ratioschedule described previously, the number


of behaviors required to obtain reinforce-ment varies. An employee may be reinforcedafter performing a given behavior twice, andthen not reinforced again until the behavioris performed five more times. Some readerswill recognize this as the reinforcementschedule on which gambling is based. Giventhe number of people who become addictedto gambling, it is fair to say that this is a verypowerful schedule of reinforcement. Like thevariable-interval schedule described previ-ously, some rewards cannot be administeredaccording to this schedule for ethical rea-sons. However, rewards such as praise andrecognition certainly can be, and often are,administered in this manner.

A second major principle of the behav-ioral approach to motivation is that of punish-ment, or any consequence that has the effectof reducing the probability of a behavioroccurring. In organizational settings, pun-ishment may be used to influence behavior,but typically it is used much less often thanreinforcement. The most common use ofpunishment in organizations is to decreasethe frequency of counterproductive behav-iors. Thus, punishment primarily motivatesbehavior by discouraging negative behavior.The most common forms of punishment inorganizations are docking employees’ pay,suspension, demotion, being given undesir-able work assignments, and, in extremecases, termination.

Although punishment may have a power-ful effect, there are important issues to con-sider before organizations use it to influenceemployee behavior. For example, althoughpunishment may produce the desired out-come in the short run, it may also produceconsiderable resentment and distrust amongemployees. In addition, it is well known thatpunishment tends to suppress undesirablebehavior rather than eliminate it completely.Another danger in using punishment to

influence behavior is that an organizationmay adopt it as the primary mode of influ-ence. Typically, in this mode, employeesare not praised when they do something wellbut are punished when they do somethingwrong. The overuse of punishment may alsolead to negative perceptions of fairness,which, as discussed previously, result in ahost of dysfunctional consequences.

In many cases, the behavior of employeesin organizations may meet with neither pos-itive nor negative consequences—that is,nothing happens. This should lead to a phe-nomenon is known as extinction, where theoccurrence of the given behavior declinesuntil it is not performed at all. The influenceof extinction on organizational behavior maybe positive or negative, depending on thenature of the behavior under consideration.For example, if an employee is rude andobnoxious during meetings, extinguishingsuch behavior is positive. On the other hand,if an employee is very helpful to others andnever receives any acknowledgment, there isa chance that the positive behavior will beextinguished. Of course, in such cases, someemployees may feel a sense of internal sat-isfaction and thus may keep performinghelpful behaviors for that reason. However,when such behavior is no longer internallyrewarding, it may cease.

Probably the most important implicationof extinction is that organizations must thinkabout the behaviors they want to encourageand the behaviors they want to see mini-mized. Too often, rewards in organizationsare administered in a way that encouragesbehaviors that are only minimally important,and extinguishes those that are the mostcrucial to organizational success. Consider,for example, an organization that adminis-ters rewards primarily on the basis of senior-ity. In behavioral terms, such an organizationis saying that the most valuable commodity is


the length of service of employees. Underthis type of system, an employee who per-forms very well but has not been employed along period of time has little incentive tomaintain a high level of performance. InChapter 9 we focus more on the strategicuse of reinforcement to motivate employees.

In many companies, behavioral princi-ples are used in training employees to learnnew skills and to adopt new behaviors. Par-ticularly when an employee is learning anovel behavior, the behavioral principle ofshaping comes into play. Essentially, shapinghas to do with the reinforcement of succes-sive approximations of a particular behavior,rather than the entire behavioral sequence.Probably the best example of the use ofshaping is in the training of animals. Readerswho have been to Sea World have undoubt-edly enjoyed the tricks performed by sealions and killer whales. To teach those tricks,trainers must work many hours and rein-force the slightest movements that are seenas leading to the ultimate behavior.

In organizational settings, shaping maybe used in ways that have implications foremployee motivation. For example, whenemployees are first learning job tasks,reinforcing ‘‘successive approximations’’ ofultimate task performance will keep an em-ployee from getting discouraged. In manyacademic departments, faculty are oftenreinforced for taking preliminary steps thatmay lead to desired outcomes, such as pub-lication and external grants. By reinforcingbehaviors such as building relationshipswith those at funding agencies, and estab-lishing collaborative relationships with otherresearchers, it is hoped that such activitieswill ultimately lead to grants and publica-tions.

A final behavioral principle that hasimportant implications for motivation isfeedback. When employees engage in any

form of behavior (or in performance-relatedbehavior in particular), it is helpful to havesome feedback about that behavior (Kluger& DeNisi, 1996). Feedback has motivationalvalue, particularly when it is positive. Mostemployees enjoy hearing positive feedbackwhen they perform well, and such feedbackoften serves as an incentive to maintain ahigh level of performance. Feedback can alsohave considerable diagnostic value whenemployee performance is lacking. When anemployee is performing poorly, feedbackserves the important function of letting theperson know that he or she is seen as notperforming well. Sometimes, it is quiteobvious when performance is lacking (e.g.,a comedian tells a bad joke), but in manycases, it is not (e.g., a manager who is makingpoor strategic decisions). Thus, in manyinstances, feedback about performance mustbe given by some external agent or the em-ployee simply will not know that he or she isperforming poorly.

Perhaps the most important diagnosticfunction of feedback is that it communicatesto employees where specific performancedeficiencies exist. Simply having the knowl-edge that one is not performing well is cer-tainly useful. However, it is more useful toreceive feedback on what specific aspects ofperformance are lacking. Once these aspectsare established, it is possible to go evenfurther and diagnose the root cause of theperformance problem.

In organizations, the application of be-haviorism is known as Organizational Behav-ior Modification (OBM), and this approachhas been used to influence a number ofbehaviors, such as safety, suggestive selling,and production efficiency. In general, evi-dence supporting the effectiveness of OBMis impressive (Weiss, 1990). That is, usingbehavioral principles has been shown toinfluence the behaviors listed above in ways


that are favorable for organizations. One lim-itation of OBM, and thus of the behavioralapproach, is that it appears to work bestwhen it is applied to relatively simple formsof behavior. That is, when jobs are relativelysimple, it is much easier to keep track ofdesirable and undesirable behaviors, andapply reinforcement accordingly. With morecomplex tasks, however, this becomes muchmore difficult to do.

As an example, suppose we tried to usereinforcement principles to motivate a scien-tist who is working on mapping the entirehuman genetic structure. Because of thecomplexity of this type of scientific activity,it would likely be quite difficult to get a goodhandle on all of the steps necessary to ulti-mately accomplish this goal. Also, becauseprogress in this type of scientific activity isvery slow and incremental, reinforcementmay be so infrequent that it would have littleimpact on motivation.

Another issue has been raised about be-haviorism: the ethics underlying this ap-proach. Some critics, for example, havecharged that by systematically analyzing thecontingencies underlying behavior andmanipulating the environment to impact be-havior, people are robbed of their choice andfree will. B. F. Skinner, in his 1971 book,Beyond Freedom and Dignity, countered suchcharges by stating that environmental con-tingencies will govern behavior whether ornot we choose to intervene. Behaviorism, inhis view, represented nothing more than asystematic attempt to use those environmen-tal contingencies in a way that was beneficialto society.

SELF-DETERMINATIONTHEORY

Thus far the motivation theories we haveconsidered have viewed the employee as a

calculating individual who is motivated bythe desire to pursue positive consequencesand avoid negative consequences. Thesetypes of theories are collectively referredto as hedonic theories of motivation, with anemphasis on needs, cognitive processes,and behavioral consequences being usedin the service of allowing an individual tomaximize pleasure and avoid pain (Ryan &Deci, 2001). In contrast, organismic theoriesof motivation ‘‘assume that humans areinherently motivated to develop their inter-ests and skills, to connect and contributeto other people, and to move toward theirfullest potential’’ (Sheldon, Turban, Brown,Barrick, & Judge, 2003, p. 358). Organismictheories are growth-oriented, emphasizinghuman beings’ innate need to develop.

Self-determination theory (SDT) is anorganismic theory that has been very influ-ential in areas outside of organizational psy-chology, but it has only recently been appliedto employee motivation at work (Sheldon etal., 2003). Sheldon and his colleagues notethat SDT is actually a collection of a numberof smaller theories that all emphasize theimportance of individuals performing tasksout of intrinsic interest or a sense that taskperformance is relevant to important aspectsof an individual’s identity. Deci (1975) ini-tially emphasized the importance of intrinsicmotivation for task enjoyment and persist-ence, where an individual performs a taskbecause of an inherent enjoyment in the taskitself, rather than for extrinsic reasons suchas pleasing others or making money. Deciargued that intrinsically motivating tasks feelautonomous, coming from the individual,whereas extrinsically motivating tasks feelcontrolled, with performance having moreto do with external demands than one’sinherent interests. Employees who trulyenjoy the work they do will feel moreautonomy at their job, which should lead


them to persist in their work even whenexternal factors, such as being monitoredby a boss, are absent.

Later research emphasized the importanceof how an individual interprets the motivationbehind the various tasks he or she performs.Cognitive Evaluation Theory made the impor-tant point that how individuals interpret orevaluate the reasons for their behavior deter-mines whether that behavior ends up feelingautonomous versus controlled (Deci & Ryan,1985). For example, two different employeesmay get a bonus for performing at a certainlevel. One employee may interpret the bonusas an attempt by management to control hisor her behavior, whereas another employeemay interpret the bonus as recognition fora job well done. The latter employee is likelyto feel greater autonomy at work, and there-fore exhibit greater task persistence andenjoyment.

SDT appears to work well with inher-ently interesting jobs where individuals cangain enjoyment from the job itself, but whatabout really boring jobs that provide littleinherent interest? For example, it is hard toimagine employees showing intrinsic moti-vation for tasks that are highly monotonousand uninteresting (e.g. checking for errors indatabase after database). Recent modifica-tions to SDT accommodate these types oftasks. Ryan and Deci (2000) present an elab-orated version of SDT where extrinsic moti-vation is separated into four separate types:external motivation (where the individual isperforming for external rewards), introjectedmotivation (where the individual is perfor-ming in order to avoid feeling guilty aboutnot living up to obligations), identified moti-vation (where the individual is performing atask because it expresses his or her values oris otherwise personally important), and inte-grated motivation (where various sources ofidentified motivation are integrated into a

coherent self-conception). Ryan and Deciargue that external and introjected motiva-tion are both perceived by the individual ascontrolling, but that identified and inte-grated motivation is perceived as autono-mous, because the individual ‘‘owns’’ thebehavior.

Sheldon et al. (2003) discuss how theseforms of motivation can be applied to anemployee who builds computers. Althoughit is unlikely that many employees have anintrinsic interest in putting computers to-gether, an employee may build a computerto get a paycheck (external), to avoid beingseen as a bad worker (introjected), or be-cause he or she takes pride in producing aquality computer (identified). The authorsargue that employees motivated throughidentification will be more satisfied at workand will maintain superior performance inthe absence of supervision by others.

SDT is just beginning to be applied toemployee motivation within organizationalpsychology. In the context of work motiva-tion, Sheldon et al. (2003) detailed how thetheory could be used to provide insights intothe areas of goal commitment and trainingmotivation. The authors point out that mostconceptualizations of goal commitment em-phasize commitment to goals as a function ofexpectancy of achieving the goal and theattractiveness of the goal. A SDT approachemphasizes the importance of the internal-ization of the goal as personally important inpredicting goal commitment. Sheldon andhis colleagues have found that individualsare more likely to attain goals when theyhave internalized the goals and are pursuingthem for autonomous reasons (Sheldon &Elliot, 1998; Sheldon & Houser-Marko,2001).

Judge and his colleagues have recentlyused a model based on SDT to understandthe relationship between core self-evaluation

Self-Determination Theory 255

and job satisfaction (Judge, Bono, Erez, &Locke, 2005). Recall from Chapter 4 thatemployees with high levels of core self-evaluation have high self-esteem, high self-efficacy, an internal locus of control, and lowlevels of neuroticism. These authors con-ducted a longitudinal study where coreself-evaluations and the motivation under-lying the pursuit of work goals were assessedat Time 1 and job satisfaction and the attain-ment of work goals were assessed at Time 2(60 days later). The authors found that coreself-evaluation was related to a higher level ofautonomous motivation for work goals(pursing work goals for intrinsic and identi-fied reasons). In addition, higher levels ofautonomous motivation at Time 1 wererelated to higher goal attainment and higherlevels of job satisfaction at Time 2. Futureresearch is needed to examine the ability ofSDT to predict performance related out-comes at work.

JOB-BASED THEORIESOF MOTIVATION

Job-based theories propose that the key tounderstanding motivation lies in the contentof employees’ jobs. Job-based theories areclosely related to need-based theories, dueto the fact that need satisfaction is oftenoffered as an explanatory mechanism linkingjob content and motivation. Job-based theo-ries, however, are more likely than need-based theories to have been developed spe-cifically for the workplace. Also, focusing onjob content as the lever for influencing be-havior is inherently more practical thanfocusing on need satisfaction.

Motivation-Hygiene Theory

From a historical perspective, the firstjob-based theory to appear on the scene

was Herzberg’s Motivation-Hygiene Theory(Herzberg, 1968). The basic premise behindHerzberg’s theory, as with all job-based theo-ries, was that the primary source of motiva-tion in the workplace was the content ofpeople’s jobs. At the time that Motivation-Hygiene Theory was developed, most orga-nizations were highly influenced by ScientificManagement. Recall that the primary methodof motivation in Scientific Management wasthrough compensation and financial incen-tives. Herzberg, and others, were of theopinion that financial incentives had thepower to ‘‘motivate’’ people in the sense thatthey kept them on the job and perhapsprevented them from complaining. Usingterminology from SDT, motivation was com-pletely external. However, Herzberg believedthat, in order to truly motivate employees,the content of the jobs that people performwas key.

Herzberg proposed that the work envi-ronment could be divided into two generalcategories. The first of these, hygiene factors,included aspects of the work environment,such as pay, fringe benefits, relations withcoworkers, and essentially everything elsethat is distinct from the content of anemployee’s work. Herzberg used the termhygiene factors because these factors are nec-essary to keep employees from being dissat-isfied but do not have the power to trulymotivate them. To use a health-related anal-ogy, maintaining proper dental hygiene doesnot make a person’s teeth any better, but itprevents problems such as tooth decay andgum disease.

Herzberg labeled the second category ofthe work environment motivators. In contrastto hygiene factors, motivators reside primar-ily in the content of a person’s job. Motiva-tors include things such as the amount ofchallenge inherent in one’s work, the amountof discretion one has in carrying out one’s job


tasks, and perhaps how intrinsically interest-ing the work is. According to SDT discussedearlier, these factors all promote a moreintrinsic orientation to performing at work.According to Herzberg, in order to motivatean employee, an organization must designwork in a way that builds in motivatorsand thus makes work content intrinsicallyrewarding to employees. A summary of hy-giene factors and motivators is contained inFigure 8.2.

In terms of empirical support, Motivation-Hygiene Theory has not fared particularlywell. Herzberg, Mausner, and Snyderman(1959) provided what has probably beenthe strongest support for Motivation-HygieneTheory. In their study, engineers and ac-countants employed by the City of Pittsburghwere asked to describe critical incidents thatwere illustrative of being ‘‘very satisfied’’ or‘‘very dissatisfied.’’ Consistent with Motiva-tion-Hygiene Theory, hygiene factors werementioned more often than motivators whendescribing dissatisfaction. The reverse wastrue when describing satisfaction.

After the Herzberg et al. (1959) study,other attempts were made to test the theory(summarized by Locke, 1976), but mostwere unsuccessful. Since the mid-1970s,very little research has used Motivation-Hygiene Theory as a framework. (For anexception, see Maidani, 1991.) The reason

most analysts have given for the inability toreplicate the theory is the critical incidentmethod used in Herzberg’s original study.Specifically, in describing critical incidents,respondents tended to attribute highly sat-isfying incidents to aspects of the job thatwere most closely associated with themselves(e.g., the degree of challenge in the job) andto attribute dissatisfying incidents to aspectsof the work environment that were mostclosely associated with others (e.g., socialrelations with coworkers). Thus, many haveargued that Herzberg’s findings representeda methodological artifact rather than supportfor Motivation-Hygiene Theory. (See Locke,[1976] for an example of one of the mostvigorous criticisms of the theory.)

Despite the lack of empirical support forhis theory, one can certainly argue thatHerzberg was a pioneer because he was oneof the first organizational psychologists tofocus on job content as a source of employeemotivation. This point has often been lostover the years. In focusing on the methodo-logical shortcomings of Motivation-Hygiene Theory, many in the field havefailed to give Herzberg credit for being aheadof his time and providing the foundationfor many practices that are commonplacein organizations today. In addition, the crit-ical distinction emphasized by Herzbergbetween external and intrinsic motivationhas gained a great deal of support throughSDT described previously.

Job Characteristics Theory

Although Herzberg was one of the first toemphasize the importance of job content inmotivating employees, his theory had someimportant limitations. For example, Motiva-tion-Hygiene Theory was rather imprecise asto how to build ‘‘motivators’’ into employees’jobs. Herzberg also provided no tangible

FIGURE 8.2Summary of Hygiene Factors and Motivators

Hygiene Factors

Pay

Fringe Benefits

Relations withCoworkers

Physical WorkingConditions

Motivators

Level of Challenge

Level of Autonomyor Discretion

Intrinsic Interest

Opportunities forCreativity


measures of these job dimensions. Anotherproblem with Motivation-Hygiene Theory isthat it was based, at least implicitly, on theassumption that all employees want the samethings from their work.

Job Characteristics Theory (Hackman &Oldham, 1976, 1980), to a large extent,addressed the deficiencies in Motivation-Hygiene Theory and has become the mostinfluential job-based theory of motivation inorganizational psychology. Before describingits major components, we should note thatJob Characteristics Theory has evolved overseveral years. For example, Turner and Law-rence (1965) proposed the concept of Requi-site Task Attributes—essentially, a set of job

dimensions that they believed to be motivat-ing to employees. Building on this work,Hackman and Lawler (1971) proposed whatis generally considered the most immediateprecursor to what eventually became JobCharacteristics Theory. Job CharacteristicsTheory extended this earlier work by pro-posing the mediating linkages between jobcharacteristics and outcomes, and by speci-fying moderators of individual differences.

As can be seen in Figure 8.3, the startingpoint in the theory is labeled core job dimen-sions. These represent characteristics of aperson’s job and include the followingdimensions: skill variety, task identity, tasksignificance, autonomy, and feedback. Skill

FIGURE 8.3Job Characteristics Theory of Motivation

Skill VarietyTask Identity

Task Significance

Feedback

ExperiencedMeaningfulness

Knowledge ofResults

GrowthNeed Strength

AutonomyFelt

Responsibility

CriticalPsychological States OutcomesCore Job Dimensions

Internal MotivationJob Satisfaction

Performance QualityLow Absenteeism

Low Turnover

Source: J. R. Hackman and G. R. Oldham. (1980). Work redesign. Reading, MA: Addison-Wesley. Reprinted by permission of

Pearson Education, Inc. Upper Saddle River, NJ.


variety represents the extent to which a jobrequires that a person must use many differ-ent skills. A good example of a job with highskill variety would be that of a corporateexecutive. A person performing this jobmay have to utilize quantitative skills to pre-pare a budget, interpersonal skills to manageconflicts among others, and high-level ana-lytical skills to develop a long-term strategicplan. On the other end of the spectrum, amanual labor job may require primarilyheavy lifting and a very minimal amount ofindependent judgment.

Task identity represents the extent towhich a job requires that a person mustcomplete a whole identifiable piece of work,as opposed to a small fragment of it. Con-ducting research is an example of a job withhigh task identity because it requires a per-son to be involved in all steps in the process:reviewing the literature, developing meas-ures, collecting and analyzing data, and writ-ing a report. Low task identity might befound in a traditional assembly-line job. Anemployee may be responsible for adding onepart to a product, and thus will have only avague idea of how he or she contributes tothe finished product.

Task significance represents the degree towhich performing the job is important or‘‘counts for something.’’ In a sense, all jobsin the workforce are important, but it ispossible to argue that some are more signifi-cant than others. For example, most readerswould probably agree that the task signifi-cance for a research scientist studying themolecular structure of the HIV virus ishigher than that of a clerk in a retail store.Nevertheless, this core job dimension is stilla bit more subjective than the other two.

Autonomy represents the degree towhich employees have control and discre-tion over things such as how they performtheir job tasks and schedule their work.

Comparatively speaking, college professorsrepresent a professional group with anextremely high level of autonomy. As mostreaders know, professors have considerablecontrol over their hours of work, choice ofwork activities, and method of approachingtheir work activities. At the other extrememight be a person performing telemarketing.Most telemarketing companies provide veryexplicit instructions (e.g., scripts) to telemar-keting representatives and instruct them notto deviate from these instructions.

The final core job dimension, feedback,represents the extent to which performing ajob provides information about the perfor-mance of the job incumbent. As a rule, come-dians know very quickly whether theiraudience considered a particular joke funny.Dead silence and a sea of blank stares arepretty good indicators that a joke has‘‘bombed.’’ At the other extreme, years maypass before a corporate executive receivesfeedback about certain aspects of his or herperformance. For example, the ‘‘correctness’’of a decision to enter a new market may notbe known until the company has been in thatmarket for several years.

Notice in Figure 8.3 that the core jobdimensions are immediately linked to thenext step, which is labeled critical psycholog-ical states. These states represent what em-ployees experience, on a psychological level,by performing a job with a given set of corejob dimensions. According to the model,when jobs have high levels of skill variety,task identity, and task significance, the cor-responding psychological state is experiencedmeaningfulness. Having these three dimen-sions present leads employees to psycholog-ically experience their jobs as meaningful.

The critical psychological state associatedwith autonomy is labeled felt responsibility. Ifan employee has autonomy over how he orshe performs a job, this will evoke feelings of


responsibility for the outcomes that resultfrom that work. An executive who has com-plete autonomy to determine the strategicdirection of an organization will also likelyfeel a strong sense of responsibility for thesuccess or failure of that organization. Con-versely, an employee who simply ‘‘followsorders’’ is unlikely to feel a great deal ofresponsibility for the outcomes of his orher work.

The core job dimension of feedback islinked to the critical psychological state ofknowledge of results. Thus, an employeewhose job provided considerable feedbackwill psychologically possess knowledge ofthe results of his or her performance. Conver-sely, employees who receive little feedbackhave a correspondingly vague knowledge ofthe results of their performance.

According to the next step in the model,critical psychological states are linked to per-sonal and work outcomes. This means thatexperiencing the three previously describedcritical psychological states will lead to anumber of outcomes, one of which is highinternal work motivation. Note also in Figure8.3 that the critical psychological states arealso associated with high levels of job satis-faction and performance quality, and lowlevels of absenteeism and turnover.

The final aspect of Job CharacteristicsTheory is the role of growth-need strength.Growth-need strength represents the extentto which employees see their job as a mech-anism for satisfying ‘‘growth’’ needs such aspersonal achievement and self-actualization(Alderfer, 1969; Maslow, 1943). The specificrole played by growth-need strength is thatof moderating the relations between the corejob dimensions and the critical psychologicalstates, and between the critical psychologicalstates and the personal and work outcomes.More specifically, Hackman and Oldhamproposed that the core job dimensions will

evoke the critical psychological states onlyfor those with a high level of growth-needstrength. Similarly, the theory proposed thatthe critical psychological states will lead tothe proposed personal and work outcomesonly among those who have a high level ofgrowth-need strength. For those with a lowlevel of growth-need strength, core jobdimensions will have little impact on criticalpsychological states, and these states willhave little impact on outcomes.

Over the years, Job CharacteristicsTheory has been subjected to considerableempirical testing. Fried and Ferris (1987)conducted a comprehensive meta-analysisof job characteristics studies and reported anumber of findings that were supportive ofJob Characteristics Theory. For example, allof the core job dimensions were found to berelated to outcomes such as job satisfaction,motivation, absenteeism, and turnover.Fried and Ferris’s data, however, are a bitmore equivocal with respect to the role of thecritical psychological states proposed byHackman and Oldham. Specifically, the corejob dimensions do not correlate predictablywith their proposed critical psychologicalstates. In addition, the magnitude of the cor-relations between the core job dimensionsand the critical psychological states is notstronger than with outcomes. This is impor-tant because if the critical psychologicalstates are key mediators, as proposed byHackman and Oldham, the core job dimen-sions should be more strongly correlatedwith them than with more distal outcomes(Baron & Kenny, 1986).

Although the Fried and Ferris (1987)meta-analysis is informative, very few studieshave tested the Job Characteristics Model asa whole. This was done in Hackman andOldham’s (1975, 1976) early work, but mostresearchers after that have tested only partsof the theory. One exception is a study in


which Champoux (1991) tested the entiretheoretical model utilizing canonical corre-lation analysis. The results of this study sup-ported both the causal flow of the model andthe proposed moderating effects of growth-need strength. Subsequent studies, however,have been less supportive of the moderatingeffects of growth-need strength (Evans &Ondrack, 1991; Johns, Xie, & Fang, 1992;Tiegs, Tetrick, & Fried, 1992), as well as themediating influence of the critical psycho-logical states (e.g., Renn & Vandenberg,1995). It is worth noting that cross-culturalresearch on job characteristics and job sat-isfaction has taken advantage of some of thekey distinctions presented in Hackman andOldham’s work (see Comment 8.3).

Campion’s Multidisciplinary Approach.One of the assumptions underlying the two

previously described job-based theories wasthat job content appeals to employees at apsychological level and this, in turn, resultsin positive employee outcomes. This wouldappear to be a valid assumption, but it is alsotrue that employees view their jobs frommore than just a psychological/motivationalperspective. Consistent with this point, thedesign of jobs is an issue that is of interest toother disciplines, such as industrial engi-neering, human factors/ergonomics, andbiomechanics.

Based on this notion, Campion devel-oped the Multidisciplinary Approach to JobDesign (Campion & McClelland, 1991;Campion & Thayer, 1985). Strictly speak-ing, this is not a theory of motivation, but anapproach to guide the design and redesign ofjobs. It is covered as a theory here because

COMMENT 8.3

CULTURAL DIFFERENCES IN THE IMPORTANCE OF INTRINSIC JOBCHARACTERISTICS

JOB CHARACTERISTICS THEORY emphasizes the

importance of intrinsic job characteristicssuch as autonomy and meaningfulness. Are

intrinsic job characteristics more important

determinants of job satisfaction for some cul-

tures than others? That question was

addressed in a large mutli-national study con-

ducted by Huang and Van de Vliert (2003).

These authors examined questionnaire

responses from over 107,000 employees in49 different countries. Employees in each

country responded to the extent to which

their job contained intrinsic characteristics

such as being interesting and promoting

autonomy. The country of the employees

was coded along such dimensions as Gross

National Product (GNP), individualistic ver-

sus collectivistic, and degree of power dis-

tance. The authors found that intrinsic job

characteristics were stronger predictors ofjob satisfaction for employees in richer coun-

ties, more individualist countries, and coun-

tries with smaller power distance (less

hierarchically organized). The authors rea-

soned that intrinsic work charactersitics may

not be as important predictors of job satisfac-

tion in countries where such characteristics

are not necessarily valued. This research high-lights the importance of considering country-

level influences on motivational processes.

Source: Huang, X., & Van de Fliert, E. (2003). Where

intrinsic job satisfaction fails to work: National modera-

tors of intrinsic motivation. Journal of Organizational

Behavior, 24, 159–179.


the different approaches that are describedby Campion ultimately represent a desire fordifferent end states. Thus, even though thisis typically presented only as a method of jobdesign, at its core it is really a theory ofmotivation.

According to Campion, organizationscan use four different approaches to designjobs, and each approach is associated withcertain outcomes for both individual em-ployees and the organization as a whole. Ascan be seen in Table 8.1, the motivationalapproach has been emphasized by those inpsychology and closely related fields (e.g.,human resources, organizational behavior).Recall from the previously described job-based theories that the emphasis is on mak-ing job content intrinsically interesting andmeaningful to employees. Positive outcomesassociated with this approach includeincreased job satisfaction, internal motiva-tion, higher-quality performance, and fewerwithdrawal behaviors. Designing jobs in thisfashion also comes at a cost. For example,these types of jobs are more complex andthus may require higher skill levels, longertraining periods, and higher levels of com-pensation. Such jobs may also be stressfulbecause of the high levels of responsibilityand the complexity of the interpersonalinteractions that are required.

The next job design approach presentedin Table 8.1 is mechanistic. This approachderives its roots from scientific managementand, in more modern times, has been theprovince of industrial engineering. Consis-tent with scientific management, the empha-sis in the mechanistic approach is to designjobs with maximum efficiency in mind. Jobtasks are simplified and work cycles are gen-erally made to be short. The primary benefitof the mechanistic approach is that employ-ees performing jobs designed in this way will

be efficient, particularly if one defines effi-ciency in terms of speed of production. Jobsdesigned in this fashion will also, generally,be easier to staff, and training time will beshort due to low-level skill requirements.The primary disadvantage of the mechanisticapproach is that jobs designed in this waymay foster boredom and alienation amongemployees, which ultimately could lead to anumber of counterproductive behaviorssuch as absenteeism, lack of effort, and evensabotage.

The third approach to job design pre-sented in Table 8.1 is biological. This app-roach is focused on designing jobs tomaximize the physical comfort of employees.Those trained in ergonomics and biome-chanics tend to emphasize this form of jobdesign. By emphasizing employees’ physicalcomfort in the design of jobs, organizationsmay reap important benefits such as reducedhealth care costs and lower numbers ofworkers’ compensation claims. These out-comes may ultimately translate into higherlevels of job satisfaction, but designing jobsin this fashion may require a considerableinvestment on the part of the organization.There may also be instances where employ-ees’ physical comfort detracts from their per-formance, particularly in tasks that requiresustained attention to detail and vigilance.

The fourth and final approach to job de-sign in Table 8.1 is perceptual motor. In thiscase, jobs are designed primarily with task-related information-processing demands inmind. As such, this approach tends to beemphasized primarily by those trained inhuman factors engineering. The primaryadvantage of this approach is that it may cutdown on errors and fatigue, particularly forjobs that have heavy information-processingrequirements. Airline pilots, air traffic control-lers, and anesthesiologists are three groups


whose jobs would be relevant. Despite theseadvantages, a potential drawback to this ap-proach might be high levels of boredom, ifinformation is highly simplified. Also, like thebiological approach, this approach to jobdesign may require considerable researchand development costs.

Beyond specifying the four differentapproaches to job design, the underlyingmessage in Campion’s model is that deci-sions regarding job design require that orga-nizations weigh certain costs and benefits,and ultimately make some trade-offs. Forexample, if efficiency is a very high prioritywithin an organization, the mechanistic ap-proach would probably be preferred, despiteits inherent costs. Campion’s theory alsoreminds us that organizational psychologyis not the only discipline that has somethingto contribute in the realm of job design.

Since it has been developed more re-cently, much less empirical work has beendone on the Multidisciplinary Approach thanon Job Characteristics Theory. It appears,however, that many of the premises of thistheory have been supported. For example,Campion and McClelland (1991) found thatjob changes according to the different disci-plinary orientations led to many of the pre-dicted outcomes. Also, Campion and Berger(1990) demonstrated that redesigning jobsaccording to different approaches had anumber of predicted implications for com-pensation. Over time, it is likely that moreresearch will examine many of the proposi-tions in this useful approach to job design.

Summary of Job-Based Theories

When viewed in a historical context, job-based theories represent a major theoreticalbreakthrough in organizational psychology.Prior to Herzberg, much of motivational

theory within organizational psychologywas focused on need satisfaction. Further-more, outside of the field, because of theinfluence of Scientific Management, muchof the theory and practice in motivationwas focused only on the use of financialincentives. This is not to say that financialincentives are irrelevant. Rather, the adventof job-based theories led to the realizationthat job content can have a potent impact(positive or negative) on people.

A problem that is common to all three job-based theories is that all assume, to a largeextent, that job content is an objective attrib-ute (Salancik & Pfeffer, 1977). It has beenshown, however, that ‘‘objective’’ indexes ofthe work environment often do not correlatewell with self-report measures of the sameattributes (e.g., Spector & Jex, 1991). Anotherweakness in the job-based theories describedin this section is that they are somewhat defi-cient with regard to process issues. For exam-ple, in Job Characteristics Theory, the core jobdimensions lead employees to experiencecritical psychological states, and these stateslead to a number of outcomes. The theory isnot very explicit, however, as to the reasonsunderlying these propositions. Given the timeperiod in which Job Characteristics Theorywas developed (mid-1970s), and the pro-posed moderating role of growth-needstrength, one might deduce that the mecha-nism is need satisfaction (Hackman &Oldham, 1976). However, it is also possiblethat jobs high on the core job dimensions alsohave higher levels of compensation and pres-tige. Both of these factors, rather than needsatisfaction, may be necessary for the steps inthe model. The more general point is that byfocusing so heavily on job content, thesetheories have come up a bit short on theprocesses by which job content is translatedinto outcomes.


THE PRACTICAL VALUEOF MOTIVATION THEORIES

Having reviewed what are generally consid-ered to be the major theories of employeemotivation, we now ask: How valuable arethese theories to managers in organizations?We do not believe it would be possible orworthwhile to rank-order the theories in thischapter in terms of practical value. However,it is possible to draw some general conclu-sions about the practical usefulness of thefive general types of theories described. Gen-erally, Need Theories probably fare theworst. Needs may be highly specific to indi-vidual employees, so it may be extremelydifficult for a manager to either determine agiven employee’s level of need satisfaction ortake steps to respond to it. Also, because agiven need may be satisfied in multiple ways,motivating on this basis would be quite time-consuming and cumbersome for managers.

Job-based theories, in contrast, fare con-siderably better in terms of practical value.Job content is something that most managerscan relate to, and in fact have some controlover. Thus, if a manager sees that an em-ployee lacks autonomy in his or her job,steps may be taken to increase autonomy.In a similar manner, the insights providedby SDT stress the importance of craftingjobs so that employees experience owner-ship and believe in the personal importanceof their work. On the other hand, in somecases, changing a person’s job is simplynot practical. For example, job contentmay be governed by a union contract, or per-haps changing one employee’s job wouldhave such wide-ranging effects throughoutan organization that the cost would be pro-hibitive.

Cognitive process theories may also haveconsiderable practical value, although theexact value varies considerably for each

theory. Expectancy Theory, for example,has much to offer managers in designingreward systems and in the diagnosis of per-formance problems; that is, performance-based rewards will be effective only if em-ployees are able to see a connection betweentheir performance and the level of rewardsthey attain (e.g., Instrumentality). Similarly,a performance problem may be linked to abelief that effort will make no difference(e.g., low expectancy), a belief that perfor-mance will make no difference (e.g., lowinstrumentality), or the fact that an employeesimply does not value the rewards thatan organization is providing. Goal-settingtheory has also proven to be very usefuland, in fact, is employed extensively in orga-nizations.

Equity Theory, at least in its originalform, probably has less practical valuethan Expectancy Theory and Goal Setting.Because perceptions of inputs and outcomesrepresent cognitions, they may be highlyindividualized and thus may be of little helpto managers in motivating people. Also,readers will recall that many Equity Theorypredictions are negative; that is, EquityTheory predicts that in some situations em-ployees will reduce effort, or perhaps evenleave a situation, in order to resolve feelingsof under-reward. However, for most manag-ers, motivation is a positive enterprise andtrying to prevent negative behaviors is notnearly as useful. Nevertheless, recent devel-opments and research in Equity Theoryfocusing on fairness perceptions of treatmentand policies emphasizes the negative organi-zational consequences that can occur whenemployees believe they are not being treatedfairly.

The behavioral approach to employeemotivation may also be very useful to man-agers. Principles of behaviorism are relativelyeasy for most managers to grasp, even if they


do not have behavioral science training. Par-ticularly when jobs are not highly complex, itis not difficult to determine the contingen-cies governing different behaviors. Finally,from a practical point of view, the best thingabout the application of behavioral princi-ples is that it works.

Locke and Latham (2004) have recentlyillustrated how current theories of motiva-tion could be improved. One important rec-ommendation involves integrating thedifferent motivation theories into a compre-hensive model of work motivation. Given thediversity of theories discussed in the presentchapter, this task is easier said than done.However, it is worth emphasizing that all ofthe theories proposed likely address a part ofwhat we mean by work motivation, andtherefore each theory yields novel recom-mendations for managers in order to moti-vate employees to perform well and feel goodabout their work. Well-articulated and well-supported theories provide managers withconsiderable informed guidance as theyattempt to motivate employees. If no theoriesof motivation were available, managers’attempts to motivate people would essen-tially be random, or perhaps would be basedon each manager’s idiosyncratic view of theworld. In the next chapter, we examine howthese motivation theories are applied in orga-nizations in order to influence a multitude ofemployee behaviors.

CHAPTER SUMMARY

In this chapter, we reviewed what are con-sidered the major theories of motivation inorganizational psychology. These theorieswere organized into five general categories:Need-Based Theories, Cognitive ProcessTheories, The Behavioral Approach, Self-Determination Theory, and Job-Based Theo-ries. According to Need-Based Theories, mo-

tivation is largely rooted in the human desireto satisfy needs. Theories falling under thiscategory included Maslow’s Need Hierachy,Alderfer’s ERG Theory, and AchievementMotivation Theory. In general, support forNeed-Based Theories has been rather weak,due largely to the difficulty of conceptualiz-ing and measuring needs.

Cognitive Process Theories are aimedat describing the cognitive processes involv-ed in employee motivation. These theories,for example, focus on things such asdecision making, levels of aspiration, andself-regulation. Theories discussed underthis category included Equity Theory,Expectancy Theory, Goal Setting, and Con-trol Theory. Although all of these theorieshave been supported, Goal Setting hasclearly received the greatest support andhas had the most impact within organiza-tions. In the future, as Cognitive ProcessTheories become more complex, a challengewill be to translate these into a form that canbe readily used by managers.

The Behavioral Approach to employeemotivation involves using principles adaptedfrom behaviorism in order to influence be-havior in organizations. The principle usedmost frequently is reinforcement, althoughothers, such as punishment, shaping, andextinction, may be used in certain situations.Applications of the behavioral approach inorganizations, in the form of OrganizationalBehavior Modification (OBM), have pro-duced impressive results. This approach,however, appears to work best in situationswhere the jobs being performed are nothighly complex.

Self-determination theory emphasizesthe importance of not only the quantity, oramount, of motivation, but the quality of thatmotivation. Proponents of this theory of em-ployees driven by autonomous motivation atwork will be more satisfied and perform

Chapter Summary 265

better than employees who experience theirbehavior as controlled by others. This theoryhas received a great deal of support outsideof the organizational context, and will likelybe applied to organizational psychology incoming years.

According to Job-Based Theories, thecontent of employees’ jobs is the key factorinfluencing motivation. Theories coveredunder this category included Herzberg’s Mo-tivation-Hygiene Theory, Job Characteristics

Theory, and Campion’s Interdisciplinary Ap-proach to Job Design. Job-Based Theorieshave proven quite useful and have generallybeen supported much better than Need-Based Theories. One problem that plaguesJob-Based Theories is the distinctionbetween objective and subjective attributesof jobs.

In the concluding portion of the chapter,we examined the value of motivation theo-ries to managers in organizations. This may


EDWIN LOCKE AND AN EVIDENCED-BASED THEORY OF GOAL SETTING

I first became interested in goal setting when I

read about a study done in 1935 in England

by C. A. Mace. Some of his results were

reported in Art Ryan’s and Pat Smith’s 1954

text, Principles of Industrial Psychology. Bothwere professors at Cornell where I was work-

ing towards my Ph.D. in the early 1960’s. At

the same time Ryan was writing a book Inten-

tional Behavior, eventually published in 1970,

and I got to read drafts of some of the chapters.

It was an unpleasant time in the field of

psychology, because it was then dominated by

the doctrine of behaviorism (with B. F. Skinneras its leader). This very irrational doctrine had

dominated psychology for most of the century.

It held that consciousness was an epiphenom-

enon, that human action could be understood

without reference to consciousness, and thatpeople were determined by their environment.

Neither Ryan, nor Smith, nor I accepted this

doctrine (which started to collapse in the

1970’s) but we were in the minority then.

I decided, in agreement with Ryan, that

looking at conscious goals or intentions was a

good way to begin the study of human moti-

vation—that is, by looking at proximal causes.So I did my dissertation on goal setting and

started a program of research. Gary Latham

discovered goal setting independently based on

some field research, and then we began to work

together—which we have done off and on for

over 40 years.

The most important thing we learned?

Build theories inductively based on the accu-mulation of a great many studies over a long

period of years. Don’t base them on deduction

or on just a few empirical studies. Another

important point: Always use your independent

judgment. Don’t accept doctrines or theories

just because they are popular. Look at the

evidence and draw your own conclusions.

Edwin A. LockeR.H. Smith School of Business

University of Maryland


vary from theory to theory, but it was con-cluded that, in general, motivation theoriescan be quite useful to managers. Specifically,theories provide managers with a ‘‘road map’’for motivating employees. Without motiva-tion theories, managers would be forced torely entirely on intuition and their ownimplicit theories of human behavior. Wecontinue to explore the importance of moti-vation in the next chapter as we examine thespecific tactics used by organizations toinfluence employee motivation.


George, J. M. (1995). Asymmetrical effectsof rewards and punishments: The case ofsocial loafing. Journal of Occupational andOrganizational Psychology, 68, 327--338.

Griffin, R. W., & McMahan, G. C. (1994).Motivation through work design. InJ.Greenberg (Ed.), Organizational behavior:The state of the science (pp. 23--43).Hillsdale, NJ: Erlbaum.

Latham, G. P., & Pinder, C. (2005). Workmotivation at the dawn of the twenty-firstcentury. Annual Review of Psychology, 56,485--516.

Locke, E. A., & Latham, G. P. (2002).Building a practically useful theory of goalsetting and task motivation: A 35 yearodyssey. American Psychologist, 57, 705--717.

Locke, E. A., & Latham, G. P. (2004). Whatshould we do about motivation theory?Six recommendations for the twenty-firsttheory. Academy of Management Review, 29,388--403.


Chapter Nine

OrganizationalApplications ofMotivation TheoryF

or any organization to be success-ful, employee behavior must bechanneled in directions that con-tribute positively to that success. Asexamples, a car dealership wants its

salespeople to work hard to sell cars, andan elementary school wants its teachersto strive to educate students. Organizationsalso want to prevent employees from engag-ing in behaviors that stand in the way oforganizational success. For example, a con-struction company wants to discourage itsemployees from being late to work, and anauto manufacturer wants employees to re-frain from drug use on the job.

The purpose of this chapter is to build onChapter 8 by describing the various ways inwhich organizations apply motivation theo-ries in order to influence employee behavior.It should be noted at the outset that themethods described in this chapter are notthe only ways that organizations can in-fluence behavior. Indeed, an organizationcould manipulate, coerce, and even physi-cally threaten its employees in order to influ-ence their behavior. However, in the longrun, these methods tend to have undesira-ble effects. Thus, organizations typically usemore positive methods.

This chapter describes methods that areeither directly or indirectly based on the theo-ries of motivation described in Chapter 8.An obvious advantage of doing this is thatit serves to maintain continuity from chapterto chapter. A more important reason, whichwill hopefully be brought out in this chapter,is that methods of influence that are firm-ly grounded in well-supported motivationtheories are generally more effective than

those based purely on intuition or specula-tion.

SOME BASIC ASSUMPTIONS

Before getting into specific methods that orga-nizations use to influence employees’ behav-iors, it is useful to examine some basicassumptions surrounding this process. Oneunderlying assumption, which is so basic thatwe rarely question it, is that an organizationhas the right to influence the behavior of itsemployees. In essence, the relationshipbetween organization and employee is viewed

269

as a psychological contract whereby eachis entitled to certain things (e.g., Morrison &Robinson, 1997). From an employee’s per-spective, the employment relationship typi-cally carries with it certain entitlements suchas pay, fringe benefits, and, possibly, otherperquisites. In return, an organization expectsemployees to behave in ways that benefit theorganization. When their behavior is notbenefiting the organization, employees areexpected to modify their behaviors.

Another assumption is that employeeshave at least some freedom of choice regard-ing the behaviors they are capable of engag-ing in that positively or negatively influencean organization’s performance. If employeeshad no freedom of choice, organizationswould have very little to do in the way ofmotivating their employees. In fact, if em-ployees had no freedom of choice, all anorganization would have to do is order em-ployees to behave in ways that supportedorganizational goals. This would obviouslymake life much simpler in organizations. (Itwould also make for a very short Chapter 9.)In reality, though, employees in most orga-nizations do have some level of control. Cer-tain forms of behavior (i.e., attending work)may be required to maintain organizationalmembership. The choice to go beyond thesebasic behaviors typically rests with the em-ployee.

A third assumption underlying applica-tions of motivation theory is there are nomajor internal or external constraints on em-ployees’ behaviors. Internal constraintswould be things such as a lack of job-relevantskills or abilities among employees. As wasshown in Chapter 4, motivation is only onedeterminant of productive behavior in or-ganizations. When an organization attemptsto influence employees’ behaviors throughcompensation, for example, it is assumed(not always correctly) that employees have

the skills and abilities necessary to performtheir jobs well.

External constraints, on the other hand,represent factors in the external organiza-tional environment that make it difficult foremployees to translate their skills and abil-ities into performance (Peters & O’Connor,1988). Although situational constraints is atopic that has typically been explored in theoccupational stress literature (see Chapter 7),it is relevant here as well. When organiza-tions attempt to motivate employees by pro-viding higher levels of job autonomy, forexample, an implicit assumption is that thereare no organizational conditions blockingthe increase in autonomy.

A final underlying assumption of organi-zational attempts to motivate employees isthat behavior is at least somewhat malleable.Put differently, it is assumed that people arecapable of changing their behaviors. Thisseems like a fairly commonsense notion,but the evidence in the psychological liter-ature regarding behavior change is not clear-cut. For example, Hellervik, Hazucha, andSchneider (1992) conducted an extensivereview of the behavior change literatureand concluded that, in general, empiricalevidence supports the notion that behavioris amenable to change. They were quick topoint out, however, that behavior cannot bechanged quickly or easily (see Comment 9.1).

BEHAVIORS ORGANIZATIONSATTEMPT TO INFLUENCE

Figure 9.1 contains four forms of behaviorthat are most typically targeted by organi-zational influence attempts. If we take asequential view of motivation, organizationalattempts to influence behavior begin beforeemployees actually become organizationalmembers. Specifically, organizations first tryto influence behavior during the attraction

270 Organizational Applications of Motivation Theory

stage. Through tangible means such as salaryand benefits, and more intangible means suchas promotion potential and organizationalimage, organizations strive to influenceskilled individuals to seek membership inthe organization, and ultimately to becomemembers of the organization.

Once an individual becomes an em-ployee, there are a number of behaviors thatorganizations attempt to influence. Themost visible of these, and the one that hasbeen explored the most in the motivation

literature, is productive behavior. Organiza-tions want employees to perform their basicjob tasks well and, in some cases, go beyondand perform extra-role behaviors (e.g., help-ing other employees, striving extra hard).They would also like employees to comeup with innovative and creative ideas forthe organization’s benefit, and, given theincreasing competition and rates of change,organizations often want employees to learnnew things and periodically update theirskills.

COMMENT 9.1

CAN PEOPLE CHANGE?

BEHAVIOR CHANGE IS an important issue that has

been studied and debated by psychologists

for many years. The importance of behavior

change, however, goes far beyond psychologyand other behavior sciences. For example,

people’s views about behavior change have

implications for the relationships that we

develop with others and, in many cases, pub-

lic policy decisions. Many people also spend a

considerable amount of money in order to

change what they consider to be undesirable

behaviors (e.g., smoking, overeating, beingsedentary).

What does psychological research have

to say about behavior change? Hellervik,

Hazucha, and Schneider (1992) conducted a

comprehensive review of the behavior change

literature and came up with a number of in-

teresting conclusions. The good news is that

research evidence generally supports thenotion that it is possible for people to change

behavior. Their review contained studies show-

ing evidence that people are able to change

behaviors such as level of knowledge, job

performance, safety behavior, and mental

health. However, their review also showed that

behavior change is complex and depends on a

number of factors—perhaps most importantly,

the behavior one is trying to change. For exam-

ple, it is unlikely that underlying traits such as

cognitive ability and personality traits can bechanged. On the other hand, much simpler

things, such as interpersonal skills, probably

can be modified.

Another important conclusion from this

review is that behavior change is not easy.

People have to be motivated to change, and

interventions designed to change behavior

need to be well designed and, in many cases,need to take place over a fairly long period of

time. So, yes, it is possible for people to change

some forms of behavior, but such change does

not occur overnight. Failing to recognize this

could lead to problems if organizations either

attempt to change behaviors that cannot be

modified, or fail to use proper interventions to

change other behaviors that must be modified.

Source: L. W. Hellervik, J. F. Hazucha, and R. J. Schneider.

(1992). Behavior change: Models, methods, and a review

of the evidence. In M. D. Dunnette and L. M. Hough

(Eds.), Handbook of industrial and organizational psychology

(2nd ed., Vol. 3, pp. 823–895). Palo Alto, CA: Consulting


Behaviors Organizations Attempt to Influence 271

This strong focus on performance in themotivation literature has unfortunately shift-ed the focus away from other behaviors theorganizations wish to influence. For exam-ple, organizations obviously want to discour-age employees from being absent frequently,and from engaging in a multitude of counter-productive behaviors such as theft, sub-stance use, and sabotage—to name a few.Although we typically don’t think of thesebehaviors as the focus of organizational mo-tivation programs, they really are in the sensethat organizations are trying to persuade em-ployees not to engage in them.

Another behavior that is frequently thefocus of organizational applications of moti-vation theory is retention. In comparison toother behaviors, motivating employees toretain their membership in an organizationis a bit different because it requires the orga-nization to balance a number of factors. As isoften the case with compensation, an orga-nization may be in the position of having tomake difficult choices when deciding whichemployees are worth retaining, and how

much the organization is willing to pay tokeep them. If an organization retains oneemployee by providing a large pay increase,this may very well prompt other employeesto look elsewhere. Thus, an organization isoften in the unenviable position of having toweigh the cost of internal harmony againstthe cost of a skilled employee’s leaving.

Regardless of the behavior organizationswish to influence, applying motivation the-ories involves some choice on the part of anorganization, and such choices are oftenvalue driven. For example, the founder ofan organization may make a very consciouschoice to reward his or her employees on thebasis of performance. In other cases, thevalues communicated by motivational prac-tices are far more implicit and, in some cases,are in conflict with the espoused values ofthe organization (Kerr, 1975; Lawler, 1990).Many organizations say they value perfor-mance and even institute reward systemsthat are meant to reflect this philosophy.However, despite the espoused value of per-formance, rewards in many organizations areonly very weakly related to performance.Thus, in exploring the application of moti-vation theories, we must keep in mind thatsuch applications always involve importantvalue-laden choices on the part of organiza-tions.

ORGANIZATIONAL REWARDSYSTEMS

By far, the most common method of moti-vating and influencing employee behavioris through organizational reward systems.There could potentially be an infinite num-ber of ways an organization could rewardits employees, so it is useful to distinguishbetween two types of rewards: tangible andintangible. Tangible rewards are those thatare most familiar to readers: salary, fringe

FIGURE 9.1Types of Behaviors That Are Typically the Focusof Applications of Motivation Theories inOrganizations

Attraction

Productive Behavior

Applications ofMotivation Theories

CounterproductiveBehavior

Retention


benefits, and bonuses. Intangible rewardsinclude things such as recognition, praise,and increased freedom for employees. Tan-gible rewards will be discussed first. Organi-zational reward systems draw primarily fromthe behavioral approach to motivation de-scribed in Chapter 8, with the logic beingthat rewarding desired behaviors will lead toan increased likelihood of their occurrence.

Tangible Rewards

One misconception within organizationalpsychology, perhaps due to the rise of job-based theories over the years, is that moneydoes not motivate people in the workplace(Flannery, Hofrichter, & Platten, 1996; Law-ler, 1990). Quite to the contrary, few peoplewould work for an organization for no salary.Furthermore, people engage in a variety ofillegal behaviors, ranging from selling illegaldrugs to selling government secrets, primar-ily to make money. Why is money impor-tant? In a general sense, money is obviouslyimportant because it provides the means forpeople to purchase life’s necessities andluxuries. In the workplace, employees’salaries are important because theycommunicate something about the employ-ees’ value to the organization. Within a givenorganization, if one employee has an annualbase salary of $20,000 and another employeeis paid $100,000, it is fairly evident thatthe second employee is more highly valuedthan the first. Salary is also importantbecause many people use it as at least anindirect barometer of their career success.For example, a person may define success ashaving a six-figure salary before the age of40. Given the informational and practicalvalue of money, it is not surprising thatmoney is linked to employee performanceand retention (Stajkovic & Luthans, 1997,2003).

Despite the importance of pay, it is alsotrue that pay is one of many motivating fac-tors in the workplace. In fact, when peopleare asked about the most important thingsthey are looking for in a job, pay tends to beranked lower than things such as a chance todo interesting work, and an opportunity touse their skills (Hugick & Leonard, 1991).

As an attraction mechanism, pay can behighly effective. According to Gerhart andMilkovich (1992), research evidence showsthat organizations who adopt a strategy ofpaying top dollar for talent have greater suc-cess in attracting skilled employees thanorganizations choosing not to do so. Thereasons for this would appear to be ratherobvious. When all other things are equal,many applicants will choose to work for anorganization that pays them well. In addi-tion, organizations that pay premium salariestend to develop a positive reputation; thus,more applicants will be attracted to them(see Comment 9.2).

Despite the apparent utility of paying topsalaries to attract top talent, this strategy canbe quite risky for organizations. Given thehigh payroll costs involved, those hired mustperform extremely well in order to justifythis cost (Lawler, 1990). In addition, if sev-eral organizations within the same industryadopt this strategy, salaries may be drivento a much higher level than would bewarranted by normal market forces (e.g.,scarcity of labor). This has clearly been thecase in professional sports, where salarieshave reached astronomical levels. It has be-come increasingly difficult for small marketteams to compete for talent and ultimately tobe successful. It should be noted, also, thatmany teams with huge payrolls have beenunsuccessful (see Comment 9.3).

Pay is also frequently used as a mecha-nism for motivating behaviors such as per-formance and retention in organizations.


The most common way of doing this isthrough merit pay; that is, employees receivean annual percentage increase in their pay,based on the outcome of a formal perfor-mance review (Lawler & Jenkins, 1992).Ideally, in a merit pay system, employeeswho receive the most favorable performancereviews receive the greatest percentage in-creases. From an organizational viewpoint,the hope is that employees will see the con-nection between performance and the size oftheir annual increase.

According to Lawler and Jenkins (1992),there is ample evidence that a well-designedand properly administered merit pay pro-gram can be highly effective in motivatingemployees. Merit pay systems, however, areoften not effective because they are eitherpoorly designed or administered improp-erly. A clear theme in the compensationliterature is that pay systems should bedesigned to support the strategic objectivesof an organization (Flannery et al., 1996;Lawler, 1990; Wilson, 1995). Thus, if an

COMMENT 9.2

BEING KNOWN AS A HIGH-PAYING COMPANY

IT’S CLEAR FROM the compensation literature

that organizations known to pay very high

wages tend to be more successful in recruiting

employees, compared to organizations thatare less generous. Therefore, it may be espe-

cially important for organizations to convey

this in their various recruiting tools. In Chap-

ter 3 we discussed how applicants to organi-

zations make judgments about organizations

in much the same way that consumers make

judgments about different products. Leivens

and Highhouse (2003) have emphasized thathow organizations portray themselves in

advertisements and other media may strongly

affect the ability of the organization to recruit

employees. Being known as a high-paying

company may be an important aspect of com-

pany image.

On the surface, the reason for this seems

rather obvious: Who doesn’t want a high sal-ary? Paying high wages also helps with recruit-

ing, for other reasons. For example, the fact

that an organization pays well may be seen by

potential employees as a sign that the organi-

zation ‘‘takes care of its employees,’’ and may

even give the organization a somewhat elite

image among potential applicants. Attracting a

great number of applicants may allow an orga-

nization to be highly selective, and, ultimately,

to hire the best talent available.

So why don’t all organizations attempt to

be known as high-paying companies? Onereason is that many organizations simply can’t

afford the expense. Paying premium wages is

costly, and that cost tends to compound over

time. Typically, only large organizations and

those that have been extremely successful can

afford to have such high payroll costs. Another

reason some organizations do not choose this

strategy is that wages can get out of control ifmany organizations in the same industry

choose to adopt this strategy. In fact, this

strategy may drive wages to a level that is out

of line with the skills and talents of those who

are being hired. (See the section on executive

compensation.) Furthermore, when organiza-

tions are paying everyone well, employees must

be able to contribute almost immediately, andthere may be little time for newcomers to ease

into their roles.

Source: B. Gerhart and G. T. Milkovich. (1992). Employee

compensation: Research and practice. In M. D. Dunnette

and L. M. Hough (Eds.), Handbook of industrial and orga-

nizational psychology (2nd ed., Vol. 3, pp. 481–569). Palo

Alto, CA: Consulting Psychologists Press.


organization’s strategy is focused on cus-tomer service, the merit pay system shouldencourage positive customer service behav-ior. A common mistake in many organiza-tions is that very little thought is put intoexactly what behaviors are being encouragedby the merit pay system.

For the proper administration of aperformance-based merit pay system, threefactors are particularly crucial. First, for sucha system to work, an organization must beable to accurately measure and documentperformance differences among employees.This can be done in some companies, but,more often, it is nearly impossible to dowith any level of accuracy, particularly when

jobs require a great deal of collaboration orinterdependence. If performance cannot beaccurately measured and documented, aperformance-based merit pay system will bedoomed to failure. In fact, where performancecannot be measured well, a performance-based merit system may do more harm thangood because employees may see merit paydecisions as being very arbitrary.

Second, the system must be administeredfairly; that is, employees must believe there issome validity to the performance-based paydecisions that result from the system (Eskew& Hennenman, 1996; Scarpello & Jones,1996). This is related to, but goes beyond,the first point. Fairness involves not only the

COMMENT 9.3

CAN MONEY BUY ORGANIZATIONAL SUCCESS?

BESIDES THE HIGH costs that go along with paying

high wages, this strategy is risky for another

reason: An organization paying high wages

must be very successful in order to justifythose costs.

In professional sports, this is a particularly

relevant issue, given that salaries have become

so high and success is so cut and dry. Do teams

that pay enormous salaries to players tend to be

more successful than teams paying lower sal-

aries, due either to a lack of resources or simply

a refusal to pay high salaries? There are cer-tainly a number of examples to support this

hypothesis. In major league baseball, the New

York Yankees have traditionally had one of the

higher payrolls, and they have been quite

successful of late. In professional basketball,

the Chicago Bulls were quite successful during

the period when they had one of the highest

paid players (Michael Jordan) on their team.There are, however, notable examples of

professional sports teams that have been

unsuccessful in trying to spend their way to

success. In professional football, for example,

the San Francisco 49ers in 1999 had the dis-

tinction of having one of the highest payrolls

and one of the worst records in the conference.In their case, injuries, combined with age-

related declines, contributed to this dubious

distinction. In professional baseball, the Balti-

more Orioles spent a considerable amount of

money on free-agent players in 1999, and

ended up losing nearly as many games as they

won. In addition, as discussed in Chapter 4,

Michael Lewis (2003) describes how BillyBeane, general manager of the Oakland Ath-

letics, was able to put together a top-notch team

without keeping their highest paid players.

In professional sports, as with any other

type of organization, buying a deep supply of

talent is necessary but not always sufficient for

success. Thus, organizations need to be con-

cerned not only with acquiring talent, but withthings such as how that talent will mesh to-

gether, and how to design a system in which

that talent will be best used.


accuracy of performance measurement, butalso whether employees perceive that meritincreases reflect actual performance differ-ences. This obviously mirrors the actualadministration of the merit pay system, butit also depends heavily on communication.In many cases, merit pay procedures areineffective simply because organizations failto adequately explain the basis for meritincreases.

Third, for a merit pay system to motivateperformance, the amount of money availableto fund merit increases has to be enough toallow the amount of the increases to be per-ceived as meaningful (Lawler, 1990). Whatis seen as a meaningful pay increase is some-what subjective; it depends on many factors,such as the current rate of inflation and whatother comparable organizations are paying.However, a clear trend during the past 20years has been for organizations to reducethe size of merit pay increases (Lawler &Jenkins, 1992). Thus, many merit pay sys-tems that are well designed in some respectshave only a negligible impact on employeeperformance simply because the amount ofthe merit increases is not meaningful to em-ployees.

In addition to these factors, Shaw and hiscolleagues have recently examined theimportance of positive affectivity (PA) in howemployees respond to merit pay increases(Shaw, Duffy, Mitra, Lockhard, & Bowler,2003). These authors argued that employeeshigh in PA are especially sensitive to rewards,and will therefore react positively to evenmodest pay increases. On the other hand,individuals low in PA are not especially sen-sitive to rewards, and therefore require largerincreases in merit pay to register a positivereaction. This reasoning lead to the hypoth-esis that the relationship between merit payraise size and reactions to the pay raisewould be less strong among individuals high

in PA. The authors tested this hypothesis ina longitudinal study of employees at a largehospital representing diverse job types (e.g.,physician, administrative). The authors wereable to objectively determine the amount ofmerit pay increase between the Time 1 andTime 2 assessments (which were separatedby 4 months). The authors found strongsupport for their hypotheses. There wereno relationships between size of merit payand affective (how happy they were) andbehavioral (intentions to work harder) reac-tions for individuals high in PA. These indi-viduals tended to report positive affectiveand behavioral reactions irrespective of themerit pay size. On the other hand, strongrelationships were obtained between size ofthe merit pay increase and affective andbehavioral reactions among employees lowin PA. These results emphasize the impor-tance of examining the role of personality inreactions to motivational incentives.

As an alternative or supplement to meritpay, some organizations have shifted towardpay for performance, through the use ofincentive pay. In a typical incentive pay sys-tem, an employee’s pay is directly linked to aquantifiable level of performance. The mostcommon form of incentive pay, which datesback to Scientific Management, is piece-ratecompensation. Under a typical piece-ratesystem, employees are paid a certain amountbased on the number of products or partsproduced. It is also common, in piece-ratesystems, for employees to have a chance toearn bonuses by producing at a very highlevel. For sales jobs, a familiar form of incen-tive pay is described as sales commissions.

Bonuses are another way that organiza-tions often attempt to tie pay to performance.In principle, bonuses are very similar to in-centive pay; however, bonuses are oftengiven out based on different criteria. Forexample, a manager may receive an annual


bonus that is contingent on his or her depart-ment’s meeting a given performance goal.Another difference is that bonuses are oftendistributed as lump-sum payments, whereasincentive pay is typically distributed frompaycheck to paycheck.

Compared to merit pay, an advantage ofincentive pay and bonus compensation isthat both are more concretely tied to perfor-mance. For example, a real estate agentknows that variations in his or her paycheckare linked to the number of homes sold. Amanager who receives an annual bonus typ-ically knows why it is being paid. Because ofthe timing of merit pay, it is often difficult foremployees to draw any connections betweentheir increase and their performance. An-other advantage, at least with lump-sumbonuses, is that they are more psychologi-cally meaningful to employees. When anemployee receives a 5% merit increase(which is fairly typical), this makes only anegligible difference in take-home pay and isoften taken for granted. In contrast, a lump-sum payment of 5% of one’s annual salary ismore likely to attract the attention of theemployee.

Incentive pay and bonuses are alsoadvantageous to organizations for financialreasons because these systems make it mucheasier for the organizations to link their laborcosts with their ability to pay (Lawler, 1990).For example, under incentive and bonussystems, employees are paid well when theorganization is financially successful. Duringlean years, however, incentive and bonuspayments are much lower.

At least in the case of incentive pay, re-search clearly supports a positive impacton performance. Jenkins et al. (1998) con-ducted a meta-analysis of 39 studies con-ducted in both laboratory and field settingsand found the corrected correlation betweenfinancial incentives and performance quan-

tity to be .34. Financial incentives had noimpact on performance quality. This meta-analysis also revealed that the impact offinancial incentives was greatest in experi-mental simulations and when studies wereconceptually grounded in either expectancyor reinforcement theory. It has also beenshown that financial incentives work wellin comparison to other methods of increas-ing employee productivity (Guzzo, Jette, &Katzell, 1985).

More recently, Peterson and Luthans(2006) conducted a quasi-experimentalstudy to investigate the impact of financialincentives (in the form of bonus pay) andnon-financial incentives on multiple dimen-sions of business-unit performance. Thenon-financial incentives are addressed inthe next section of the chapter. In the studythe experimental condition of financial in-centives involved individual restaurantsreceiving a bonus of increasing amountsbased on the performance of a specifiednumber of objective behaviors. Employeeswere educated about the bonus program,learning that the restaurant receiving thebonus would mean that they could expectbetween $25 and $75 being added to theirpaycheck based on engaging in the desiredbehaviors. A control group of restaurants didnot receive any intervention. The authorsmeasured multiple dimensions of perfor-mance at four time periods: before the inter-vention, and 3, 6, and 9 months followingthe intervention. Performance was assessedin terms of gross profit of the restaurant,average amount of time taken by a drive-through order (a key predictor of customersatisfaction), and turnover of employees. Theauthors found that restaurants receivingthe financial intervention did not differ fromthe control group before the intervention.However, compared to the control group,those restaurants receiving the financial


incentive evidenced a higher gross profit,faster drive-through time, and lower turn-over at all time periods after the intervention.These results reveal the strong effects of arelatively modest financial incentive.

Despite the positive aspects of incentivesand bonuses, some clear negatives are associ-ated with these methods of motivating em-ployees’ performance. Though not a problemin all incentive and bonus systems, some ofthese systems can lead to adversarial relationsbetween employees and management. Histor-ically, this has been most typical in theimplementation of piece-rate compensationsystems. In a typical piece-rate system, em-ployees are expected to produce a certainamount, considered the standard, and are paidmore if they produce more than the standard.In many companies, there is considerable dis-agreement over what is an appropriate stand-ard, and employees may intentionally slowdown production while the standard is beingdetermined. Management sometimes exacer-bates this problem by arbitrarily raising thestandard if they feel that the employees aremaking too much money. Ultimately, thefocus can be more on ‘‘beating the system’’than on increasing productivity.

Problems may also occur when some jobsin an organization are covered by incentiveand bonus systems and others are not. Theemployees who do not work under suchplans may resent the employees who do.Employees who work under such a systemmay be reluctant to accept a transfer to otherjob responsibilities for which they may bebetter suited, because they would have totake a pay cut to do so.

One thing that merit pay, incentives, andbonuses all have in common is that most aretypically based on an individual employee’sperformance. Many of today’s organizations,however, want employees to take a broaderperspective and focus their efforts on enhan-

cing the success of the work group or orga-nization. One of the most common ways thatorganizations tie rewards to organizationalperformance is through employee stock own-ership plans, commonly referred to as ESOPs(Rosen, Klein, & Young, 1986). Althoughthe ownership of stock is often associatedonly with executive compensation, manyorganizations make wider use of this formof compensation. Perhaps the most notableexample of this policy is Wal-Mart discountstores, where hourly employees have alwaysbeen allowed to purchase stock in the com-pany. In addition, UPS has used ESOPs toencourage motivation among all its employ-ees (Soupata, 2005).

From an organization’s point of view,there are many advantages to having employ-ees share an ownership stake through stockpurchases. For example, if an organization isperforming well financially, this can be anexcellent way to attract talented employees.Also, because the value of stock often appre-ciates over time, employees may lose outfinancially if they leave the organization aftera short period of time. Therefore, stock own-ership can also be a way of enhancing reten-tion. It is also possible that stock ownershipwill encourage positive attitudes and a senseof responsibility among employees (Klein,1987)—things that many organizations tryto promote by designating employees asowner-representatives. Finally, from a purelyfinancial point of view, having employeesown stock is beneficial because it is a goodway for an organization to raise capital andthus avoid hostile takeovers from larger orga-nizations or investors (Lawler & Jenkins,1992).

Hallock, Salazar, and Venneman (2004)reviewed research suggesting that ESOPswere related to reduced absenteeism andturnover, and increased employee produc-tivity, firm profitability, and firm stock


performance. These authors argued that inorder for ESOPs to serve as effective motiva-tors of performance, employees need to seesome connection between their individualperformance and the stock market perfor-mance of their organization. Therefore, Hal-lock et al. (2004) examined the determinantsof employee satisfaction with ESOPs amongemployees in a trucking firm located in theUnited States. The authors found that the pri-mary predictors of satisfaction with the ESOPwere employee’s perceived influence overstock performance, perceived influence overdecision making, and age. Employees weremost satisfied with the program when theyfelt their efforts influenced the company’sstock performance, when they had influenceover various decisions related to such issuesas working conditions and supervisors, andwhen they were older.

Although ESOPs have been shown topredict employee performance, they areprobably less likely to motivate employeeswhen compared to other financial incentivesthat establish a clearer connection betweenthe employee’s performance and an outcome(see Kraizberg, Tziner, & Weisberg, 2002).With the exception of high-level executives,most employees are unable to see a strongconnection between their own perfor-mance and the organization’s stock price.This has been particularly true in recentyears’ increased volatility in financial mar-kets. Furthermore, in most ESOPs, participa-tion is not based on performance. The onlycriterion employees must meet before theyare able to purchase shares of company stockis that they must be employed for a certainperiod of time (6 months, for example).

Organizations also attempt to motivateemployees’ organizational performancethrough profit-sharing and gain-sharing pro-grams. According to Florkowski (1987), in atypical profit-sharing program, an organiza-

tion designates a target profit margin that itwishes to achieve. When profits exceed thistarget margin, a percentage of these excessprofits is shared with employees. For exam-ple, if an organization decides that a 5%profit margin is acceptable and the actualprofit margin is 7%, a portion of the addi-tional 2% is shared with employees. Profitsharing has the potential to decrease compe-tition and enhance cooperation among em-ployees. Employees under such a systemstand to gain much more by working togeth-er for the good of the organization as awhole, rather than trying to outdo eachother. Also, like ESOPs, profit sharing mayhelp in attracting and retaining high-qualityemployees. Although there has not been agreat deal of research evaluating profit shar-ing, there is some evidence that it is associ-ated with such positive effects as enhancedorganizational productivity and positive em-ployee attitudes (Florkowski, 1987; Flor-kowski & Schuster, 1992).

Profit sharing, however, is also unlikelyto be a powerful motivator of individuals’performance. As with ESOPs, most employ-ees fail to see a strong connection betweentheir own behavior and the profitability oftheir organization. Some organizations canaddress this issue by basing profit-sharingpayments on divisional or even unit profit-ability, although this can’t always be done.The other problem with profit sharing as amotivational tool is that profit-sharing pro-grams often pay out only once or twice ayear. Thus, even if an employee is able tosee the connection between his or her per-formance and profits, the temporal lagbetween performance and the profit-sharingpayment makes it very difficult for such pay-ments to have much motivational impact.

Gain sharing is similar to profit sharing inthat some portion of pay is based on theperformance of the organization as a whole.


According to Lawler (1990), however, it isdifferent in two respects. First, the paymentsmade to employees from gain-sharingprograms are based on cost savings ratherthan on profits. For example, an organizationmight determine, based on past data, thatlosing 10% of cost, due to production defects,is acceptable. Given this 10% target, if a lowerpercentage can be achieved, some portion ofthe additional savings will go to employees.

Second, gain sharing is a comprehensiveorganizational change intervention; profitsharing is strictly a compensation program.Given the objectives of gain sharing, thismakes sense. Cost reduction typicallyrequires the efforts of individuals at manylevels of an organization’s hierarchy; thus,input from all employees is vital.

Evidence on the effects of gain-sharingprograms has been positive (e.g., Hatcher &Ross, 1991; Petty, Singleton, & Connell,1992). Specifically, such programs havebeen shown to result in significant costreductions, and may in fact have positiveeffects on employees’ attitudes. It is mucheasier for employees to see the connectionbetween their behaviors and cost reduction,as opposed to their impact on stock prices orprofit margins. Also, compared to profitsharing, gain sharing represents a more fun-damental change in management values, soemployees working under gain-sharing pro-grams may find that many other aspects oftheir work situation improve.

One potential drawback to gain sharingis that, in some cases, it is difficult for anorganization to establish cost-reductionbenchmarks. If an organization is relativelynew, for example, or has not collected a greatdeal of historical data, the cost-reductionbenchmarks that are set may be regardedby employees as being arbitrary. If this isthe case, such a plan may do a great dealmore harm than good. Gain sharing can only

work if the cost-reduction benchmarks areseen as being objective. If they are seen asbeing arbitrary, employees may feel a greatdeal of resentment.

Up to this point, all the forms of tangiblecompensation that have been describedinvolve either direct cash payments to em-ployees or the prospect of some future pay-ment. However, not all forms of tangiblecompensation involve direct cash paymentsto employees. Fringe benefits represent a sig-nificant portion of most employers’ totalcompensation costs. The most typical fringebenefits offered by employers include healthand dental insurance, some form of lifeinsurance coverage, and pension benefits.Some organizations also offer their employ-ees benefits such as 401(k) plans, visioncoverage, and tuition reimbursement.

Unfortunately, very little research hasexamined the motivational impact of fringebenefits. However, the research that has beendone suggests that their impact may be fairlyminimal. Often, employees simply lackknowledge about their organization’s fringebenefits program (Milkovich & Newman,1990), and they typically underestimate itsfinancial value (Wilson, Northcraft, & Neale,1985). Given that barrier, it is hard to imag-ine that fringe benefits would have muchmotivational value.

Fringe benefits, however, can have a pos-itive impact on the attraction and retention ofemployees (Gerhart & Milkovich, 1992). Forexample, given the high cost of health care,even small differences in health coverageplans can have significant financial implica-tions for employees. Thus, a very good healthcoverage plan may potentially provide anorganization with a competitive advantagewhen it tries to attract employees and isreasonably competitive with respect to sal-ary. The same can also be said for retirementand pension plans.


With regard to retention, pension plansprobably have the greatest impact becausethe value and portability of one’s benefitsoften depend on organizational tenure. Orga-nizations must deal with one question, how-ever: Is it desirable to attempt to retainemployees primarily on the basis of theirsunk costs in a pension program? Such plansmay help to retain employees, but employeeswho remain in an organization primarily forthat reason may not be highly productiveor particularly motivated (Meyer & Allen,1997).

Another common form of noncash com-pensation comes in the form of perquisites,more commonly know as perks. The specificperks offered by organizations vary widelyand depend, to a large extent, on the level ofeach employee. For example, for servers infast-food restaurants, typical perks includemeal discounts and free uniforms. Most retailstores offer employees discounts on mer-chandise. At the other end of the spectrum,perks for high-level executives can reachalmost outrageous proportions. For exam-ple, it is not unusual for executives to receiveperks such as country club memberships,free use of a company-owned resort, trans-portation to and from work, and travel via acorporate aircraft. Organizations have cutback on executive perks in recent years,due to changes in tax laws, but most execu-tives are still treated very well by their orga-nizations (see Comment 9.4).

As with fringe benefits, very little re-search has examined the impact of perks inorganizations. It is doubtful, however, thatperks have a great impact on employees’ day-to-day behaviors because, for most people,perks represent a relatively small portion oftheir compensation. In addition, within mostindustries, the nature of the perks providedto employees is fairly standard. For high-level employees, however, perks may make

a difference, particularly when a companyneeds their skills. For example, to lureaway an executive from a competing firm,an organization may attempt to ‘‘sweeten thedeal’’ with perquisites. This is also a verycommon practice in professional sports.For example, when the Los Angeles Dodgerssigned free-agent pitcher Kevin Brown in1998, a clause in his contract called forunlimited use of a corporate jet to fly to hishome in Georgia.

Intangible Rewards

Although organizations often reward em-ployees with tangible resources such asmoney and fringe benefits, these representonly a subset of the rewards organizationsmay use to influence employees’ behaviors.Many organizations recognize employees’performances with what can be describedas intangible rewards. An intangible rewardis defined as one from which the employeedoes not realize financial or material gain.Although clearly not as powerful as financialrewards, intangible rewards are used fre-quently and, in many cases, are highly val-ued by employees. Intangible rewards arerelevant to many of the theories of motiva-tion described in Chapter 8. They serve as asource of reinforcement in the context of thebehavioral approach, they satisfy the needsof employees for belonging and competence,and they promote a more autonomousorientation to work.

One of the most common intangiblerewards in organizations is a combination ofrecognition and awards. In some companies,recognition and citations come with tangiblerewards attached, but the tangible reward isoften less meaningful to the employees thanthe recognition and appreciation that are con-veyed. For example, many organizations for-mally recognize employees after a certain


number of years of service, or when they haveachieved some change in status, such as apromotion. Awards may also be given outfor specific work-related accomplishmentssuch as coming up with a novel work processor a cost-saving measure.

One large organization that uses recog-nition and awards to a great extent is the U.S.Military. The Armed Services have a hier-archically structured series of awards rang-ing from awards reflecting good discipline tothe Medal of Honor, awarded for the highestacts of courage and self sacrifice. When usedconsistently to reward superior perfor-mance, these awards can be effective moti-vators of service member behavior underextremely stressful conditions. Castro andhis colleagues (Castro, 2006; Thomas &Castro, 2003) have discussed the fair distri-bution of awards and recognition in support-ing service member motivation, retention,

and performance. Recently Britt and his col-leagues examined job recognition as a pre-dictor of morale among soldiers deployed ona peacekeeping mission to Kosovo (Britt,Dickinson, Moore, Castro, & Adler, 2007).The authors found that when soldiers feltthey were recognized for their contributions,they were more likely to report higher levelsof morale during the operation. Morale dur-ing the operation then predicted derivingbenefits from participating in the operationmonths after it was over. Although the judi-cious use of awards and recognition in themilitary can sustain motivation, it is alsoworth emphasizing that the administrationof awards that do not appear contingent onperformance can negatively affect motivation(see Thomas & Castro, 2003).

Another frequently used intangible re-ward is praise. For example, a supervisormay verbally praise a subordinate when a

COMMENT 9.4

EXECUTIVE PERKS: LIFE IS GOOD AT THE TOP

PERQUISITES (PERKS FOR short) are special

privileges that employees receive over and

above their salary and fringe benefits. For

most employees, perks represent a relativelysmall and insignificant portion of their

total compensation. For high-level executives,

however, perks can represent a significant

portion of total compensation and may be

quite lavish. Common perks for executives

include a car and driver, country club mem-

bership, use of company-owned vacation

properties, first-class travel privileges, andseparate dining facilities, to name a few. In

many organizations, top executives are almost

treated like royalty.

In recent years, however, organizations in

the United States have become somewhat less

extravagant in granting perks to executives.

One of the reasons for this is that, in the

mid-1990s, the Internal Revenue Service ruled

that some perks represented a form of income

to executives and thus were subject to incometax. Another reason is that more and more

organizations are trying to create an egalitarian

atmosphere in which all employees are treated

equally. Granting lavish perks only to execu-

tives serves to highlight status differences

within organizations, and thus runs counter

to an egalitarian philosophy. Finally, public

outrage over executive compensation in gen-eral has probably led organizations to be a bit

more judicious about the perks they grant to

executives.

Despite these cutbacks, perks still offer a

very comfortable existence to executives in

most organizations.


work assignment is well done. The use ofpraise should have some impact on an em-ployee’sbehaviorbecause it is likely thatpraisefrom one’s immediate supervisor has somereinforcement value (Latham & Huber,1992) and may enhance employees’ feelingsof competence (Bandura, 1986). Key issues,however, are the timing and sincerity of super-visory praise. Praise is likely to have the great-est impact when it follows closely after thedesired behavior. If a supervisor praises asubordinate for a report that was written 6months earlier, this is likely to have little effecton the employee. The effect will undoubtedlybemuchgreater if the praise isdelivered ontheday after the report is completed.

Praise from a supervisor is also muchmore effective if employees believe that it issincere. Undoubtedly, a number of factorscan determine sincerity, but two are partic-ularly important. The first has to do with thefrequency with which praise is given out. If asupervisor is constantly praising his or hersubordinates, the motivational value of thispraise will likely diminish over time. On theother hand, if praise is very rarely given out,subordinates may become highly suspiciouson those few occasions when they do receiveit. Thus, for praise to be effective, supervisorsmust strike a balance between giving toomuch or too little.

A related issue, though no less important,is the level of performance that must beachieved in order to receive praise. Ifsupervisors heap lavish amounts of praiseon subordinates for mediocre performance,this will decrease the value of praise whenhigh levels of performance are actuallyachieved. Praise is also typically more effec-tive if employees feel that they have somecontrol over the behavior for which theyare being praised (Koestner, Zuckerman, &Olsson, 1990). It is unlikely, for example,that an employee would be influenced if

praise were given for something he or shehad little control over.

In the section on tangible awards we intro-duced a study by Peterson and Luthans(2006) investigating the impact of financialincentives on the performance of restaurants.These authors also examined the effects ofnon-financial incentives on performancethrough the use of performance feedback andsocial recognition. Specifically, the authorshad managers develop a chart for employeesreflecting key performance measures (averagetime at the drive-through, cash register errors)for performance feedback. For social recog-nition, the managers were instructed to praisegroups of employees based on meeting andexceeding performance standards. Managerswere also instructed to give positively tonedweekly memos to each employee reflectinghow pleased they were with the employeesmeeting group goals.

Compared to a control group of restau-rants that received no intervention, restau-rants that received the intervention focusedon recognition and praise had higher grossprofits, faster drive-through times, andreduced employee turnover 6 and 9 monthsafter the intervention. In terms of comparingthe impact of financial versus nonfinancialincentives, at 6 and 9 months following theintervention, there were no differencesbetween the two intervention groups in termsof gross profits or drive-through times. Bothhad resulted in equivalent improvements inperformance. However, financial incentivesdid have a stronger positive impact on em-ployee retention compared to non-financialincentives.

Another intangible reward that organiza-tions sometimes use to motivate employees isstatus symbols. Status symbols are simplyways that organizations communicate an em-ployee’s worth or value to the organization.Typical status symbols include the size and


location of one’s office, an impressive-sounding title, and, in some areas, the loca-tion of one’s parking space. Unfortunately,there is very little research on the impact ofstatus symbols on employees’ behaviors.One would assume, however, that theseprobably do not have a great deal of impacton employees. Most people tend to takestatus symbols for granted, although theymay be happy with them initially. Also, sta-tus symbols often go hand in hand withother more tangible forms of compensation.

A final intangible reward that organiza-tions sometimes use to influence employees’behaviors is increased autonomy and freedom.Over time, as employees become more pro-ficient and demonstrate that they can betrusted, supervisors may grant them in-creased autonomy and freedom (Spector,Dwyer, & Jex, 1988). This may be done ina number of ways. For example, supervisorsmay give employees broad latitude on howthey perform their work, and perhaps allowemployees the freedom to choose their ownhours or even to work at home on occasion.Although supervisors granting this type ofautonomy and freedom do not see it as beinga reward, it is quite possible that subordi-nates do see it that way.

One of the clear benefits of granting in-creased autonomy and freedom is that it willlikely enhance job satisfaction and perhapsdecrease employees’ stress (e.g., Fried &Ferris, 1987; Spector, 1986). It is also possiblethat this will enhance performance and reten-tion, although little empirical evidence existsto support either outcome. Granting increasedautonomy and freedom may have a positiveeffect on performance, especially when em-ployees are highly talented and motivated.Increased autonomy and freedom may helpthese employees to reach their full potential.

Autonomy and freedom may contributeto retention as well. Often, autonomy and

freedom are perks that employees haveacquired over time by demonstrating theirtalent and loyalty within an organization.Thus, if an employee were to leave his orher present organization, it is unlikely thatthe same degree of autonomy and freedomwould be present at the new company, atleast initially. The autonomy and freedomthat employees have in a given career keepthem from taking other career paths. This isundoubtedly one of the reasons why manyuniversity professors choose to stay in aca-demia rather than pursue more lucrativecareers in business or government. Whenpeople have experienced a great deal of free-dom in their jobs, it is very difficult to give itup, even if they will be paid significantlymore elsewhere.

Executive Compensation

Up to this point, we have covered the majortypes of rewards that organizations use toinfluence employees’ behaviors. Thesesame rewards are typically used to influencethe behavior of executives, but the compen-sation of executives is quite unique, for anumber of reasons. One reason is theamount of compensation of executives, com-pared to other employees. Although exec-utive salaries vary by type of organization,it is not at all unusual for the total compen-sation of executives in private organizationsto exceed several million dollars per year(Crystal, 1995).

Other than the sheer amount of compen-sation, another difference is that executives’compensation is typically based more on theperformance of the organization than is thecompensation of other employees. In fact,executives may receive 50% or more of theirtotal compensation in the form of bonuses,or through stock options. Bonuses are oftendetermined by the profitability of the


company; typically, the company must per-form well for an executive to benefit hand-somely from exercising his or her stockoption. Thus, compared to other organiza-tional employees, a greater portion of execu-tives’ compensation is at risk if the organizationdoes not perform well (Gomez-Mejia, 1994).

Given the high level of compensationthat is received by executives, many haveraised the issue of whether this reward isdeserved (e.g., Crystal, 1991). Argumentsin favor of high executive salaries typicallycenter around two facts: (1) high-level exec-utive skills are in relatively short supply, and(2) decisions made by these individuals canhave a tremendous financial impact on anorganization and its shareholders.

On the other hand, it has also beenargued that executive compensation pack-ages have become excessive and are out ofline with the actual impact that exec-utives have on organizational effectiveness(Barkema & Gomez-Mejia, 1998; Crystal,1991). According to this argument, execu-tive skills are scarce but not scarce enoughto warrant the levels of compensation cur-rently given to executives. For example, eventhough the salaries of professional athletesare often viewed as excessive, it could beargued that the skills possessed by theseindividuals are much rarer than the skillspossessed by executives. For example, it islikely that far fewer people possess the bas-ketball skills of Michael Jordan or the base-ball skills of Alex Rodriguez, in comparisonto the administrative skills of Michael Eisner.

Another reason for treating executivecompensation separately is that, even withinthe same organization, the processes typi-cally used to set executive compensationlevels are often quite different from thoseused to determine compensation for otheremployees. According to Crystal (1991),high-level executive compensation packages

typically result from negotiations betweenthe executive and the compensation sub-committee of the organization’s board ofdirectors. This negotiation process is oftenaided by an outside compensation consul-tant who determines whether the board iscompensating the executive at a level that iscommensurate with executives employed atcomparable organizations.

Given these procedures for determiningexecutive compensation levels, there may bereasons why executives are at a distinctadvantage in this process. Members of cor-porate boards of directors are often execu-tives who inhabit the same social circles asthe organization’s executives. Such similaritymay positively bias board members in favorof the executives. Also, many board mem-bers have their compensations determined inthe same fashion, so it is certainly in theirinterest to have members of their peer groupwell compensated. Thus, members of corpo-rate boards may not be the most objectivejudges of what is a fair level of compensation.Interestingly, research has not shown arelationship between the composition ofcompensation committees and executivecompensation (Daily, Johnson, Ellstrand, &Dalton, 1998), a finding these authors attrib-ute to pressure from shareholders to curbexecutives’ salaries.

Another aspect of this process that oftenfavors executives is the use of external com-pensation consultants. As stated earlier, therole of a compensation consultant is typicallyto assess an executive’s compensation inrelation to the compensation of executivesin comparable organizations. On the surface,then, it would appear that compensationconsultants would be the most objectiveplayers in the whole process. Keep in mind,however, that compensation consultantsmust give their recommendations to execu-tives and board members who typically want


to see the level of executives’ compensationsrise. If a compensation consultant does notrecommend a highly lucrative compensationpackage, he or she may not be hired the nexttime compensation is determined (Crystal,1991).

Within organizational psychology, re-search on executive compensation is relativelynew, although some consistent findingshave begun to emerge. For example, theamount of executives’ compensation is onlyweakly related to the performance of orga-nizations (Finkelstein & Hambrick, 1988;Gerhart & Milkovich, 1990). This findingwould appear to be a bit disconcerting,given the vast amounts of money paid toexecutives. However, it could simply be astatistical artifact, due to the restriction ofrange in executive salaries. It may also bedue to the fact that even though executivesare important, their actions representone of a multitude of factors that contributeto organizational success. For example, evenif top executives make sound strategic deci-sions, these will not lead to success if anorganization lacks the necessary talent, atlower levels, to translate these decisions intohigher levels of profitability. Organizationsare also impacted by a number of externalforces that are outside of executives’ control:economic cycles, changes in governmentregulations, shifts in consumer preferences,and so on.

Another stream of research has begun toexamine the various determinants of execu-tive compensation. One fairly consistentfinding is that executive compensation ispositively related to organizational size(Finkelstein & Hambrick, 1988; Gomez-Mejia, 1994; Gomez-Mejia & Welbourne,1989). Larger organizations have greaterfinancial resources, and thus are simply ableto pay higher salaries than small organiza-tions. It has also been found that executives

appear to be paid, to a large extent, based onthe amount of discretion they have overdecision making in their organizations (Fin-kelstein & Boyd, 1998). This makes a gooddeal of sense, considering that when execu-tives have a great deal of discretion in deci-sion making, they have a much greaterchance of impacting (positively or nega-tively) organizational performance.

More recently, Wasserman (2006) exam-ined differences in financial compensationfor executives who were founders versusnon-founders of new venture companies.Wasserman noted that most prior work onexecutive compensation has been focused onagency theory, where executives are paidhandsomely to work on behalf of the orga-nization because there may be times whenthe interests of the executives and principalmembers of the organization diverge. Agencytheory is argued to be more relevant to non-founders of an organization who do not havean intrinsic interest in the viability and prof-itability of the company.

In contrast, stewardship theory refers tothe tendency for executives to be motivatedless by self-interest, and more by an intrinsicdesire for the organization to succeed. Ac-cording to Wasserman (2006), founders ofan organization are much more likelyto be ego invested in an organization, andtherefore may require less financial compen-sation to motivate behavior to maximizingthe effectiveness of the organization. Theauthor argues that founders will accrue morepsychic rewards by seeing their organizationsucceed. The motivational forces underlyingagency versus stewardship theory are similarto the distinction made in Chapter 8 betweencontrolled and autonomous motivation inthe context of Self-Determination Theory.

Wasserman (2006) conducted a study inwhich he examined the predictors of execu-tive salary (plus bonuses) among 1,238


executives from 528 private companies. Hisprimary hypotheses were that founderswould receive less financial compensationthan non-founders, that a greater numberof employees would be related to higherlevels of compensation, and that the relation-ship between number of employees andcompensation would be stronger for non-founders than founders. The logic for thelast hypothesis was that because of a found-er’s stewardship motivation, he or she wouldbe less likely to require additional compen-sation because of factors such as increasingemployees. Wasserman found support forhis hypotheses. On average, founders werepaid $25,000 less than non-founders. Thenumber of employees in a company was alsopositively related to compensation of theexecutive. Finally, the relationship betweennumber of employees and compensation wasmuch stronger for non-founders than forfounders. These results indicate that evenin the high-stakes realm of executive pay,psychological motivators of commitmentand responsibility can decrease (somewhat)the need for financial motivation.

We return now to a question that wasasked about other types of compensation:What impact does executive pay have onemployees’ behaviors? It appears fairly ob-vious that executive compensation has agreat deal of impact on attraction. In fact,without such compensation packages, mostorganizations would find it difficult to attracthigh-level executive talent. The impact ofexecutive compensation on attraction isclosely tied to its impact on retention; thatis, organizations must compensate execu-tives very well in order to retain their serv-ices. There are, however, certain ways thatexecutive compensation packages can bestructured to have a greater impact on reten-tion. For example, when granting an execu-tive stock options, a board of directors has

some discretion over the length of time thestock shares must be held before the execu-tive may sell them (Crystal, 1991). Thus, toretain an executive, a board may specify arelatively long period of time before the stockoptions may be exercised.

Another common way for corporateboards to enhance executive retention is bygranting executives so-called golden para-chutes—lucrative pension benefits that arecontingent on remaining with the organiza-tion for a given period of time. Although jobtenure is a feature built into most pensionprograms, with executives the stakes are sohigh that remaining with an organizationmay make a difference of millions of dollarsin pension benefits. However, some execu-tives may be so wealthy that the loss ofpension benefits may mean very little if an-other organization is more competitive withits offer of compensation.

Finally, does executive compensationhave any impact on performance? As statedearlier, there is a very weak connectionbetween the amount of an executive’s com-pensation and organizational performance.However, because so many factors contrib-ute to organizational performance, it is stillpossible that compensation does motivateexecutives. Executive compensation pack-ages are typically heavily loaded with stockoptions, and such packages are likely tomotivate executives to make decisions thatwill increase stock prices. This can obviouslybe done through increasing profits, but itmay also be done by more negative meanssuch as layoffs or ill-advised acquisitions.

MOTIVATION THROUGH THEDESIGN OF WORK

Although reward systems represent themajor mechanism for influencing employees’behaviors, organizations also attempt to


motivate employees through the design ofwork; that is, they try to design employees’jobs, departmental structures, and even thewhole organization itself in ways that engen-der high employee involvement and motiva-tion. In this section, we focus primarily onthe design of employees’ jobs as a motiva-tional tool used in organizations. Moremacro-level approaches to the design ofwork will be covered in Chapter 14. In dis-cussing increasing motivation through jobdesigns, a number of theories discussed inChapter 8 will be relevant. Obviously, job-based theories are the primary approachemphasized in considering how to bestdesign jobs to maximize motivation. How-ever, we will also see that Self-DeterminationTheory also underlies attempts to maximizeautonomy and personal importance at work,and that many attempts to redesign workrecognize the needs that employees have ina work context.

Job Design: A Brief History

According to Moorhead and Griffin (1998),prior to the nineteenth century, most nationswere agrarian societies. Families farmedand were largely self-sufficient. Gradually,this model of self-sufficiency gave way towhat might be described as general craft work.Specifically, people gradually reduced theirproduction of food and instead concentratedtheir efforts on the production of goods (e.g.,clothing, furniture) that were then traded forfood. Over time, this general craft model gaveway to greater specialization. For example, inthe production of clothing, people began spe-cializing in weaving, sewing, and tailoring.

The single event that had the greatestimpact on the design of work was the Indus-trial Revolution, which occurred in theUnited States in the late 1800s, and hadspread throughout Europe in the late

1700s and early 1800s. With the IndustrialRevolution came systematic study of job spe-cialization by men such as Adam Smith andCharles Babbage. Their work ultimately ledto the development of the assembly line and,eventually, to the introduction of Taylor’sScientific Management system. As readersmay recall, the primary motivational mech-anism used in Scientific Management wascompensation; employees were paid on thebasis of the amount they produced. Jobdesign was also a key part of motivatingemployees under the Scientific Managementsystem. To the extent that jobs were designedso employees could maximize their effi-ciency, this would lead to higher wagesbecause employees were being paid on thebasis of how much they produced.

Due largely to worker dissatisfactionover Scientific Management, there eventuallyemerged an approach to work design that istypically identified with the Human Rela-tions school of thought. According to advo-cates of this approach, work should bedesigned in ways that provide employeeswith an opportunity to have input and tofully maximize their skills and abilities. Amajor assumption behind this approachwas that people work not only to makemoney but also for more intrinsic reasonssuch as intellectual stimulation and creativeexpression. Although a number of work de-sign interventions came out of the HumanRelations movement, a common thread run-ning through most was that jobs weredesigned to be more interesting and to giveemployees greater discretion over work-related decisions.

Since the Human Relations era, the focushas been on refining this basic approach. Forexample, Hackman and Oldham’s (1980) JobCharacteristics Model brought greater specif-icity to the design of jobs and produced atool for the diagnosis of the motivational


properties of jobs (the Job Diagnostic Sur-vey). Campion’s interdisciplinary approachto job design has represented the meldingof approaches to job design that come fromdisciplines other than organizational psy-chology. We now consider specificapproaches to the design of work.

Humanistic Job Design

One of the simplest forms of humanistic jobdesign is job rotation. With job rotation, theactual content of jobs is not changed; how-ever, employees are allowed to periodicallyrotate among different jobs. The reasoningbehind job rotation is that performing differ-ent jobs will provide an employee withgreater variety and, perhaps, an opportunityto learn new skills. Although job rotation hasbeen shown to produce some positive out-comes and is still used (Campion, Cheraskin,& Stevens, 1994), the content of employees’jobs does not actually change. This type ofjob design is probably most effective forrelatively simple jobs that can become boringover time. For example, amusement parkswill frequently rotate employees from oneride to another in order to prevent boredomand maintain vigilance.

A second approach to job design, stem-ming directly from humanistic principles, isjob enrichment. The term job enrichment wasfirst coined by Frederick Herzberg andwas used to describe those aspects of peo-ple’s jobs that were labeled as motivators.(Recall the description of Motivation-Hygiene Theory in the previous chapter.)Motivators were linked to the content ofpeople’s work, such as the amount of controlworkers had over decisions, the level of intel-lectual challenge in the work, and the level ofcreativity workers were able to use in perfor-ming the work. Job enrichment representedan attempt to build greater levels of motiva-

tors into jobs and, hence, to make them moremotivating.

The primary mechanism used for theenrichment of jobs was termed vertical load-ing, which simply means providing employ-ees with more tasks to perform, as well asgreater freedom and discretion as to howthey perform those tasks. As an example ofhow vertical loading might work, the job of ajanitor might be vertically loaded by allowingthis individual to schedule his or her clean-ing tasks and to assume responsibility forordering new cleaning supplies. In thisway, the job has been changed by addingtasks and allowing the employee to assumethe responsibility for tasks that may havebeen previously performed by a supervisor.

Although few empirical studies haveevaluated the design of jobs according toHerzberg’s principles of job enrichment,there is some indication that this approachwas successfully implemented in some orga-nizations. For example, Ford (1973)reported positive outcomes, such as in-creased productivity and reduced turnover,resulting from the enrichment of clerical jobsat AT&T. Texas Instruments also had a greatdeal of success with the enrichment of jan-itorial jobs (Weed, 1971). Finally, Campionand McClelland (1991) reported positivemotivational consequences (e.g., increasedjob satisfaction) of enlarging employees’ jobsto provide them with more responsibility.

Not all of the early job-enrichment pro-grams were successful, however (Griffin &McMahan, 1994). One of the reasons is thatearly job-enrichment programs were con-ducted under the assumption that everyonewould be motivated by having their jobsenriched. Given the vast number of ways inwhich humans differ, this would appear tobe a rather naıve assumption at best. Earlyjob enrichment represented a rather narrowand imprecise way of designing jobs to


enhance motivation. Although vertical load-ing may make sense in some cases, there mayalso be other ways that jobs could bechanged in order to enhance motivation. Itis largely in response to these two criticismsof job enrichment that the next approach tojob design evolved.

The Job Characteristics Approachto Job Design

As you’ll recall from Chapter 8, Hackmanand Oldham’s (1980) Job CharacteristicsTheory posits that most jobs can be de-scribed according to these core dimensions:autonomy, variety, significance, feedback,and identity. Furthermore, to the extent thatthese core dimensions are present, peoplewill tend to experience their jobs as psycho-logically meaningful, feel a sense of respon-sibility about their work, and have anunderstanding of how they are performingtheir jobs. If people experience these positivepsychological states, a number of positiveoutcomes will occur, one of which is highinternal motivation. Hackman and Oldhamproposed that this model would only beapplicable for those with high growth-needstrength, and, more recently, it has beenproposed that satisfaction with the workcontext (e.g., pay, working conditions) isalso a key moderator. For over 25 years,Job Characteristics Theory has served as thetheoretical foundation for a great deal ofapplied work on job design in organizations.It is to that aspect of Job CharacteristicsTheory that we now direct our focus.

According to Hackman and Oldham(1980), an important first step in any po-tential job redesign effort is some form ofdiagnosis of the job (or jobs) within an orga-nization that might be the target of such aneffort. Fortunately, in the course of develop-ing Job Characteristics Theory, Hackman

and Oldham also developed the Job Diagnos-tic Survey (JDS), which can be used to meas-ure each of the components in the theory.The JDS has undergone a number of refine-ments over the years and, at present, is areasonably good psychometric instrument.

When used for a preliminary diagnosis,the pattern in the measures of the five corejob dimensions is most important. Let’s say,for example, that a group of employees hold-ing similar clerical positions completed theJDS, and the pattern of results resembledthose in Figure 9.2. In this case, the scoresthat we see reflect a reasonably high level ofsignificance, variety, and feedback. How-ever, it is clear that the employees perfor-ming this job feel that they lack bothautonomy and task identity. In clerical jobs,it is not unusual for incumbents to have littleautonomy over how their work is done, andthe low task identity may reflect a feeling thatthey are only contributing to the goals of theorganization in a very narrow way.

Now that a diagnosis has been done andwe have some indication of how a job mightbe changed, Hackman and Oldham (1980)recommend that the next step is to assessthe feasibility of job redesign. Although a

FIGURE 9.2Sample Profile of Core Job Dimension Scoresfrom the Job Diagnostic Survey (JDS)

JDS

Scor

es

Variety Identity Significance Autonomy FeedbackCore Job Dimensions

0

1

2

3

4

5

6

7


number of factors influence feasibility, thegeneral categories to examine are the em-ployees and the organizational system. Withrespect to employees, a major issue iswhether employees want their jobs rede-signed. Also, because job redesign often in-creases the skill requirements of jobs, theskill levels of employees must be taken intoaccount (e.g., Campion & McClelland,1991). If employees either do not want jobredesign, or possess very limited skills, jobredesign is unlikely to be successful.

Some of the organizational factors thatwould likely influence the decision ofwhether to pursue job redesign would bethe presence of union representation of em-ployees, the prevailing management philos-ophy of the organization, and the likely costof redesigning jobs. Although the presence ofa union does not automatically mean thatjobs cannot be redesigned, it does make itmuch more difficult because collective bar-gaining agreements often cover the contentof jobs that employees perform. Thus, if anorganization decides to redesign jobs, aunion may see this as the organization’sattempt to get more work out of its memberswithout raising their pay.

The prevailing management and controlmechanisms in an organization are impor-tant because job redesign interventions ofteninvolve granting employees greater discre-tion over how they perform their jobs. Thus,at some level, job redesign often requires thatmanagers and supervisors delegate some oftheir authority to subordinates. If the pre-vailing management philosophy in an orga-nization is very authoritarian, there is a goodchance that most forms of job redesign willnot work—they simply will not fit the orga-nization. Thus, the prevailing managementculture within an organization must not viewdelegation of authority negatively if job rede-sign is to work.

Redesigning jobs in an organization canbe a costly undertaking. Job redesign oftennecessitates the use of outside consultants,requires training employees in new skills andwork methods, and may introduce changesin jobs that interface with those being rede-signed. Generally speaking, if an organiza-tion must completely redesign its productionprocesses or technology in order to redesigna job or set of jobs, this may make the cost ofjob redesign prohibitive.

When the conditions in an organizationappear to favor job redesign, a major choiceorganizations face is how to implement thischange. Although it is possible to simplymandate the redesign of jobs, it is generallyadvisable to involve those organizationalmembers who are major stakeholders in thejob redesign process. A stakeholder is any-one who may be influenced by the redesignof a job. The obvious set of stakeholders inany job design effort are job incumbents—workers whose lives will be changed by theredesign of the job. Other common stake-holders are supervisors, union representa-tives, and anyone else in the organizationwho must regularly interact with job incum-bents. There is no standard way to involvestakeholders, but a common mechanism is tocreate a temporary committee or task forcecomposed of major stakeholders to provideoversight of the job redesign process.

Assuming that there is cooperation frommajor stakeholders, the next step in thisprocess is to decide on specific changes tobe made in jobs. These changes should obvi-ously be driven by the initial diagnosis. It isalso useful to have additional informationabout the job (from a job analysis, if avail-able). According to Hackman and Oldham(1980), jobs can be redesigned in a numberof different ways, depending on the core jobdimension that one is trying to enhance. Acommon method of changing jobs, and one


that is really based on the principle of jobenrichment, is vertical loading. In the JobCharacteristics Theory approach to job rede-sign, vertical loading is essentially the sameas it was in Herzberg’s approach; that is,employees are given more tasks and greaterlevels of control and discretion over howthey perform their job duties.

If one thinks back to Job CharacteristicsTheory, vertically loading a job could influ-ence nearly all of the core job dimensions,although the greatest impact is typically in-creased autonomy, because employees areprovided with higher levels of discretion.However, this intervention may potentiallyhave a positive impact on the other coredimensions of skill variety, task identity, andtask significance, as well. Skill variety is oftenenhanced because, when one has greater con-trol, this often necessitates the use of skills thatwere previously not used. Task identity maybe enhanced because vertical loading oftenmakes it possible for employees to completea whole cycle of work rather than a smallfragment. Task significance is enhancedbecause, when a job is vertically loaded, it isoften easier for an employee to see the impor-tance of the work he or she is performing.

Another common job design interven-tion based on Job Characteristics Theory iscombining tasks. This simply involves com-bining small, specialized tasks into largerunits of work. For example, a clerical em-ployee whose job has involved primarilyword processing may have filing and phonecoverage added to his or her job duties.Combining tasks undoubtedly has the great-est influence on the core job dimension ofskill variety because the employee is doingdifferent things that may require differentskills. It is possible, however, that combiningtasks may also influence task identity if thecombination of tasks represents a more inte-grated experience for the employee.

A third intervention based on Job Char-acteristics Theory is that of forming naturalwork units. This is similar to combining tasks,but, in practice, operates quite differently.Forming natural work units involves givingan employee responsibility for a logical,identifiable body of work. For example, aninsurance company could redesign the jobsof claims adjusters so that, instead of handl-ing claims on an ad hoc basis, each adjustercould have primary responsibility for theclaims of a certain group of clients. This typeof job change often allows higher levels oftask identity to be built into jobs because,when natural work units are formed, em-ployees are more likely to complete a jobfrom start to finish and to see the job as anintegrated whole. This intervention also hasthe potential to increase task significancebecause doing an integrated piece of workhelps employees to see the big picture—thatis, to understand how their work contributesto the more general mission of the organiza-tion.

A fourth job redesign intervention isestablishing client relationships. This simplyinvolves making it possible for an individ-ual who is performing a job to interfacedirectly with internal or external customers.A manufacturer may use this type of inter-vention by providing a phone number on afinished product so that, if there is a prob-lem, the customer can contact the individ-ual or team that produced the product.From the employees’ perspective, this mayenhance autonomy by giving the job asomewhat entrepreneurial feel. It also hasthe potential to enhance skill variety, sincethe skills necessary to interface with cus-tomers may be quite different from thoserequired for other aspects of the job. Estab-lishing client relationships is also an excel-lent way to build greater levels of feedbackto any job.


Connecting employees and clients mayalso provide the setting for employees to seethe beneficial effects of their work on others.Grant and his colleagues have recently devel-oped a model emphasizing how illustratingthe positive impact that employees have onothers can sustain motivation and lead to anincrease in performance (Grant, in press;Grant et al., in press). Grant refers to inter-ventions designed to increase employees’awareness of the positive impact of theirwork on others as relational job design,and provides convincing evidence thatsuch interventions can increase employeepersistence and performance. Grant et al.conducted a longitudinal study among em-ployees in a fundraising organization. Callersin the primary intervention condition inter-acted with a beneficiary of their fundraisingefforts (an undergraduate student whoreceived a scholarship as a result of fund-raising efforts). Employees in a letter-onlycontrol condition received a letter from abeneficiary containing essentially the sameinformation as the interpersonal contact.Employees in the second control conditionhad no contact with students. One monthlater the authors examined pre- to postinter-vention change in the length of phone timewith potential donors and the amount ofmoney raised. In a dramatic illustration ofthe impact of contact with the beneficiary,those in the primary intervention group in-creased their phone time by 142%, andended up raising 171% more money. Thosein the control groups did not show an in-crease on either measure. Future researchwill undoubtedly continue to examine thebenefits of relational job design on motiva-tion and performance.

A final intervention based on Job Char-acteristics Theory is opening feedback chan-nels. This simply involves redesigning a jobin a way that provides employees with an

opportunity to receive feedback on their per-formances. In manufacturing, this may sim-ply involve providing employees with accessto quality control data on a regular basis. Itmay also involve the elimination of feedbackfrom supervisors, in favor of direct feedbackto employees. This intervention is targetedspecifically at the core job dimension of feed-back.

Hackman and Oldham (1980) reportedthe results of several studies that showed thesuccessful implementation of job redesignbased on Job Characteristics Theory, andothers were published over the years (e.g.,Griffin, 1991; Griffin & McMahan, 1994;Parker & Wall, 1998). Generally speaking,redesigning a job based on Job Character-istics Theory has been shown to have a fairlyrobust effect on job satisfaction (i.e., satis-faction tends to increase when jobs are rede-signed), although the impact on actual jobperformance is somewhat mixed. For exam-ple, Griffin (1991) examined the redesign ofjobs for bank employees and found thatimmediately after the job redesign perfor-mance actually went down. Over time, how-ever, performance ultimately exceeded initiallevels. This suggests that it may take time foremployees to learn redesigned jobs, but ulti-mately redesigning jobs may enhance perfor-mance.

Despite the successes, Hackman andOldham (1980) also note instances in whichthe redesign of jobs based on Job Character-istics Theory has been unsuccessful (e.g.,Frank & Hackman, 1975). Although theremay be numerous reasons for job redesignfailures, there are likely to be some commonthemes. For example, organizations often donot anticipate the complexities of job rede-sign. A common mistake in this regard isfailing to anticipate ripple effects, or the widerimpact when a job is redesigned. For exam-ple, when a job is vertically loaded, many of


the decisions that were previously in thehands of the supervisor are passed down tothe employee. If supervisors are notinformed of this at the beginning of a jobredesign project, they may ultimately not becooperative.

A second common reason for difficultiesin job redesign is failure of the organizationto do the necessary preliminary work. Forexample, if a proper diagnosis is not con-ducted, a job may be redesigned when itdoes not need to be or when employees donot want it. An organization may also makethe mistake of failing to involve key stake-holders in the job redesign process. If, forexample, an organization were to attempt toredesign a job without consulting a union, itis very likely that the redesign would not besuccessful.

The Interdisciplinary Approachto Job Redesign

Given its relatively recent development, lesshas been written about the application of theinterdisciplinary approach, at least in compar-ison to Hackman and Oldham’s (1980) JobCharacteristics Theory. Nevertheless, this ap-proach has been used successfully in organi-zations. When job design is guided by theinterdisciplinary approach, the first step inthe process is a thorough diagnosis of thejob(s) that is being considered for redesign.In the process of developing the interdiscipli-nary approach, Campion and colleaguesdeveloped the Multimethod Job Design Ques-tionnaire to assess jobs on each of the fourdisciplinary approaches (e.g., Campion &Thayer, 1985), although research suggeststhat it measures more than four dimensions(Edwards, Skully, & Brtek, 1999). Thisinstrument can be used, for example, to high-light whether a job may be lacking on themotivational, mechanistic, biological, or per-

ceptual motor approaches to job design.Recall that the motivational approach focuseson increasing autonomy and employee par-ticipation; the mechanistic approach focuseson specialization and efficiency; the biologicalapproach focuses on preventing physicalproblems and environmental stressors; andthe perceptual approach focuses on facilitat-ing effective information processing. A diag-nosis may indicate, for example, that a job iswell designed for efficiency and speed of in-formation processing, but lacks characteris-tics that will facilitate internal motivation andphysical comfort.

If a job is lacking in one or more of thefour approaches to job design, an organiza-tion must decide whether the costs ofenhancing the job on that approach wouldbe offset by the benefits of improvement. Insome cases, the choice is relatively obvious.For example, if a job is lacking on the bio-logical approach, and several workers’ com-pensation claims have been filed as a result,an organization may have little choice but toimprove the physical comfort level of thoseperforming this job.

In other cases, however, the choices arenot as clear-cut. For example, if a job islacking on the motivational approach,enhancing it may lead to desirable outcomessuch as high job satisfaction and internalmotivation, and decreased absenteeism andturnover. Changing a job in this manner,however, may also increase the skill require-ments, which may force the organizationto pay higher wages (Campion & Berger,1990). In some cases, this is a trade-off thatis favorable for an organization; in othercases, it is not.

Assuming that a diagnosis was performedand an organization decided to enhance a jobon any of the four approaches to job design,how would the organization do it? In themotivational approach, the previous section


on the Job Characteristics Theory approachto job design answers the question. Makingchanges—such as vertical loading, combin-ing tasks, forming natural work units, estab-lishing client contact, and opening feedbackchannels—will likely maximize outcomesassociated with this approach. However,these changes may result in a number of coststo the organization, such as increased skillrequirements, longer training times, and,possibly, higher wages.

Changing jobs to enhance the mechanis-tic job design dimension is rather unfamiliarto organizational psychologists, but it is stilldone frequently by industrial engineers. Inmany cases, this involves a time and motionstudy to assess whether the job in questionhas been designed efficiently (e.g., Salvendy,1978). Such a study may reveal, for example,that the way the job is currently designedallows for a number of unnecessary motionsand efficiency is being compromised. Elimi-nation of these wasted motions may result inmuch greater efficiency; hence, productivitymay be enhanced. Often, mechanistic rede-sign also involves redesigning incentive sys-tems so that employees are motivated touse these more efficient work methods andprocedures.

Redesign of a job to enhance the biolog-ical approach may be initiated by the orga-nization or, in some cases, by an individualemployee who is having physical difficultiessuch as back problems or repetitive-motioninjuries (Hollenbeck, Ilgen, & Crampton,1992; May & Schwoerer, 1994). This typeof redesign typically begins with someform of ergonomic assessment of the job(s)in question. This involves an analysis of thejob by an individual trained in ergonomicsor, in some cases, in occupational health.What such a person would be looking fordepends on the particular job being ana-lyzed. However, common problems in ergo-

nomic job design include the existence ofrepetitive motions or the design of the workstation. For many clerical jobs, the height ofthe desk and the positioning of a worker’scomputer are key variables that one wouldassess for possible improvement. Dependingon the problem that is identified, there maybe a number of ways to enhance ergonomicjob design. For example, some jobs can bechanged by eliminating some repetitivemotions. For other problems, the solutionmay lie in the redesign or replacement ofequipment or work stations. Frequently,the solution to an ergonomic problem maybe quite expensive.

If an organization were to enhance a jobon the perceptual motor component, thiswould typically involve some analysis ofthe job by an expert trained in Human Fac-tors or Engineering Psychology (Wickens &Hollands, 2000). In analyzing the job, such aprofessional would be largely focusing onthe nature of the information the incumbentmust work with and how this information ispresented. This type of analysis sometimesindicates that the incumbent is simply beingrequired to process too much information;thus, changes may be recommended toreduce information load. Such an analysismay reveal rather straightforward changesthat can be made in a job or in the work aids(e.g., computers) associated with the job.However, changes in information presenta-tion may be quite involved and expensive.The problem may not be the amount ofinformation being processed, but the way itis being presented to the incumbent. Forexample, in the design of automobile instru-ment panels and computer software, it hasbeen found that information is easier to proc-ess when it is presented as icons or symbols,as opposed to text.

Morgeson and Campion (2002) con-ducted a comprehensive study evaluating


different forms of job redesign among em-ployees in the data analysis division of a largepharmaceutical company. These authorscompared pre- and post-intervention job sat-isfaction, training requirements, and worksimplicity. Employees within the organiza-tion received one of three interventions: anintervention geared toward increasing boththe motivational and mechanistic qualities ofthe job; an intervention geared toward max-imizing only the motivational qualities; or anintervention geared toward maximizing onlythe mechanistic qualities. The motivationalintervention focused on increasing the auton-omy, identity, skill variety, and significance ofa job. The mechanistic intervention focusedon integrating fragmented activities and elim-inating activities that detracted from the job.

One of the primary purposes for theauthors was to demonstrate that motiva-tional and mechanistic approaches werenot inevitably in conflict with each other.The authors found significant increases injob satisfaction from pretest to posttest inall three redesign groups. In the groupemphasizing both motivational and mecha-nistic improvement, a measure of job moti-vation involving autonomy and skill useincreased, but a mechanistic measure of jobmotivation did not. In line with expectations,those receiving only the motivational inter-vention increased on the motivational meas-ure, and those receiving only the mechanisticmotivation increased on the mechanisticmeasure of motivation.

Integrating Approaches to JobRedesign

A recent trend in the job design literature isthe integration of the measures and tech-niques used to redesign work in ways thatmaximize different forms of motivation andpositive outcomes for the employee and the

organization. An example of this integration isthe recent creation of the Work Design Ques-tionnaire (WDQ), a measurement instrumentthat assesses job design and the nature ofwork through combining measures based onJob Characteristics Theory (JCT) and theinterdisciplinary approach to job redesign(Morgeson & Humphrey, 2006). The meas-ure assesses 21 different work characteristics,including attributes introduced by JCT (e.g.,autonomy, task variety, task significance, taskidentity) and attributes emphasized by theinterdisciplinary approach (e.g., informationprocessing, complexity, specialization, ergo-nomics). The authors provide impressive ini-tial evidence for the measure’s reliability andvalidity.

Another sign of an integrated approach towork redesign is a recognition of the obstaclesthat face any attempt to change an organiza-tion in order to enhance employee motivationand performance. Campion and his col-leagues have recently discussed eight suchobstacles, as well as the opportunities thatresult from successfully dealing with theobstacles (Campion, Mumford, Morgeson,& Nahrgang, 2005). The eight obstacles areprovided in Table 9.1. The first obstacle hasalready been discussed in recognizing thatdifferent approaches to work design (e.g.,motivational versus mechanistic) can affectdifferent types of outcomes (e.g., satisfactionversus efficiency). This obstacle can be dealtwith by trying to redesign work in ways thatenhance both satisfaction and efficiency. Forexample, a redesign effort in a bank couldsimultaneously modify the efficiency of dataprocessing and allow employees to work on acomplete product.

The second obstacle is related to the first,and refers to trade-offs between differentapproaches to job design being inevitable. Forexample, organizational psychologists mayrealize that giving workers more autonomy


over their jobs will result in additional train-ing, which may result in greater costs for theorganization. In such situations those incharge of the redesign need to justify theadditional costs in terms of the ultimate bene-fits in employee attitudes and future perfor-mance.

The third obstacle involves difficultiesin deciding upon the most appropriate unitof intervention for the redesign effort.Should the jobs of individual employeesbe modified, or should the tasks that differ-ent employees perform be considered sep-arately for redesign? Campion et al. 2005note that redesign efforts can be attemptedat the level of the job, duties involved in ajob, tasks that make up jobs, or task clustersthat naturally go together. The authors rec-ommend task clusters as especially effectiveunits of analysis in a redesign effort,because task clusters refer to specific activ-ities engaged in by workers, reveal interde-pendencies between different jobs, and arerecognized by employees as representing awhole piece of work. Campion et al. furthernote that most jobs possess between 10 and15 different tasks. The research describedby Morgeson and Campion (2002) withpharmaceutical data analysts used the pro-cedure of designing multiple tasks intoclusters before deciding how the workwould be redesigned.

The fourth obstacle is that ‘‘It is difficult topredict the nature of a job before it exists’’(Campion et al., 2005, pg. 375). The authorspoint out that all attempts to redesign a job aregoing fromaknown job toa job that isnewandmodified. Although organizational psycholo-gists may believe that the newly designed jobwill be better for the motivation of employees,it is always possible that unexpected issueswill arise when the newly redesigned jobsare actually implemented. One way to addressthis obstacle is to recognize ahead of time theinterdependencies that exist between givenjobs within an organization and considerhow reconstituted jobs will affect each otherin the new design. It may also be necessary tomake adjustments to the newly designed jobsas unforeseen issues arise.

The fifth obstacle is that ‘‘individual dif-ferences complicate job redesign’’ (Campionet al., 2005, pg. 377). It perhaps goes with-out saying that employees differ and maytherefore respond differently to redesign ef-forts. As described earlier when discussingJob Characteristics Theory, employees differin their growth-need strength, or the impor-tance they attach to challenging and mean-ingful work. Differences between employeesmay be especially important when teamsare redesigned, as we address in more detailin Chapter 12. After reviewing research onindividual and organizational moderators of

TABLE 9.1Eight Obstacles to Using Work Redesign to Increase Employee Motivationand Performance

Obstacle #1: Work design influences multiple outcomes

Obstacle #2: Trade-offs between different work-design approaches

Obstacle #3: Difficulty in choosing an appropriate level of analysis

Obstacle #4: It is difficult to predict the nature of a job before it exists

Obstacle #5: Individual differences complicate job redesign

Obstacle #6: Job enlargement can occur without job enrichment

Obstacle #7: New jobs need to be created as part of growth or downsizing

Obstacle #8: Long-term effects may differ from short-term effects


the effects of work redesign, Campion et al.conclude that, although individual differ-ences may influence reactions to work rede-sign, they are unlikely to undermine theoverall effectiveness of interventions.

The sixth obstacle refers to the problemthat results when employees are givengreater responsibilities (i.e., their job isenlarged), but the motivational benefit ofjob enrichment does not occur. This obstaclerefers to the critical point that a discrepancycan exist between beliefs of managers andbeliefs of employees regarding whether in-creased duties and/or responsibilities make ajob more satisfying and rewarding. En-largement without enrichment can lead tocosts such as decreased satisfaction (Cam-pion & McClelland, 1993). Campion et al.(2005) stress the importance of consideringhow a job can be enriched and not justenlarged.

The seventh obstacle addresses situationsin which an organization is going through aperiod of rapid growth or downsizing. Underthese conditions, the nature of employees’jobs may change dramatically, as fewer em-ployees are available to do the work andmust take on additional roles and responsi-bilities. Campion et al. (2005) discussnumerous options for dealing with the needto combine or separate jobs in the context ofthese types of transitions. However, the mostimportant recommendation they make is tokeep employees informed of the changingroles associated with their jobs and to beclear about how their jobs are being restruc-tured.

The final obstacle addressed by Campionet al. (2005) involves the observation that thelong-term effects of a redesign effort maydiffer from the short-term effects. The authorsnote that, although efforts to redesign work toincrease motivation may result in positiveconsequences in the short-term, these bene-

fits may decrease or even reverse over a longerperiod of time. Apparently interventions thatenrich employees’ jobs are initially met withenthusiasm and interest, but over the longterm employees habituate to the new joband perhaps have mixed feelings about thedifficulty of their more complex jobs.

The obstacles and opportunities identi-fied by Campion et al. (2005) illustrate thecomplexity involved in applying motiva-tional theory to real-world jobs. However,we would argue that this complexity shouldnot deter practitioners from making use ofmotivational theory and research to enrichthe jobs of employees. One of the most im-portant areas of future research on job re-design is investigating the determinants ofeffective redesign efforts in different types oforganizations.

ORGANIZATIONALDISCIPLINARY PROCEDURES

To this point, we have covered methods bywhich organizations attempt to motivate em-ployees to engage in productive behaviors.Another use of motivation theory is to dis-courage employees from engaging in coun-terproductive behavior. In Chapter 5, oncounterproductive behavior, interventionsaimed at curbing specific behaviors such asabsenteeism, accidents, drug use, and work-place violence were briefly discussed. There-fore, the focus of this section will be on moregeneral organizational disciplinary proce-dures.

Progressive Discipline

Although specific disciplinary proceduresvary widely across organizations, it is quitecommon for organizations to have what havebeen described as progressive disciplinary pro-cedures. A progressive disciplinary procedure


indicates a progression in the severity of theconsequences when a work site has continu-ing or escalating infractions (Arvey & Jones,1985). For example, such a progressive ap-proach may be applied to safety violations ina manufacturing plant. The first safety-related violation may result in a verbal warn-ing to the offending employee. If moresafety-related violations occur, the conse-quence may increase in severity—a writtenwarning may be followed by a formal writtenreprimand, a suspension, and eventually, ifenough violations occur, dismissal.

What determines the specifics of a pro-gressive disciplinary policy? One obviousfactor is the nature of the behavior an orga-nization is attempting to discourage. Whencounterproductive behaviors are relativelymild, an organization can tolerate a numberof infractions before severe consequences arehanded out. However, for some behaviors,even one instance cannot be tolerated. Forexample, most reasonable people wouldprobably agree that an organization cannottolerate an employee assaulting a coworkeror demonstrating blatant forms of sexualharassment. For such behaviors, an organi-zation may opt for a zero tolerance policy andterminate an employee at the first infraction.

Another important factor that must beconsidered when determining a disciplinarypolicy is the legal environment. Many unioncontracts contain clauses dealing with em-ployee discipline (Bemmels & Foley, 1996).Some organizations may want to be tough oncertain forms of counterproductive behavior,but may be constrained by either the terms ofa collective bargaining agreement or a threatof litigation brought by a union. Organiza-tional disciplinary procedures must also beconsistent with state and federal laws gov-erning employment. Legislation in somestates may constrain an organization fromdisciplining specific forms of counterpro-

ductive behavior. As an example, alcoholand drug abuse meet the criteria for beingdisabilities under the terms of the Americanswith Disabilities Act of 1990. Organizationsare probably much more likely to exhibittolerance by providing employees with treat-ment for alcohol and drug abuse problems,even if they would like to adopt more puni-tive measures.

Unfortunately, relatively little is knownabout the impact of progressive disciplinaryprocedures. Despite this lack of empiricalevidence, it is likely that the effectiveness ofprogressive disciplinary procedures dependson a number of factors. One important factoris whether employees are aware of thesedisciplinary procedures. This may seem likea rather obvious point, but organizationsvary widely in their effectiveness in commu-nicating policies to employees. It is possiblethat an organization could have a progressivedisciplinary procedure and its employees aresimply unaware of it. If employees don’tknow about a policy, there is little chancethat it will impact their day-to-day behaviors.

When disciplinary policies are applied,those procedures must be fulfilled in a fairmanner (Trevino, 1992). Put differently, it isimportant to have policies applied with adegree of procedural justice. It is also impor-tant, in the implementation of disciplinaryprocedures, that employees are treated withrespect and dignity. This often involves pro-viding an employee with an opportunity totell his or her side of the story and rebut anyaccusations. The term used to describethis form of fairness is interactional justice(Colquitt et al., 2001).

In an effort to be fair, some organizationshave developed formalized grievance proce-dures to redress employees’ complaintsregarding disciplinary procedures. In manycases, grievance procedures are mandated bycollective bargaining agreements, although

Organizational Disciplinary Procedures 299

that is not always the case (McCabe, 1988).Some organizations voluntarily create griev-ance procedures that closely resemble thosecontained in collective bargaining agree-ments. Although grievance procedures varyby organization, most typically allow em-ployees to file formal grievances if they feelthey have been unfairly disciplined. Aftersuch a filing, grievance procedures normallyallow for disputes to be settled informally. Ifthis is not possible, more formal proceduresare used.

As with disciplinary policies, grievanceprocedures work best when employees per-ceive them to be fair and those who utilizethem are treated with respect (Bemmels &Foley, 1996). Grievance procedures regard-ing disciplinary actions can often be avoidedif an employee and his or her immediatesupervisor can resolve the dispute informally(Cleyman, Jex, & Love, 1995; Klaas, 1989).

CHAPTER SUMMARY

This chapter covered the most commonapplications of motivation theory in organi-zations. Without a doubt, the most widelyused mechanism that organizations use formotivating behavior is reward systems. Tan-gible rewards include merit pay, incentivepay, bonuses, fringe benefits, perquisites(perks), and status symbols. Research overthe years has shown that tangible rewardssuch as pay can be very powerful motivatorsof employee behavior. In many organiza-tions, however, the way in which thesereward systems are administered makes itvery difficult for employees to make the con-nection between rewards and performance.

Organizations also motivate employeesthrough the use of a number of intangiblerewards such as recognition, praise, and highlevels of job autonomy. Compared to theimpact of tangible rewards, much less is

known about the impact of intangible re-wards. Some research and considerable anec-dotal evidence, however, suggest that thesemay often be powerful motivators. It is doubt-ful that intangible rewards can substitute forlow levels of tangible rewards, however.

Compensation of executives was treatedas a separate topic, primarily because themanner in which executives are compen-sated differs greatly from that of rank-and-file employees. Executives’ compensationsare typically much more dependent on orga-nizational performance than are the compen-sations of other employees. Research hasshown that the size of executive compensa-tion packages is positively related to organi-zational size and to the amount of discretionexecutives have in decision making. Thus, itappears that, to a large degree, executives arepaid based on the potential impact of theirdecisions on organizational performance. Itwas also shown that executives who adoptmore of a stewardship approach to theircompanies appear less motivated by financialcompensation.

The most troubling finding in the execu-tive compensation literature is that executivecompensation is largely unrelated to organi-zational performance. This has led to veryheated public criticism of what are seen asexcess levels of compensation among execu-tives. Despite such criticisms, it is veryunlikely that the level of executive compen-sation will go down appreciably. The proce-dures that are used to determine thesecompensation practices appear to be firmlyentrenched. Also, to compete for executivetalent, organizations often have little choicebut to pay these high levels of compensation.

Other than reward systems, the othermajor method of motivating employees isthrough the design of work. The basic ideais that the content of people’s jobs may havea profound impact on whether they are


motivated. Approaches to motivationthrough the design of work have evolvedconsiderably. The oldest was described as

the humanistic approach, which was typifiedby Herzberg’s job enrichment. This ap-proach involved primarily building high


RICHARD HACKMAN AND DETOURS IN WORK DESIGN

It’s all about discovering and accepting two

things about yourself. One, what you are natu-

rally good at. And two, what drives you batty.

My major at MacMurray College was math-

ematics. I got mostly A’s in my math coursesbut only by working my tail off. Some other

math majors, I noticed to my dismay, were

getting A’s effortlessly. Then I happened to take

a social psychology elective. I got an A. Effort-

lessly. Soon thereafter I made the best decision

of my college career: I went ahead and finished

up my math major but applied for doctoral

study only to psychology programs, hopingagainst hope that I really did have some natural

talent for work in that field. That set my course.

Today, I probably would not recognize a differ-

ential equation if I tripped over one.

Friends and family tease me that I have

higher needs for autonomy and privacy than

anyone else they know—and yet I study

groups. To them, it seems to be quite a mis-

match. It’s not. I study groups because, too

often, what I most want to do when I’m in one is

to escape. Precisely because I find groups sofraught and frustrating I’ve had plenty of

energy for research about how to make them

better (to see what we’ve learned, check out my

book Leading Teams).

But wait. Why the focus on groups? Isn’t

Hackman the one who did all that research

with Oldham on work design? It’s true. But that

line of research was a side trail off my mainresearch path. In one of my first small group

studies, done while still a doctoral student at

the University of Illinois, I used a number of

different group tasks to increase the generality

of the findings. The attributes of those tasks

turned out to shape group performance more

powerfully than anything else we measured or

manipulated. Then, after I had joined thefaculty at Yale University, my colleague Ed

Lawler offered me the chance to move beyond

the experimental laboratory to assess the

impact of task attributes on job behavior in

organizations. Once again, task effects were

both powerful and provocative, and more

research clearly was called for. Greg Oldham,

then a doctoral student, joined up and webegan the series of studies that eventually

resulted in our Work Redesign book. Only after

that book was published did I return to my

main line of research on group behavior and

performance. I’ve been there ever since.

J. Richard Hackman


Harvard University

Chapter Summary 301

levels of control and discretion into jobs. Jobenrichment was applied successfully, but iteventually gave way to the job characteristicsapproach. This approach involves changingjobs in order to build in greater levels of thecore job dimensions from Job CharacteristicsTheory (e.g., job autonomy, variety, signifi-cance, identity). A recent development rele-vant to this approach is the importance ofillustrating to employees the positive impacttheir performance has on others.

Campion’s interdisciplinary approach tojob design has suggested a number of waysthat jobs can be changed to enhance a varietyof outcomes, some of which are relativelyunfamiliar to organizational psychology.The most recent trend in this area is tointegrate Job Characteristics Theory withCampion’s interdisciplinary approach inorder to redesign jobs in an attempt to max-imize both motivation and productivity.Regardless of the approach taken, it shouldalways be remembered that job redesign is acomplex undertaking that requires carefuladvance planning and, often, considerablefinancial resources.

Organizations also use motivation theoryto discourage other forms of behavior. Themost typical way of doing this is through theuse of progressive disciplinary policies. Suchpolicies differ by organization. Their actualcontent depends on factors such as the beha-vior being discouraged, collective bargainingagreements, and other legal constraints. Ulti-mately, the success of a progressive discipli-nary policy depends on how well it iscommunicated and whether it is appliedconsistently and in a fair manner.

Many organizations often develop griev-ance procedures to accompany progressivedisciplinary measures. These allow for em-

ployees to dispute disciplinary actions if theyare not considered fair. As with progressivedisciplinary procedures, the effectiveness ofgrievance procedures depends on whetherthey are seen as fair by employees. In manycases, formal grievance procedures can beavoided if supervisors and subordinates areopen to informal problem solving.


Campion, M. A., Mumford, T. V., Morge-son, F. P., & Nahrgang, J. D. (2005).Work redesign: Eight obstacles and oppor-tunities. Human Resource Management, 44,367--390.

Mitchell, T. R., & Mickel, A. E. (1999).The meaning of money: An individual dif-ference perspective. Academy of Manage-ment Review, 24, 568--578.

Mitra, A., Gupta, N., & Jenkins, G. D., Jr.(1997). A drop in the bucket: When is apay raise a pay raise? Journal of Organiza-tional Behavior, 18, 117--137.

Shaw, J. D., Duffy, M. K., Mitra, A., Lock-hart, D. E., & Bowler, M. (2003). Reac-tions to merit pay increases: A longitudinaltest of a signal sensitivity perspective. Jour-nal of Applied Psychology, 88, 538--544.

Wasserman, N. (2003). Stewards, agents,and the founder discount: Executive com-pensation in new ventures. Academy ofManagement Journal, 49, 960--976.

Westphal, J. D. (1999). Collaboration inthe boardroom: Behavioral and perform-ance consequences of CEO-board socialties. Academy of Management Journal, 42,100--110.


Chapter Ten

Leadership andInfluenceProcessesL

eadership is a topic that has been ofinterest to organizational psychol-ogists for several decades. Indeed,volumes have been written aboutleadership, though not all have

been products of organizational psycholo-gists. Authors ranging from business execu-tives to collegiate athletic coaches havewritten books about what it takes to succeedas a leader. Because much of leadershipinvolves getting things done through otherpeople, power and influence represent coreactivities of leaders. In fact, power and influ-ence are deemed so vital to leaders that someauthors have defined leadership largely as aform of influence (Yukl, 1989, 2006).

In this chapter we examine leadership aswell as power and influence processes. Cov-erage of the general approaches to leadershipis followed by descriptions of well-knownleadership theories. Consistent with recentadvances in the study of leadership, thechapter devotes much more attention to con-tingency and process approaches to leader-ship, in comparison to those that focusexclusively on the traits and behaviors ofleaders.

Compared to other treatments of leader-ship, this chapter is somewhat unique in thatpower and influence are covered in the samechapter as leadership theories. This wasdone intentionally to acknowledge that theessence of leadership is influencing otherpeople’s behavior. Whether one is leading achurch congregation, a Fortune 500 corpo-ration, or a major league soccer team, muchof what one does involves influencing others’behaviors. Furthermore, a leader’s success ininfluencing others, as well as the means by

which he or she chooses to do so, will de-pend heavily on the amount and nature ofpower held. Power and influence are clearlythe ‘‘nuts and bolts’’ of leadership.

DEFINING LEADERSHIP

If you were to pick 10 people at random andask them to define leadership, there is a goodchance that you would get a variety of defi-nitions. According to Yukl and Van Fleet(1992), leadership is difficult to definebecause of the complexity of the leadershipprocess. Because leadership involves interac-tions between leaders and subordinates

303

(typically, the members of a work group),leadership can be viewed in many ways. Forexample, we can view leadership as consist-ing of the behaviors that are enacted by thegroup leader. These may include organizingthe work, obtaining resources for the group,providing encouragement to group mem-bers, and ultimately evaluating the group’soutput (Guzzo & Shea, 1992).

On the other hand, one could just aseasily view leadership as a series of functionsthat need to be carried out in order for agroup to be effective. The nature of a group’stask may need to be clarified, resources mayneed to be obtained, the spirits of groupmembers may need lifting, and the group’soutput must eventually be evaluated. Thesefunctions can be but don’t necessarily haveto be performed by a leader. Any groupmember with relevant expertise may helpto provide task clarification, or someonewith an outgoing personality may motivateothers. By viewing leadership in this way, weare saying that it resides within groups, andnot with one specific individual.

Definitions of leadership often differ inwhether they emphasize leadership behav-iors or the results of those behaviors. Ideally,when a leader attempts to influence his orher subordinates, these individuals will dowhat the leader wants, and do it willingly.Sometimes, however, an influence attemptby a leader will result only in grudging com-pliance or may even be actively resisted bysubordinates. According to some definitionsof leadership, compliance or resistance doesnot represent true leadership. On the otherhand, according to other definitions of lead-ership, influence attempts that lead only tocompliance or resistance still represent lead-ership, albeit unsuccessful leadership.

Another issue that complicates the task ofdefining leadership is the frequent distinctionbetween leadership and management. A leader,

some have argued, is a person who obtainscommitment from his or her subordinatesand, in some cases, may even inspire them.A manager, on the other hand, is someonewho makes sure the ‘‘trains run on time’’ andprimarily obtains compliance from his or hersubordinates. A manager is someone whodoesn’t make things worse for his or her workgroup, but doesn’t get them too excited either.Interestingly, the leadership–managementdistinction is much more of an issue in thepopular leadership literature than it is amongleadership scholars. This may explain whypeople have strong feelings about the issue(see Comment 10.1).

Despite all the factors that complicate themeaning of leadership, it is possible to iden-tify some common ground among thenumerous definitions. Yukl and Van Fleet(1992) define leadership as ‘‘a process thatincludes influencing the task objectives andstrategies of an organization, influencingpeople in the organization to implementthe strategies and achieve the objectives,influencing the group maintenance andidentification, and influencing the cultureof the organization’’ (p. 149). This definitionis summarized in Figure 10.1. Vroom andJago (2007) have recently defined leadershipmore succinctly as ‘‘a process of motivatingpeople to work together collaboratively toaccomplish great things’’ (pg. 18).

There are several things to note aboutthese definitions. First, leadership involvesthe influencing of others’ behaviors. Second,leadership is viewed as a process and not asan outcome. It is possible, based on thisdefinition, for a leader to engage in unsuc-cessful influence attempts. Third, these def-initions imply that leadership requires avariety of skills. Influencing task objectivesand strategy may require strong analyticaland conceptual skills; influencing people toimplement those strategies and objectives

304 Leadership and Influence Processes

requires interpersonal and persuasive skills.Finally, leaders are frequently importantagents of change in organizations. Changingthe culture of an organization is a tall order,although it may be necessary at times if anorganization is to survive. Because of theinfluence they have, leaders are often in thebest position to facilitate cultural change.

The Importance of Leadership

What exactly do leaders do that is so impor-tant? Leaders are often needed to providestrategic direction and vision to groups and,in many cases, to entire organizations (Bass,1998). Work-group members are often toobusy with routine task completion, and with

meeting deadlines, to think about where thegroup is headed in the future. In manygroups, strategic planning and visioning ac-tivities are shared among group members,but the leader is typically the focal point ofsuch efforts. In a sense, then, leaders helporganizations to channel productive behav-ior in directions that are beneficial and thatmeet relevant strategic objectives.

Another important function of leaders,particularly those in small groups, is to en-gage in motivation and coaching behaviors.Even highly experienced employees occa-sionally need encouragement and, insome cases, help in solving difficult work-related problems. As with strategic planningand visioning, motivation and coaching

COMMENT 10.1

MANAGEMENT VERSUS LEADERSHIP

LIKE MANY AREAS in organizational psychology,

leadership has had its fair share of problems

with definition of important terms and con-

structs. One issue that often comes up,particularly among those who work in orga-

nizations, is the distinction between manage-

ment and leadership. A manager is typically

defined as an individual who engages in tra-

ditional administrative behaviors such as

planning, helping to organize the work of sub-

ordinates, and exerting control over their be-

havior. A leader, on the other hand, is a personwho not only fulfills required administrative

functions, but also is able to inspire and moti-

vate employees to strive for excellence, and, at

times, facilitates meaningful change in organi-

zations.

One of the reasons that we find this ‘‘man-

agement versus leadership’’ distinction inter-

esting is that it seems to be more of an issuefor employees, and less of an issue for leader-

ship researchers. Although recent theories of

charismatic and transformational leadership

address this issue to some degree, leadership

researchers have not focused a great deal of

effort on the issue. In contrast, I (SMJ) have

found that in courses I have taught during thepast 10 years, the issue is always raised and

discussed with a great deal of enthusiasm. To

most people, at least in my experience, man-

agers and leaders are distinct groups.

If people do indeed distinguish between

management and leadership, and have strong

feelings about it, this suggests two things to me.

First, employees in organizations want to workfor people who are true leaders and are not

there just to perform administrative duties.

Second, there is a shortage of real leaders in

organizations. There may be many reasons for

this; it may be due to the fact that real leaders

are often agents of change. If those in positions

of authority simply carry out administrative

duties, this allows an organization to maintainthe status quo, and no pressure for change is

created.

Defining Leadership 305

activities could potentially be shared amongthe members of a group. However, it is oftenmore efficient, as well as less confusingfor group members, to have one individualwho is primarily responsible for fulfillingthese functions. In most cases, that personis the leader.

A third important function of leaders inorganizations is enforcement and interpre-tation of organizational policies. For most em-ployees, leaders serve as ‘‘linking pins’’ topeople in higher levels of the organiza-tion (Likert, 1967). Because of this concept,leaders are often required to interpret andenforce organizational policies. Again, it iscertainly possible that a group could infor-mally ‘‘police itself,’’ but having a formallydesignated leader makes it much more likelythat organizationally mandated rules andprocedures will be followed.

Finally, leaders are important becausethey are typically responsible for obtainingresources for groups. Leaders essentially rep-resent the interests of their work groupswithin the broader organizational environ-

ment. Because of this, groups often relyheavily on the persuasive skills of leaders toobtain resources for task completion. With-out a leader, the members of a group may allbe trying to obtain resources and, at times,may get in each other’s way.

The four leadership functions just men-tioned are not meant to be exhaustive, butthey make a fairly compelling case for theimportance of leadership. Furthermore, withorganizations becoming flatter, skilled lead-ership is even more crucial to the success oforganizations. In flatter organizational struc-tures, leaders have a much wider span ofcontrol (i.e., they supervise a larger numberof employees), and the impact of each lead-er’s behavior is much greater than in organi-zations with a great many levels. Thus,leadership is very important, if not vital, tothe success of an organization.

GENERAL APPROACHESTO LEADERSHIP

Like many of the topics covered in this book,leadership has been of interest for centuries,although much of the early writing on lead-ership came from philosophers, historians,and political scientists. Only within the pasthalf-century have organizational psycholo-gists become heavily involved in the studyof leadership. During this time, distinct ap-proaches to the study of leadership haveevolved. In this section, we review three ofthese approaches: the trait approach, thebehavioral approach, and the contingencyapproach.

The Trait Approach

The basic premise behind the trait approachto leadership is actually quite simple: Thosewho are effective leaders possess traits thatare different from those who are less effective

FIGURE 10.1Summary of Yukl and Van Fleet’s (1992)Definition of Leadership

Development of TaskObjectives and Strategies

Implementation of TaskObjectives and Strategies

InfluenceBehavior

Group Maintenanceand Identification

OrganizationalCulture


leaders. Leadership research guided by thetrait approach is aimed primarily at identify-ing traits that discriminate between effectiveand ineffective leaders. Indeed, a good dealof early leadership research was based on thetrait approach. More exhaustive summariesof this research can be found in Mann (1959)and Stogdill (1948).

Unfortunately, early trait-based leader-ship research failed to generate a definitiveprofile of the traits that characterized ‘‘theeffective leader,’’ partly because some of the‘‘traits’’ explored by these early leadershipresearchers (e.g., physical characteristics,gender) were not based on sound theoret-ical reasoning. In addition, the aim of mostof these early leadership researchers was touse traits to distinguish effective fromineffective leaders. Given that numerousvariables influence leaders’ effectiveness, itis understandable that using traits alone topredict effectiveness met with only limitedsuccess.

Because traits did not predict leadereffectiveness well, and because, within psy-chology, emphasis shifted to environmentalinfluences on behavior, the trait approach toleadership generally fell out of favor in the1940s and 1950s. Trait-based leadership re-search was still conducted but was clearly amuch less dominant approach to leadershipthan it previously had been. Over time, how-ever, the trait approach to leadership resur-faced and made important contributions tothe study of leadership, primarily due totwo factors. First, researchers eventuallydecreased the emphasis on the predictionof leader effectiveness, in favor of predictingleader emergence. In group situations wherethere is not a formally designated leader,someone within the group eventuallyassumes the leadership role. Leadershipemergence is simply the process by whichthis occurs.

The trait approach has also made greatstrides in identifying traits that predict leaderemergence (Foti & Rueb, 1990; Zaccaro,Foti, & Kenney, 1991). Those who are moreintelligent, have higher needs for domi-nance, are high self-monitors, and aresocially perceptive tend to emerge as leaderswhen no leader has been formally desig-nated. This profile suggests that emergentleaders are able to (1) accurately ‘‘read’’ thesocial dynamics of a situation, and (2) adapttheir behavior to meet those social demands.Although not yet researched in the trait liter-ature, it is plausible that such individuals arealso more likely to end up in leadershippositions when formal selection proceduresare used. Longitudinal studies of managerialeffectiveness would certainly suggest thatthis is the case.

Second, trait-based leadership researchhas made a comeback because the traitsinvestigated in more recent research havebeen more theoretically plausible. Accordingto Yukl and Van Fleet (1992), several traitshave been identified that predict managerialeffectiveness and advancement within orga-nizations. These include a high energy level,stress tolerance, integrity, emotional matur-ity, and self-confidence. Given the nature ofmanagerial work, it is easy to see how thesetraits would be related to success, especiallywhen they are compared to things such asphysical characteristics or gender.

Zaccaro (2007) has recently developedan integrative model of how leader traits arerelated to leader emergence, effectiveness,advancement, and promotion. Zaccaro em-phasizes that leader traits include person-ality traits, cognitive abilities, motives, andvalues, and that combinations of traits arelikely to be better predictors of leader effec-tiveness than single traits considered in iso-lation. For example, Kemp, Zaccaro, Jordan,and Flippo (2004) examined the ability of


social intelligence, metacognition (aware-ness of one’s thought processes), andtolerance of ambiguity to predict leadereffectiveness among military personnel ona 3-day decision-making simulation. Theauthors found that those leaders with highlevels on all three traits performed the mosteffectively, and that low scores on any onetrait harmed performance. This line of re-search emphasizes the importance of con-sidering how leader traits act in combinationrather than in isolation.

In Chapter 4 we talked about the ‘‘BigFive’’ personality traits as predictors ofjob performance. Recall that personality re-searchers have identified five major traits:extraversion, neuroticism, conscientious-ness, agreeableness, and openness to expe-rience. One new direction in leader traitsis to examine the relationships between theBig Five and ratings of leadership and per-formance. Bono and Judge (2004) foundthat extraversion was the most consistentpredictor of transformational leadership—a type of leadership we will turn to later inthe chapter. Transformational leadershiprefers to the ability of a leader to articulatea clear and important vision that will moti-vate followers to strive to achieve. Theauthors noted that, in general, the relation-ships between the Big Five and leadershipwere rather weak. The authors suggest thatsituational variables may influence (i.e.,moderate) the relationships between theBig Five and leadership (see also de Hoogh,den Hartog, & Koopman, 2005).

Although much has been done to revivethe trait approach to leadership, there are stillmany questions that trait researchers have yetto answer. For example, what are the prac-tical implications of trait leadership theory?One would assume that the practical value ofthis approach lies mainly in the area of selec-tion for leadership positions, but that has

not been fully articulated by trait researchers.Another issue that has not been fullyaddressed by trait researchers is the impactof various combinations of traits within workgroups. What happens, for example, if agroup consists of several individuals whopossess traits indicative of leadership emer-gence? Do these individuals share leadershipfunctions, or do they compete for this role?Despite these potential shortcomings, thetrait approach, particularly in recent years,has advanced our understanding of leader-ship processes considerably.

The Behavioral Approach

Due largely to shortcomings of early traitresearch, the focus of leadership researchshifted to the behaviors that seem to dis-tinguish effective from ineffective leaders.The best-known taxonomy of leader be-havior was developed by Ralph Stogdill andEdwin Fleishman and their colleagues atOhio State University (e.g., Fleishman,Harris, & Burtt, 1955). According to theseresearchers, leadership behavior can be bro-ken down into two basic categories: (1) ini-tiating structure and (2) consideration. Leaderbehaviors that comprise the initiating struc-ture dimension are aimed at facilitating thetask performance of groups. Examples mightinclude organizing work for subordinates,communicating performance expectations,and making sure that subordinates’ behav-iors stay focused on the tasks that they areperforming.

Consideration is represented by behav-iors that are designed to show subordinatesthat they are valued and that the leader caresabout them as people. Examples of thisdimension include showing an interest insubordinates’ families, ‘‘touching base’’ withsubordinates periodically to see how thingsare going, and being compassionate whenproblems occur.


During roughly the same time periodwhen the Ohio State leadership studieswere conducted, other researchers were in-volved in efforts to provide meaningful clas-sifications of leader behavior. For instance,Rensis Likert and his colleagues at the Uni-versity of Michigan made the distinctionbetween job-centered leadership behaviorand employee-centered leadership behavior(Likert, 1961). Blake and Mouton (1964)made a similar distinction between concernfor production and concern for people in thedevelopment of their managerial grid. Notethat all of these reflect a basic distinctionbetween leader behaviors designed to facil-itate task completion, and leader behaviorsdesigned to enhance interpersonal harmonyin a group.

Despite the apparent parsimony of clas-sifying leader behaviors into two broadcategories, a number of issues were stillunresolved. For instance, some argued thatthese two dimensions were largely indepen-dent (e.g., Blake & Mouton, 1964). In otherwords, a leader could simultaneously exhibitbehaviors indicative of initiating structureand consideration. Others argued that thesetwo forms of leader behavior are negativelyrelated (e.g., Likert, 1961). For example,initiating structure behaviors were per-formed at the expense of consideration,and vice versa.

Another issue was that some leaderbehaviors were difficult to classify as strictlyinitiating structure or strictly consideration.For instance, a leader may make a point oftalking to each subordinate each day, to seehow things are going. This could certainlybe viewed as consideration because it pro-vides the leader with an opportunity toexpress concern for these subordinates.These informal chats may also help to keepsubordinates focused on their work-relatedtasks, and may provide an opportunity to

exchange important task-related informa-tion with the leader. Thus, the behaviorsleaders engage in may be more complexthan this two-dimensional classificationwould suggest.

Although the Ohio State University two-factor approach enjoyed some success ini-tially, a number of authors expressed doubtsabout the ability of consideration and initiat-ing structure to predict leader effectiveness(Yukl & van Fleet, 1992). Judge, Piccolo,and Ilies (2004) pointed out that these pes-simistic assessments were based largely onqualitative reviews of the literature, ratherthan an objective assessment of the abilityof consideration and initiating structure topredict outcomes such as follower satis-faction and leader effectiveness. Judge et al.(2004) conducted meta-analyses of 163 in-dependent correlations between initiatingstructure, consideration, and these out-comes. The authors found the overall rela-tionship between consideration and keyoutcomes, controlling for measurementerror in the variables, was .48, and that theoverall relationship between initiating struc-ture and the key variables was .29. Judgeand his colleagues point out that these rela-tionships are rather impressive, and supportthe validity of the two major categories ofleader behavior.

However, one final issue that continuesto plague the behavioral approach (and hasfrom the beginning) is that researchers werenever able to identify a set of leader behaviorsthat were consistently associated with effec-tiveness across all types of situations. Thissuggests that there is no universal set of leaderbehaviors that will result in leader effective-ness in all situations. Rather, the behaviorsthat are needed from a leader will vary fromsituation to situation. This realization led tothe contingency approach to leadership,which will be described next.


The Contingency Approach

The contingency approach is based on theassumption that the relationship betweenleader behaviors and traits and effectivenessdepends on characteristics of the particularsituation the leader is in. The task of a leader,according to the contingency approach, is tofirst ‘‘read’’ the situation to determine whatbehaviors would be most appropriate. Oncethis is determined, the leader has to adjusthis or her behavior to meet the demands ofthe situation.

To illustrate how the contingency ap-proach works in practice, let’s say that aleader has been asked to take charge of agroup consisting of five highly skilled andexperienced design engineers. In this typeof situation, the leader would probably nothave to do a great deal of teaching andperformance-related coaching. In fact, ifthe leader tried to do this, the group mem-bers might consider him or her an annoy-ance. Instead, the leader in this situationwill be more effective if he or she concen-trates on obtaining resources for the group,facilitates professional development activi-ties for group members, and periodicallymakes an effort to boost the morale of thegroup.

Now consider a different leader who isin charge of a group of five design engineerswho are all recent college graduates. A gooddeal of this leader’s behavior will be focused ontask clarification, teaching, and performance-related coaching. In a group like this, theseactivities would not be considered anannoyance at all; in fact, they would prob-ably be welcomed. To be effective in thissituation, a leader would have to be very‘‘hands on’’ with his or her subordinates. If aleader in this situation spent the bulk of hisor her time negotiating for resources within

the organization, or remained very distantfrom the group members, he or she wouldprobably not be successful.

Most leadership theories developedduring the past 30 years are contingencytheories. Thus, it is accurate to say that thefield of leadership has accepted the generalpremise behind contingency theories. Lessconsensus, however, has been given tomany of the specifics of the contingencyapproach. For example, there is not a greatdeal of consensus regarding the specificaspects of the situation that leaders must‘‘read’’ in order to adjust their behaviors.For example, several contingency theoriespropose that ‘‘subordinates’’ are one suchfactor, but there is not a great deal of agree-ment on what specific aspects of subordi-nates are the most important.

Another area of disagreement sur-rounding contingency theories has to dowith the behaviors that leaders mustexhibit in order to be successful. As read-ers will see, contingency theories differ inthe level of adaptability they ascribe to theleader. In some theories (e.g., Fiedler,1967), it is proposed that leaders have apredetermined leadership style that is notsubject to a great deal of modification.Other contingency theories (e.g., House,1971), however, propose that leaders arefully capable of adapting their behaviors todifferent situations. This really speaks tothe more basic issue of the malleability ofbehavior, which was discussed in the pre-vious chapter (e.g., Hellervik, Hazucha, &Schneider, 1992). Based on that literature,the weight of the evidence suggests thatleaders are capable of modifying theirbehaviors to meet situational demands.What is not nearly as clear is what leadersare specifically supposed to do in responseto the situations they face.


MODERN THEORIESOF LEADERSHIP

Most leadership theories developed withinthe past 30 years can be classified as con-tingency theories. In this section, we exam-ine the contingency leadership theories thathave been most influential in the leadershipliterature. Influence is defined in terms of theresearch generated by the theories, as well asthe impact the theory has had on the practiceof leadership within organizations.

Fiedler’s Contingency Theory

The basic premise behind Fiedler’s contin-gency theory is actually quite simple. Likeall contingency theories, it proposes thatthe success of a leader depends on theinteraction between characteristics of thesituation and characteristics of the leader.According to Fiedler, situation favorabilitydepends on the three factors illustrated inFigure 10.2. The first of these, leader–member relations, reflects the extent towhich a leader gets along well with his orher subordinates. Generally speaking, sit-uations are more favorable for leaders whenthey get along well with subordinates, and,conversely, less favorable if leader–memberrelations are poor.

The next situational attribute, task struc-ture, reflects whether the subordinates areworking on a task that is very straightforwardand structured (e.g., produce 50 cars per day),or whether the task is vague and unstruc-tured (e.g., ‘‘Develop innovative products’’).Although subordinates may find a lack ofstructure challenging, from a leader’s per-spective having a high degree of structure ismore favorable than having a low degree.When task structure is high, the leader isrequired to spend less time clarifying the task

for subordinates, and decisions are typicallymuch easier to make.

The third determinant of situation favor-ability is the position power of the leader—the amount of formal authority that a leaderhas over his or her subordinates. Somedegree of authority is inherent in all leader-ship positions, but the amount of authorityactually varies considerably. Some leadersare granted the authority to assign subordi-nates to different jobs, to evaluate their work,and to dismiss those who are not performingwell. However, leadership positions do notalways carry a great deal of authority. A goodexample is the chairperson of an academicdepartment. A chairperson is technically ‘‘incharge’’ of an academic department, but thisperson has very little formal authoritybeyond that of supervisors in many othertypes of organization.

From a leader’s perspective, a high ratherthan a low position power is desirable. Whenposition power is high, subordinates willtypically do what the leader wants, and theleader does not have to exert a great deal offorce over employees. When a leader’s posi-tion power is low, subordinates may still dowhat the leader wants, but the leader mayhave to expend a great deal of effort in orderto make that happen. Consider, for example,the chairperson of an academic department

FIGURE 10.2Determinants of Situation Favorability in Fiedler’sContingency Theory

TaskStructure

Leader-MemberRelations

PositionPower

Situation Favorability


who is trying to persuade a tenured facultymember to teach a class that this individualdoes not want to teach. The chairpersonmust spend time and effort to persuade thisindividual to teach the course, and perhapsmay have to offer something in return (acourse release in the future).

Given these three situational attributes,and the fact that each has two levels, it ispossible to come up with eight unique sit-uations (called octants) in terms of favorabil-ity. These are illustrated in Figure 10.3. Themost favorable situations for leaders arethose in which leader–member relations aregood, task structure is high, and positionpower is high. In this type of situation, aleader gets along well with his or her sub-ordinates, is directing a group of employeesworking on a well-defined task, and has agreat deal of formal authority. From a lead-er’s perspective, what could be better? Aleader can then spend his or her time onactivities such as strategic planning, acquir-ing resources for the group, andperhaps helping subordinates to developtheir skills.

At the other end of the spectrum, theleast favorable situations for leaders are thosein which leader–member relations are poor,task structure is low, and the leader has verylow position power. From a leader’s perspec-tive, what could be worse? The fact that theleader does not get along well with his or hersubordinates is likely to be unpleasant. How-ever, when combined with a very vague andunstructured task and a very low level ofauthority, this is even worse. A leader in thissituation may have to spend the bulk of hisor her time trying to influence or negotiatewith subordinates in order to get anythingaccomplished. Furthermore, there is noguarantee that such influence attempts willbe successful. The leader will have consid-erably less time available for things such as

FIGURE 10.3Summary of the Eight Octants Which RepresentDiffering Degrees of Situation Favorability

L-M Relations (P)Task Structure (L)Position Power (L)

Low Situation Favorability

Moderate Situation Favorability

High Situation Favorability

L-M Relations (G)Task Structure (L)Position Power (L)

L-M Relations (P)Task Structure (H)Position Power (L)

L-M Relations (P)Task Structure (L)Position Power (H)

L-M Relations (G)Task Structure (H)Position Power (L)

L-M Relations (G)Task Structure (L)Position Power (H)

L-M Relations (P)Task Structure (H)Position Power (H)

L-M Relations (G)Task Structure (H)Position Power (H)

P = PoorG = Good

H = HighL = Low


strategic planning, resource acquisition, oremployee development.

In between these extremes are six othersituations that Fiedler referred to as havingmoderate favorability for the leader. In theinterest of brevity, all of these moderatelyfavorable situations will not be described.However, as an example of a moderatelyfavorable situation, a leader may have goodleader–member relations, high task struc-ture, and low position power vis-a-vis hisor her subordinates. From the leader’s pointof view, these situations are inherently morecomplex than situations of either very highor very low favorability.

The second portion of Fiedler’s theoryhas to do with the characteristics of theleader. According to Fiedler, leaders canbe reliably distinguished in terms ofwhether they are task-oriented versusrelationship- oriented. To measure taskversus relationship orientation in leaders,Fiedler and his colleagues developedthe Least Preferred Coworker (LPC) Scale(Fiedler, 1967). As can be seen in Table10.1, the LPC Scale consists of 18 pairs ofadjectives. Respondents completing thisscale are asked to think of a person withwhom they currently work or have workedin the past, and with whom they have hadthe most difficulty in getting work done. Ahigh LPC score indicates that a leader hasdescribed his or her least preferred co-worker in relatively favorable terms. Thisindicates that the leader is relationshiporiented because he or she is able to ratethis coworker favorably, even though theindividual is not seen as someone whowould facilitate task accomplishment. Incontrast, a low LPC score indicates thatthe least preferred coworker is describedin relatively unfavorable terms. This indi-cates that the leader is task oriented, ac-cording to Fiedler, because this coworker’s

negative impact on task accomplishmentoverrides any positive qualities this personmay possess.

Fiedler proposed that leaders who are taskoriented (herein referred to as Low LPC lead-ers) are most successful in either highly favor-able or highly unfavorable situations. Inhighly favorable situations, a Low LPC leaderwill basically leave things alone and not try tointroduce major changes. He or she will alsonot try to ‘‘get into people’s heads’’ and be-come very close to them interpersonally. Thistype of leader behavior simply is not needed.In contrast, when situations are highly unfa-vorable, a Low LPC leader is probably the onlytype that will get anything done. In thesesituations, a High LPC leader’s attempts todevelop strong interpersonal ties will likelyfall flat and will ultimately reduce the chancesof any form of task accomplishment.

When situations are moderately favor-able, Fiedler proposed that leaders who arerelationship oriented (herein referred to asHigh LPC leaders) are most effective. Thelogic here is that moderately favorable situa-tions are not ‘‘black and white.’’ Such situa-tions often require some interpersonalfinesse, and a High LPC leader has this trait.Let’s say, for example, that a leader is in amoderately favorable situation: Leader–member relations are good, but task struc-ture and position power are low. A High LPCleader is needed because the leader may haveto rely heavily on his or her relationshipswith subordinates in order to clarify the taskand ultimately get things done. A Low LPCleader would be unsuccessful in this situa-tion, primarily because he or she may not seethe complexities in the situation and maysimply demand performance. The relation-ship between LPC and situational favorabil-ity is summarized in Figure 10.4.

Considerable research has been done onFiedler’s contingency theory over the years,


TABLE 10.1Least Preferred Coworker (LPC) Scale (Fiedler, 1967)

Over the course of your life you have probably worked in many groups with other people on your job, in communitygroups, church groups, athletic teams, etc. Some of your coworkers may have been very easy to work with in attainingthe group’s goal, while others were less easy to work with.

Think of the person in your life with whom you worked least well. He or she may have been someone you knew in thepast or someone you work with now. The person does not have to be the person you like least well, but should be theperson with whom you have the most difficulty getting the job done. In this scale you will be describing this person. Youdo not need to give the person's name.

Following are pairs of words which are opposite in meaning, such as “Very Neat” and “Not Neat.” Between each pair ofwords are eight blanks to form a scale.

EXAMPLE: In describing the person with whom you least like to work, if you ordinarily think of him or her as being“Quite Neat,” you would put an “X” in the space marked 7.

If you ordinarily think of this person as being only “Somewhat Neat,” you would put your “X” in the space above the 6.

If you think of this person as being “Slightly Untidy,” you would mark the space above the 4.

If you would think of this person as being “Very Untidy” (or not neat), you would put your “X” in space 1.

Look at the words at both ends of the line before you mark your “X.” Work rapidly, your first answer is likely to be yourbest one (there are no right or wrong answers, though).Please do not omit any items, and mark each item only once.

Now use the scale to describe the person with whom you find it hardest to get the job done.

: : : : : : : : :Pleasant Unpleasant8 7 6 5 4 3 2 1

: : : : : : : : :Friendly Unfriendly8 7 6 5 4 3 2 1

8 7 6 5 4 3 2 1

8 7 6 5 4 3 2 1

1 2 3 4 5 6 7 8

1 2 3 4 5 6 7 8: : : : : : : : :Rejecting Accepting

1 2 3 4 5 6 7 8: : : : : : : : :Tense Relaxed

1 2 3 4 5 6 7 8: : : : : : : : :Distant Close

1 2 3 4 5 6 7 8: : : : : : : : :Cold Warm

: : : : : : : : :Supportive Hostile

: : : : : : : : :Boring Interesting

1 2 3 4 5 6 7 8: : : : : : : : :Quarrelsome Harmonious

1 2 3 4 5 6 7 8: : : : : : : : :Gloomy Cheerful

: : : : : : : : :Open Guarded

8 7 6 5 4 3 2 1: : : : : : : : :Considerate Inconsiderate

8 7 6 5 4 3 2 1: : : : : : : : :Agreeable Disagreeable

8 7 6 5 4 3 2 1: : : : : : : : :Kind Unkind

1 2 3 4 5 6 7 8: : : : : : : : :Backbiting Loyal

1 2 3 4 5 6 7 8: : : : : : : : :Untrustworthy Trustworthy

1 2 3 4 5 6 7 8: : : : : : : : :Nasty Nice

1 2 3 4 5 6 7 8: : : : : : : : :Insincere Sincere

Note: 1 ¼ least descriptive of the Least Preferred Coworker; 8 ¼ most descriptive of the Least Preferred Coworker.

Source: F. E. Fiedler. (1967). A theory of leadership effectiveness. New York: McGraw-Hill. Used with permission of the author.

Scores on the LPC Scale can range from 18 to 144. A score of 56 or less indicates that a person is a task-oriented leader; a score of 63

or above indicates that a person is relationship-oriented. Scores between 56 and 63 indicate that a person’s leadership style cannot be

determined.


and the evidence is mixed. For example, ithas been found that leader LPC scores pre-dict performance in situations of differingfavorability in a way that is consistent withthe theory (Chemers, 1983; Chemers, Hays,Rhodewalt, & Wysocki, 1985), but othertests have not been supportive (e.g, Schrie-sheim & Kerr, 1977; Vecchio, 1977). Themost comprehensive test of contingencytheory to date was a meta-analysis conduc-ted by Schriesheim, Tepper, and Tetrault(1994). This study found that the differencesin mean performance levels of High versusLow LPC leaders in different octants gener-ally supported Fiedler’s theory. However, interms of absolute levels of performance, theresults were less supportive. For example, inhighly favorable situations, it was found, aspredicted by Fiedler’s theory, that Low LPCleaders out-performed High LPC leaders.However, the performance of High LPC lead-ers was still above the mean, which is con-sistent with the idea of ‘‘mismatch’’ proposedby Fiedler. Schriesheim et al. (1994) recom-mended that ‘‘organizations without theability or interest in situational engineeringmight consider just trying to make all lead-

ership situations highly favorable (Octant 1)’’(p. 571).

Other than the equivocal support, the por-tion of Fiedler’s theory that has been thesource of greatest criticism is the LPC Scale.Many researchers have questioned the logicbehind the measurement strategy (e.g.,McMahon, 1972; Theodory, 1982). In fact,having given the LPC Scale to students forseveral years, we have noted they are oftenconfused by the instructions. A more seriousproblemis the lackof support for theconstructvalidityof this scale.Recall fromChapter2 thatconstruct validity reflectswhether ameasure ismeasuring the intended construct or attribute.Strong support for the construct validity of theLPC Scale simply does not exist.

At this point in time, Fiedler’s theoryno longer represents one of the major theo-retical approaches used by leadership re-searchers. Even so, it is a valuable theorybecause it has generated a great deal of re-search on leadership. It has also served as thebasis for Cognitive Resource Theory (Fiedler &Garcia, 1987), which states that groups drawon the different cognitive resources from theleader, depending on the situation. This is arelatively new approach, and not a great dealof work has been done on it as yet. It doesseem to be a promising approach, though,and ultimately may be more useful thanFiedler’s original theory.

In addition, Fiedler drew researchers’attention to the importance of consideringthe role of the situation in understandingthe leadership behaviors that are most sup-portive of effective performance (see Vroom& Jago, 2007). A recent study by Yun,Faraj, and Sims (2005) illustrates the impor-tance of situational factors as determinants ofleader effectiveness. These authors examinedthe effects of leadership in the stressful situa-tion of trauma-resuscitation teams. Theauthors examined the effects of empowering

FIGURE 10.4Effectiveness of High versus Low LPC Leaders atDifferent Levels of Situation Favorability

High

Low

Low HighSituation Favorability

Work GroupPerformance

Low LPC

High LPC


(delegating responsibility to team members,encouraging initiative) versus directive(expecting team members to follow detailedinstructions and guidelines) leadership undersituational conditions of high versus lowtrauma severity and low versus high teamexperience.

The authors used a scenario approachin which trauma-resuscitation personnelread scenarios describing realistic eventsvarying along leadership style, traumaseverity, and team experience. The authorsexamined how effective participants thoughtthe team would be in the given scenario andhow much of a learning opportunity theevent would provide. The authors found thatempowering leadership was perceived to bemost effective when the severity of traumawas low and when the amount of teamexperience was high. Directive leadershipwas perceived to be more effective whentrauma severity was high or team experiencewas low. This study showed that the mosteffective leadership style depended on real-istic features of the situation, which is theprimary tenet of the contingency approach toleadership.

Path--Goal Theory

Path–Goal Theory represents a very ambitiousattempt to blend leadership and employeemotivation into one theory (House, 1971;House & Mitchell, 1974). The basic ideabehind Path–Goal Theory is that the role ofa leader is really to help his or her subordi-nates become successful. House actuallystated this in Expectancy Theory terms(Vroom, 1964); specifically, if a leader issuccessful, subordinates’ level of expectancy(the perception that effort will lead to per-formance) is raised. Stated differently, thefunction of leaders is to show subordinatesthe ‘‘path to the goal.’’

Path–Goal Theory states that a leadermust be able to adapt his or her leadershipstyle to the subordinates being supervisedand the situation. House proposed that, tobe successful, a leader must be capable ofutilizing the four different leadership styles:directive leadership, supportive leadership,achievement-oriented leadership, and par-ticipative leadership.

Directive leadership focuses on makingsure that subordinates know what they aresupposed to be doing, and perhaps clarify-ing task responsibilities. A leader who meetswith subordinates once a week to give outwork assignments is exhibiting directive lead-ership. Supportive leadership represents behav-iors that are aimed at showing concern andcaring for subordinates. A leader who makesit a point to ask about a subordinate’s sickchild is exhibiting supportive leadership.

Achievement-oriented leadership representsbehaviors that are aimed at helping employeesto improve their performance and ultimatelyperform better. A leader may exhibit thisleadership style in a number of ways, suchas providing on-the-job coaching, settingchallenging goals, making sure training anddevelopment opportunities are available, andseeing to it that subordinates have the re-sources they need in order to be successful.Finally, participative leadership representsbehaviors that are aimed at getting the inputof subordinates on work-related matters. Aleader who regularly seeks the input of sub-ordinates before making important decisionsis exhibiting this form of leadership.

Having described the four leadershipstyles, the next issue is to determine wheneach of these leadership styles should beused. Path–Goal Theory proposes that lead-ers should consider two situational factorswhen they are deciding on the appropriateleadership style (1) characteristics of one’ssubordinates and (2) characteristics of the


work environment. With respect to subordi-nates, the two key factors that a leader mustconsider are perceived ability and personality.In considering perceived ability, what wouldbe the most appropriate leadership style forsubordinates who perceive themselves as hav-ing limited job-related abilities? For these sub-ordinates, a leader would probably need to bequite directive, because these individualslikely would want to know exactly what todo. Participative leadership may not beemphasized because individuals who perceivetheir abilities to be limited may not have agreat deal to contribute. Achievement-oriented and supportive leadership wouldprobably be used to varying degrees, depend-ing on other characteristics of the subordi-nates.

When subordinates perceive themselvesas having a great deal of task-related ability, aleader would probably need to put relativelylittle emphasis on directing. Instead, theleader may need to strongly emphasizeachievement-oriented and participative lead-ership. Those who perceive their abilities tobe high may have a strong desire to furtherdevelop those abilities; thus, achievement-oriented behaviors would be called for.These subordinates may also have a greatdeal to contribute, so it would be in theleader’s best interests to solicit input andideas from these individuals. Supportiveleadership would likely be used in varyingdegrees, depending on other characteristicsof subordinates.

The second subordinate characteristicthat leaders need to consider when decidingon a leadership style is personality. This isobviously a broad category, but one person-ality trait that Path–Goal Theory deemsimportant is subordinates’ locus of control.According to Rotter (1966), locus of controlreflects relatively stable individual differ-ences in beliefs regarding control of external

reinforcements. A person with an internallocus of control believes that he or she hasa great deal of control over reinforcements.Such a person, for example, would believethat working hard would be a good thing todo because it would lead to positive out-comes. Persons with an external locus ofcontrol believe that reinforcements in theirlives are due to external forces such as luck,fate, or, perhaps, powerful people.

As a leader, managing an individual withan internal locus of control would probablyrequire an emphasis on achievement-orientedand participative leadership, and compara-tively less on directive and supportive lead-ership. An employee with an internal locusof control believes that he or she has controlover reinforcements, and hence is also likelyto believe that if performance is increased,then positive rewards will result. Facilitatingthis process requires the use of achievement-oriented leadership. Also, because thosewith an internal locus of control (internals)may also perform well (Spector, 1982), it isoften in the best interest of the leader to seekinput from such individuals through partic-ipative leadership.

Those with an external locus of controlwill likely need greater direction from theleader; thus, directive leadership behaviorswill be needed. Also, it is very likely that thosewith an external locus of control (externals)will need more support from the leader, com-pared to internals. Having an external locusof control has been shown to be associatedwith negative mental health outcomes (e.g.,Spector, 1982; Storms & Spector, 1987);thus, externals may often be more anxious,frustrated, and dissatisfied than internals.

In addition to the characteristics of sub-ordinates, Path–Goal Theory proposes thatleaders must focus on characteristics of thework environment when they are determin-ing the most appropriate leadership style.


One aspect of the situation that is importantis the prevailing norms regarding authorityand leadership within an organization. Thisis really an aspect of an organization’s cultureand reflects, for example, prevailing views onissues such as employee involvement andparticipation, the extent to which employeesshould take the initiative to solve work-related problems, and whether managersshould get involved in subordinates’ per-sonal lives. In an organization that stronglyvalues employee involvement and participa-tion, a participative leadership style would fitmuch better than in a very autocratic orga-nization. Similarly, in an organization thatplaces a great deal of emphasis on employeeself-reliance, a very directive style of leader-ship would probably not fit very well. On theother hand, achievement oriented and par-ticipative styles would be very compatible.

Task structure is a second characteristicof the work environment that is important indetermining the most appropriate leadershipstyle. If a leader is directing a group that isworking on a highly structured task (e.g.,producing a very simple product), therewould probably be little need for the leaderto adopt a directive or a participative leader-ship style because members of the groupknow exactly what they’re supposed to do.In contrast, when a task is highly unstruc-tured (e.g., developing a new product), aleader may at times have to be directive,but may also need to be participative in orderto help the group figure out how best toapproach the task.

The final environmental characteristicproposed by Path–Goal Theory is thenature of the work group one is leading.For example, in some groups, the task ofproviding direction is done by experiencedmembers of the group rather than the lead-er. If this is the case, the leader does notneed to be directive but could emphasize

other leadership styles. Essentially, thismeans that the leader’s behavior needs to‘‘add value’’ to the behaviors being per-formed by members of the group.

House reformulated his theory in 1996 toprovide a comprehensive theory that de-scribes 10 different categories of leadershipbehaviors and identifies which behaviorsare appropriate in specific circumstances(House, 1996). The 10 categories cover alarge range of behaviors including behaviorsdesigned to clarify the work roles of subor-dinates to behaviors designed to help sub-ordinates achieve excellence in performance.House derives 22 propositions from histheory that provide guidelines for when aspecific type of leader behavior is likely tohave positive, negative, or no consequences.The essence of House’s theory remains thesame: Leaders need to possess a broad rep-ertoire of behaviors that they can strategicallycall on depending on key aspects of thesituation and characteristics of subordinates.

Given the nature of Path–Goal Theory, itis difficult to test in its entirety. However,tests of various parts of the theory havebeen relatively successful (e.g., Wofford &Liska, 1993).Britt,Davison,Bliese,andCastro(2004) also reviewed a number of studiesrelevant to the effects of military leadershipthat supported aspects of House’s overalltheory. The practical implications of Path–Goal Theory come primarily in the area ofmanagement training and development. Spe-cifically, managers need to be trained to rec-ognize meaningful differences among theirsubordinates, as well as important aspectsof the work environment, and they have tolearn to use the different leadershipstyles proposed by Path–Goal Theory. Thetheory may have implications for selectionand placement. For example, if a leader isvery good at developing subordinates (i.e.,providing achievement-oriented leadership),


an organization may wish to place this personin charge of a group consisting of a number ofyoung, high-potential employees. Conversely,if a leader is very adept at participative lead-ership, an organization may want to place thisperson in charge of a group that must makemany consensus decisions.

Vroom--Yetton--Jago Model

The Vroom–Yetton–Jago model (Vroom &Jago, 1988, 2007; Vroom & Yetton, 1973)is a contingency theory of leadership thatfocuses on one aspect of leadership: decisionmaking. This model is also more prescriptivethan the other theories discussed; that is, thistheory is focused on providing leaders with aset of guidelines for which decision-makingstyle to adopt. According to this model, lead-ers will be more effective to the extent thattheir decision-making style is compatiblewith the situations they face.

The first component of the Vroom–Yetton–Jago model to consider is the variousstyles that a leader could use in making adecision. As can be seen in Table 10.2, in thefirst decision-making style (AI), the leadermakes a decision alone after consideringrelevant information. The next decision-making style (AII) also involves the leader

making the decision alone, but, in this case,information is obtained from subordinatesbefore making the decision. Decision-makingstyle CI involves sharing the problem witheach subordinate individually, and then mak-ing the decision alone. Decision-making styleCII involves sharing the problem with sub-ordinates as a group and then making thedecision alone. The final decision-makingstyle (GII) involves making the decision bygroup consensus.

According to the model, in order todetermine which decision-making style ismost appropriate leaders must analyze asituation for the presence or absence ofthe following attributes: (1) the need for aquality decision; (2) whether the leader hassufficient information to make the decisionalone; (3) the degree to which the problemis structured; (4) whether subordinates’ ac-ceptance is needed for implementation; (5)whether subordinates will accept the lead-er’s decision; (6) the degree to which sub-ordinates share the organization’s goals; (7)whether there will likely be conflict amongsubordinates as to the most preferred deci-sion; and (8) whether subordinates haveenough relevant information to make a de-cision on their own.

According to the model, these eight sit-uational attributes will determine a ‘‘fea-sibility set’’ of decision-making strategies.The feasibility set simply represents thosedecision-making strategies that may be ap-propriate for a given situation. Figure 10.5shows how this process works. Notice thatthese situational questions are asked in asequential fashion that resembles a flow-chart. Specifically, the leader’s response toeach question narrows the feasibility set untileventually one decision-making style is rec-ommended. For a leader to use this theory,he or she would simply answer each of thequestions about the decision to be made,

TABLE 10.2Decision-Making Styles Proposed by the Vroom–YettonJago Model of Leadership

AI—Leader makes the decision alone after consider-

ing the relevant information.

AII—Leader makes the decision alone after obtaining

relevant information directly from subordinates.

CI—Leader shares the problem with each subordi-

nate individually and then makes the decision alone.

CII—Leader shares the problem with subordinates as

a group and then makes the decision alone.

GII—The decision is made by group consensus.


and, ultimately, a preferred method of deci-sion making would emerge.

Research on the Vroom–Yetton–Jagomodel has shown that managers are more

effective when they adopt decision-makingstyles that are consistent with the model’sprescriptions (Margerison & Glube, 1979;Paul & Ebadi, 1989; Vroom & Jago, 1988,

FIGURE 10.5The Recommended Decision-Making Sequence Proposed by the Vroom-Yetton-Jago Model

A B C D E F G

Statethe

Problem

No

No

No

No

No

No

No

No

No

No

No

No

No

No

No

No

No

YesYes

Yes Yes Yes

Yes

Yes

Yes

Yes

YesYes

Yes

Yes

Yes

YesYes

Yes

1: AI, AII, CI, CII, GII

2: GII

3: AI, AII, CI, CII, GII

4: AI, AII, CI, CII

5: GII

6A: CII6B: CI, CII

7: AII, CI, CII

8: AI, CI, CII, GII

9: CII

10: CII, GII

11: GII

12: CII

A. Does the problem possess a quality requirement?

B. Do you have sufficient information to make a high-quality decision?

C. Is the problem structured?

D. Is acceptance of decision by subordinates important for effective

implementation?

E. If you were to make the decision by yourself, is it reasonably certain

that it would be accepted by your subordinates?

F. Do subordinates share the organizational goals to be attained in

solving this problem?

G. Is conflict among subordinates over preferred solutions likely?

Source: V. H. Vroom and P. W. Yetton. (1973). Leadership and decision-making Pittsburgh, PA: University of Pittsburgh Press.

Copyright # 1973 by University of Pittsburgh Press. Reprinted with permission.


2007). However, a major methodologicallimitation of most tests of the model is thatthey have relied primarily on retrospectivedescriptions of decisions made by managers.This raises the question of whether managersrevise their recollections of decisions in away that is consistent with the model. Morerecent research that has not relied on retro-spective reports (Field & House, 1990;Parker, 1999) has provided more limitedsupport for the theory.

From a practical point of view, theVroom–Yetton–Jago model is one of themore useful leadership theories that has beendeveloped. Compared to other theories, thismodel provides leaders with some specificguidelines for making decisions, rather thanmerely describing leadership processes. Thebiggest problem with the Vroom–Yetton–Jago model is that it tends to oversimplifythe conditions under which leaders makedecisions. For example, in many cases, it isdifficult for a leader to provide ‘‘Yes–No’’answers to the questions posed earlier. Fur-ther revisions of this model will be needed toovercome these weaknesses.

Leader---Member Exchange(LMX) Model

Anyone who has been part of a work group, orwho has been a leader of one, knows thateveryone is not always treated the same. Tothe contrary, leaders typically develop aunique relationship with each subordinate,and some of these relationships are more pos-itive than others. Based on this idea, Danser-eau, Graen, and Haga (1975) developed theVertical Dyad Linkage Model of leadership. Theterm Vertical Dyad was originally used to de-scribe this theory because of its emphasis onthe unique relationship between leaders andsubordinates. Over time, however, the nameof the theory eventually became Leader–Member Exchange because this relationship is

really one that reflects social exchangebetween the leader and the subordinate.

According to Dansereau et al. (1975),within work groups there are typically twosets of employees: the in-group and the out-group. The in-group consists of employeeswho are trusted confidants of the leader.These are typically individuals who performwell, have a desire to assume greater levels ofresponsibility, and simply get along well withthe leader. Members of the out-group consistof the group of subordinates who have moreformal relationships with the leader. Membersof the in-group are typically privy to moreinformation from the leader than are membersof the out-group, and they are also given morediscretion over how to do their jobs. Membersof the out-group are typically individuals whomay not perform as well, may not desire agreat deal of responsibility, or simply may notget along as well with the leader as do mem-bers of the in-group.

Gradually, less emphasis has beenplaced on the in-group/out-group distinc-tion, and more emphasis is on how leader–subordinate relationships develop over time(Graen, 1976). According to Graen (1976),when a subordinate is first assigned to aleader, the leader has relatively limitedinformation as to this person’s capabilities.Thus, over time, the leader tests the subor-dinate by giving him or her assignments ofincreasing responsibility. To the extent thatthe subordinate is successful, a positiveexchange relationship develops. From thesubordinate’s point of view, there maybe some degree of negotiation as to specificrole responsibilities. Other factors that influ-ence the development of this exchange rela-tionship are perceived similarity betweensubordinates and leaders, as well as the levelof interpersonal attraction (Liden, Wayne, &Stilwell, 1993). Exchange relationships arelikely to be most positive when subordinates


are competent, when they and the leaderperceive some degree of mutual similarity,and when subordinates and leaders like eachother.

What are the consequences of the ex-change relationship that develops betweena subordinate and a leader? Gerstner andDay (1997) conducted a meta-analysis of 79studies that examined correlates of Leader–Member Exchange. They found that LMXwas positively related to job performance,job satisfaction, and organizational commit-ment, and negatively related to outcomessuch as turnover and role stressors. One ofthe most perplexing findings in their meta-analysis was the relatively small correlationbetween leaders’ and subordinates’ reportson the quality of the exchange relationship(corrected r ¼ .37). Thus, although leadersand subordinates tend to agree on the qualityof the relationship that exists between them,this level of agreement is not great. At pre-sent, it is unclear why agreement on thequality of the exchange relationship is nothigher, what factors influence agreement, orthe impact of disagreements over the qualityof the exchange relationship.

LMX Theory is useful for both theoret-ical and practical reasons. In terms oftheory, it presents leadership in a morerealistic light, compared to many previoustheories. Subordinates are not simply pas-sive recipients of leaders’ influence. Interms of practical implications, LMXTheory suggests that it is desirable for lead-ers to develop positive exchange relation-ships with their subordinates. This may notbe possible 100% of the time, but organi-zations may be able to facilitate thedevelopment of high-quality exchange rela-tionships by training managers in suchskills as communicating with subordinates,providing feedback, and engaging in coach-ing activities.

LMX Theory faces a number of chal-lenges. One of the most important of theseis continued refinement of what actuallyconstitutes the exchange relationship itself.To measure the exchange relationship,Liden and Maslyn (1998) developed a scalethat consisted of four distinct dimensions:(1) affect, which represents the levels ofmutual interpersonal attraction between aleader and subordinate; (2) loyalty, whichrepresents the amount of public supportprovided by each member of the leader–subordinate dyad; (3) contribution, whichrepresents what each member of the leader–subordinate dyad contributes positively tothe goals of the organization; and (4) pro-fessional respect, which represents thedegree to which each member of the leader–subordinate dyad has built a reputation,within and/or outside of work, because heor she excels in his or her line of work.Previous LMX scales have treated it as aone-dimensional construct.

Another challenge for LMX Theory isexpansion of its scope. For most people, theunique relationship they develop with theirimmediatesupervisorisoneofthemostimpor-tant dimensions of their work experience. Assuch, it may influence many work outcomes.For example, Kokotovich, Jex, and Adams(2000) found that a high-quality LMX moder-ated the relationship between role ambiguityand job satisfaction. Employees reporting ahigh-quality LMX actually reacted positivelyto role ambiguity. One study also found thatLMX was related to the organizational citizen-ship behavior of altruism (Wayne Andy &Green Hown, 1993). Researchers should fur-ther investigate these types of relationships.Recent studies have also suggested that LMXmay interact with the cognitive ability of em-ployees to also predict creativity (Tierney,Farmer, & Graen, 1999). More recent studieshavealsoarguedthatother forms of leadership


(e.g., transactional, transformational) influ-ence employee performance through LMX.We now turn to a discussion of these areas ofleadership.

Charismatic, Transformational,and Transactional Leadership

These last three leadership theories are thenewest to be developed. Although less re-search has been conducted on these ap-proaches overall, the theories representwhere the field of leadership is heading andhave become quite influential. Because theseapproaches to leadership are highly related,they will be discussed together.

The idea of Charismatic and Transforma-tional leadership is that there are certain leaderbehaviors and traits that not only influencesubordinates but may also inspire them toperform well beyond their capabilities. An-other defining characteristic of Charismaticand Transformational leadership is that bothhave the potential to induce meaningfulchange in organizations. The terms charis-matic and transformational leadership areoften used interchangeably, and when a dis-tinction is made between the two forms ofleadership, it is noted that charismatic lead-ership is but one component of transforma-tional leadership.

The term that is sometimes used to de-scribe the opposite of Charismatic and Trans-formational leadership is Transactionalleadership. A transactional leader is one whomakes sure that subordinates get the jobdone and follow the rules of the organization.Transactional leaders typically use behavioralprinciples of reward and punishment tomake clear the behaviors that are expectedof the employee (Judge & Piccolo, 2004).Transactional leaders, however, do notinspire subordinates or facilitate meaningfulchange in organizations.

It is important to note, however, thatleaders can be capable of engaging in bothtransactional and transformational leader-ship depending on the situation. In fact, Bass(1998) has argued forcefully that transac-tional leadership often forms the base fortransformational leadership. The logic hereis that a leader must be able to clearly applyrules and contingencies for employees tofollow before embarking on the more moti-vating behaviors involved in transforma-tional leadership.

Judge and Piccolo (2004) have argued thattransformational leadership is composed offour primary dimensions. The first dimensionis idealized influence (charisma). This compo-nent refers to leaders setting the example ofexemplary performance and dedication to theorganization through conviction and emo-tional investment. Those who are charismatictend to have a number of common traits: acaptivating tone of voice, direct eye contactwith the listener, animated facial expressions,and a powerful, confident, and dynamic com-munication style. This type of communicationstyle obviously helps a leader to communicatehis or her vision and to generate enthusiasmfor it. It also helps more generally by increas-ing the leader’s appeal to his or her followers.Charismatic leaders have great ‘‘presence’’and make a tremendous impression on thosearound them.

The second dimension is inspirationalmotivation. One task that is often cited in thisregard is providing a vision. According toHouse (1977), a vision is a very generalizedideal state that typically represents sharedvalues and often has moral overtones. Anexample of a vision for a universitymight be to enlighten the students; a visionfor a military organization might be touphold freedom around the world; a visionfor an auto manufacturer might be toenhance the mobility of society. A vision


applies to all members of the organizationand can thus serve as a general ‘‘rallyingpoint’’ for everyone. Many examples of lead-ers, particularly in the political arena, can bedistinguished on the presence or absence ofvision (see Comment 10.2).

The third dimension is intellectual stimu-lation. This dimension refers to the trans-formational leader’s ability to challengesubordinates and encourage them to be cre-ative and take appropriate risks. Leaders mayencourage subordinates to think outside thebox in order to come up with innovative solu-tions that will instill a competitive advantageto the organization.

The final dimension is individualizedconsideration. This dimension refers to theleader’s ability to attend to the needs ofemployees and make the employees feelunderstood and appreciated. In many ways

this dimension incorporates aspects of LMXtheory into what it means to be a transforma-tional leader. Transformational leaders tendto have a charismatic communication style.

Research over the years has shown thattransformational leadership is related topositive outcomes such as employees’ per-formance, satisfaction, and positive percep-tions of leaders (Bass & Avolio, 1993;Shamir, House, & Arthur, 1993). Judgeand Piccolo (2004) recently conducted ameta-analysis of over 87 studies examiningthe correlations between transformationaland transactional leadership and variousperformance outcomes (e.g., follower jobsatisfaction, follower satisfaction with theleader, leader job performance). Theauthors found an overall validity coefficientof .44 for transformational leadership and.39 for transactional (contingent reward)

COMMENT 10.2

THE VISION THING

ONE OF THE key components of Charismatic

and Transformational leadership is vision. A

vision is essentially an ideal or desirable end

state that often has moral overtones. A leaderwith vision ‘‘stands for something’’ and has a

sense of purpose that is communicated to his

or her followers.

Vision has become particularly important

in the political arena. When candidates run for

national office, the vision that they are able to

communicate to voters can literally make or

break their chances of being elected. In 1980,Ronald Reagan defeated Jimmy Carter for the

U.S. Presidency largely based on the vision that

he communicated to the American public.

Reagan’s vision, based heavily on conservative

principles, struck a chord with voters who

wanted lower taxes and a stronger national

defense. Whether or not one agreed with Rea-

gan’s ‘‘vision,’’ there is no denying that he

communicated it well and was quite successful

at convincing the public to embrace it.

Just as having a vision propelled RonaldReagan to victory, a lack of vision may have

been one of the major reasons George H. Bush

lost the presidency to Bill Clinton in 1992.

Although Bush showed excellent crisis-

management skills during the Gulf War, he

was unable to articulate a coherent vision in

the way Reagan did many years earlier. For

many voters, it was difficult to tell exactlywhat Bush stood for. Clinton, in contrast,

was very successful at communicating a vision

based on economic opportunity, and in many

instances seemed to connect with voters

much better on a personal level. The end result

was that Clinton won a convincing victory over

Bush and third-party candidate Ross Perot.


leadership. One interesting finding was therather strong positive correlation (.80)between transformational and transactionalleadership in the meta-analysis. This find-ing strongly suggests that transformationaland transactional leadership are not oppos-ing ends of a single dimension ofleadership. Bass and his colleagues havealso found that both transformational andtransactional leadership predict the perfor-mance of light infantry platoons in combat-simulation exercises (Bass, Avolio, Jung, &Berson, 2003).

One emerging trend in research on trans-formational leadership is examining how suchleadership predicts performance outcomes.Two studies investigating similar determi-nants were recently conducted by Piccoloand Colquitt (2006) and Purvanova, Bono,and Dzieweczynski (2006). Both papersaddressed the idea that transformational lead-ers cause their employees to be engaged inmore meaningful work (e.g., report higher jobcharacteristics such as variety, significance,and autonomy), which then leads their em-ployees to perform better. In support of thishypothesis, Piccolo and Colquitt (2006)found that such job characteristics mediatedthe relationship between transformationalleadership and both task performance andorganizational citizenship behavior. Purva-nova et al. (2006) also found that perceivedjob characteristics mediated the relationshipbetween transformational leadership and adifferent measure of citizenship performance,even when controlling for objective job char-acteristics. Taken together, the results indicatethat transformational leaders produce em-ployees who perform better because the em-ployees take greater ownership of their workand feel their work is more significant.

Other researchers have argued that trans-formational leadership is related to followerperformance through leader-member ex-

change (LMX; Wang et al., 2005). The logichere is that transformational leaders end upforming a stronger interpersonal bond withtheir followers, which leads their followers toperform better. Wang et al. examined leader-follower dyads in organizations across thePeople’s Republic of China. They found thatthe relationship between transformationalleadership and the employee’s task perfor-mance was completely mediated by a meas-ure of LMX focusing on the quality of therelationship between the employee andsupervisor.

Authentic Leadership

One of the most recent approaches to lead-ership coming from the field of organiza-tional psychology is authentic leadership.Avolio, Gardner, Walumbwa, Luthans, andMay (2004) define authentic leaders as ‘‘thoseindividuals who are deeply aware of howthey think and behave and are perceived byothers as being aware of their own andothers’ values/moral perspective, knowledge,and strengths; aware of the context in whichthey operate; and who are confident, hope-ful, optimistic, resilient, and high on moralcharacter’’ (pp. 802–804). A key aspect ofauthentic leadership is leaders being ‘‘whothey are’’ and harnessing the energy of fol-lowers by causing them to connect with thegoals of the leader and group.

Avolio et al. (2004) developed a model ofauthentic leadership illustrating how authen-tic behavior on the part of the leader resultsin followers being more likely to personallyidentify with the leader and collective (i.e.,organization), which then leads the followersto experience hope, trust, and positive emo-tions. These positive emotional states thencreate the favorable work attitudes of com-mitment, job satisfaction, meaningfulness,and engagement, which result in the positive


outcomes of performance, extra effort, andless withdrawal. The theory of authenticleadership is in its early stages of develop-ment, but will likely get much more attentiongiven recent highly publicized incidences ofleader corruption and lack of character.

Before ending our discussion of leader-ship, it is worth noting that, until recently,cross-cultural research on leadership waslacking. However, Comment 10.3 discussesan ambitious project examining conceptions

of leadership across 62 different countries.Future research will be addressing the impli-cations of this project for multiple conceptu-alizations of leadership.

POWER AND INFLUENCEIN ORGANIZATIONS

Regardless of whether one is a chief execu-tive officer of a Fortune 500 company or thesupervisor of a janitorial crew, a big part of

COMMENT 10.3

LEADERSHIP AND CULTURE

CULTURAL DIFFERENCES IN leadership have not

been heavily researched by organizational

psychologists. One of the few extensive trea-

tises of cultures and leadership was recentlycompleted by House and his colleagues

(2003), who examined 17,000 leaders from

951 organizations in 62 societies across the

globe. The authors were interested in explor-

ing differences between cultures in how they

viewed leadership and the practices that lead-

ers used in the different cultures. The authors

also explored value differences across the 62societies. Six global leader behaviors were

identified: charismatic/value-based leader-

ship (leader ability to inspire and motivate

others), team-oriented leadership (emphasis

on team-building and being diplomatic), par-

ticipative leadership (involving others in lead-

ership decisions), human-oriented leadership

(compassion and consideration toward sub-ordinates), autonomous leadership (individu-

alistic decision making and independence/

separation from subordinates), and self-

protective leadership (focus behaviors on

protecting the individual leader and group

through behaviors designed to enhance status

and save face).

House and his colleagues found that allcultures believed in the importance of leaders

possessing a team orientation and effectively

communicating a vision to subordinates. The

authors found the most variation in the global

leader behaviors of autonomous leadershipand self-protective leadership. Autonomous

leadership was seen as slightly effective in

countries of Eastern Europe (with the excep-

tion of Hungary) but was seen as ineffective in

most countries of Latin America, the Middle

East, and Anglo countries. Self-protective lead-

ership was seen as being slightly effective for

the following countries: Albania, Taiwan,Egypt, Iran, and Kuwait. However, this type

of leadership was seen as ineffective in other

countries, especially in Northern European

countries such as France. The book describes

the similarities and differences across cultures

in perceptions of effective and ineffective lead-

er behaviors, and attempts to link these per-

ceptions to differences in the values of thecultures involved. The research discussed in

this book will likely influence future investi-

gations of how leadership differs across cul-

tures.

Source: House, R.J., Hanges, P.J., Javidan, M., Dorfman,

P.W., & Gupta, V. (Eds.) (2003). Culture, leadership, and

organizations: The GLOBE study of 62 societies. Thou-

sand Oaks, CA: Sage Publications.


one’s job is influencing others to behave inways that are consistent with the goals of theorganization. Furthermore, the extent towhich a leader can influence others depends,to a large extent, on his or her social powerover others. In addition, employees otherthan leaders can use power and influencein either adaptive or dysfunctional wayswithin organizations. In this section, powerwill be discussed first, followed by influencetactics.

Defining Power

The term power is often used in a negativefashion, even though it is not inherently bador evil. Power simply represents a person’spotential or capacity to influence others(French & Raven, 1959). When one attemptsto influence another person’s behavior, theoutcome of that influence attempt generallytakes one of three forms (Kelman, 1958):compliance, identification, or private accep-tance. Compliance refers to an influenceattempt where the target of influence doeswhat is requested, but does not necessarilydo it willingly. When a child is told by aparent that he or she cannot have a cookie,the child typically complies with this direc-tive but, if given the choice, would certainlyeat the cookie (at least that’s the way it worksin our houses!). An example of compliance inthe workplace might be an employee wearinga piece of safety equipment, even though heor she doesn’t want to and does not believe itwill necessarily be effective.

The second potential outcome of influ-ence is referred to as identification. In thiscase, the employee does what the leaderwants, primarily because he or she likes theleader. As with compliance, when behavioris changed on the basis of identification,there is a change in behavior but not inattitudes; that is, the employee still doesnot really want to do what the leader wants

done. A work-related example of identifica-tion would be employees staying late to helptheir well-liked leader meet an impendingdeadline, even though they do not inherentlybelieve in the value of the project.

The third result of influence is referred toas private acceptance or internalization. In thiscase, the employee does what the leaderwants because he or she believes that it isthe right thing to do. Compared to compli-ance and identification, private acceptanceis, in the long run, much more efficient forleaders. Therefore, if subordinates believethat what the leader wants them to do iscorrect, the leader will need to spend muchless time either monitoring to ensure com-pliance, or making sure that subordinatesstill like him or her. Keep in mind, however,that it is not always necessary for a leader toobtain private acceptance from subordinates.For example, employees often must complywith safety guidelines, even if they don’tagree with them.

The fourth and final outcome of influ-ence that might occur is resistance. In thiscase, the employee simply does not do whatthe leader asks. Resistance may take the formof an overt refusal, but, more typically, anemployee will simply be evasive when theleader inquires about whether the subordi-nate has carried out the request. This can bea very frustrating situation for a leader, and itis obviously the least desirable outcome froma leader’s perspective.

Bases of Power

Leaders are not automatically endowed withan unlimited amount of power over subor-dinates. Leaders also differ in terms of thesources or bases upon which power oversubordinates can be exerted. The mostwidely cited model of power was proposedby French and Raven (1959) over 40 years


ago. According to this model, power restsupon six bases. Some readers may recognizethe fact that most treatments of French andRaven’s model describe only the first fivebases, but the original model did containsix. The first base of power is labeled coercivepower. The basis of this influence tactic is thatone person can punish another. Thus, asubordinate may do what a leader requestsbecause the leader has the power to fire thesubordinate. Although the threat of punish-ment may give a leader considerable powerover subordinates, coercive power generallyis not a very efficient base of power. If sub-ordinates do what the leader wants onlybecause they are threatened with punish-ment, the leader’s power is diminished con-siderably if he or she is not around tomonitor the ongoing behavior and adminis-ter punishment if necessary.

The second power base described byFrench and Raven is labeled reward power.This is essentially the opposite of coercivepower. That is, subordinates do what theleader wants because the leader has the abil-ity to reward them in some way. For exam-ple, a subordinate may comply with aleader’s request that he or she work overtimebecause the leader has the power to grantthis employee a larger pay increase whenraises are given out. Unfortunately, as withcoercive power, reward power is not a highlyefficient power base. It requires the leader tomonitor subordinates’ behaviors and rewardthem at the appropriate time. An individualwho possesses either coercive or rewardpower is likely to also possess the corre-sponding type of power. Leaders differ, how-ever, in the extent to which they ultimatelychoose to use reward versus coercive power(see Chapter 9).

The third power base is labeled legiti-mate power. This power emanates from theposition that one holds in an organization.

In most organizational settings, the fact thatone employee is another employee’s super-visor means that the supervisor has a legit-imate right to make requests of the otherperson. Note that this legitimate right isindependent of the person holding theposition. Compared to coercive and rewardpower, legitimate power is more efficient. Itdoes not require surveillance on the leader’spart because, in most organizations, thelevel of legitimate authority that goes withany given position is typically known. Infact, in many cases, it is even documentedin job descriptions and other formal docu-ments. A limitation of legitimate power,however, is that if it is used exclusively, itmay elicit only compliance from subordi-nates and, in the long run, may engender agreat deal of resentment among them. Peo-ple generally do not like to be told to dosomething simply because ‘‘I’m your super-visor.’’

The fourth power base is expert power.This is power based on the fact that anindividual is perceived as an expert on some-thing that is important to the target of influ-ence. If the leader of a group of designengineers is also an expert design engineer,this will make subordinates more likely to dowhat he or she says. One thing that is impor-tant to note about expert power is that it isthe perception that is important. For this tobe a viable power base, subordinates mustperceive that the leader is an expert. Regard-less of the level of one’s true expertise, if thisis not perceived, then no expert power exists.It is also possible for nonleaders within agiven workgroup or organization to possessexpert power. For example, if a certain em-ployee is known as the individual possessinga large amount of knowledge regarding aparticular topic or procedure, that employeewill have expert power even if he or she is nottechnically a leader.


The fifth base of power in French andRaven’s model is referent power. This ispower based on subordinates’ liking of aleader. Here, as in the identification modeof influence described earlier, subordinatesdo what the leader wants because they likehim or her. Although this form of powerdoes not require surveillance, it is also some-what more tenuous than expert powerbecause interpersonal attraction is consider-ably more volatile than expertise. If subordi-nates no longer have positive feelings towardthe leader, then a great deal of his or herpower over subordinates is lost.

The sixth and final base of power isreferred to as informational power. As statedearlier, this is typically not presented as oneof the bases of power in the French andRaven model, but it was included in theinitial model (Raven, 1993). A leader hasinformational power to the extent that heor she has high-quality information that willbe convincing to subordinates. For example,a person trying to convince someone else towear a seatbelt would have a great deal ofinformational power if valid data could becited showing that the odds of being fatallyinjured are much lower if a seatbelt is beingworn.

After the development of the initial modelof power bases, French and Raven made anumber of further refinements to the model(Raven, 1993). For example, they differenti-ated between personal and impersonal formsof reward and coercive power. Rewards andpunishments can come in the form of personalapproval or disapproval. Conversely, they canalso come in more impersonal forms such as araise ora formal reprimand. Frenchand Ravenalso refined the concept of legitimate powerconsiderably. They proposed, for example,that legitimate power was based not just onone’s formal organizational position, but alsoon the principle of reciprocity (‘‘I did this for

you, so you should feel obligated to do this forme’’), equity (‘‘I have worked hard and suf-fered, so I have the right to ask you to dosomething to make up for it’’), and responsi-bility or dependence (‘‘I cannot help myself, soyou are responsible for helping me’’).

Expert and referent power were furtherdistinguished in terms of being positive andnegative. As originally conceived, both ex-pert and referent power were positive.French and Raven, however, later pointedout that both could be negative as well.Negative expert power represents situationsin which a person is seen as having superiorknowledge but, at the same time, is seen asusing the superior knowledge only in orderto further his or her own interests. Negativereferent power occurs when a person is seenas someone who is disliked rather than liked.If this person were a leader, subordinatesmay be inclined to do the opposite of whatthis individual wants them to do.

Informational power was distinguishedin terms of being direct or indirect. Wheninformational power is direct, this meansthat the leader presents logical argumentsto subordinates directly. When it is indirect,the information does not come from theleader directly, but may instead come fromanother subordinate or another leader. Thisdistinction is important because social psy-chological research on influence (e.g., Petty& Cacioppo, 1981) has shown that, in somecircumstances, information that is conveyedindirectly is given greater weight by the tar-get of influence than information communi-cated directly.

No competing models of power baseshave been proposed, but there has been atleast one effort to add to the power basesoriginally proposed by French and Raven.Finkelstein (1992) examined bases of powerwithin top management teams and, althoughsome of the power bases he proposed


corresponded to those in French and Raven’smodel, there were two that were unique.Ownership power represents the extent towhich the member of a top managementteam has an ownership stake in the organi-zation, through either stock ownership orfamily relations. Within a top managementteam, an executive who is a significant share-holder or is related to the organizationalfounder often wields tremendous power.

The other unique power base proposedby Finkelstein (1992) was prestige power.This represents the extent to which the mem-ber of a top management group has acquiredprestige and status outside of the organiza-tion. Finkelstein measured this by the num-ber of corporate boards a manager serves on,the level of prestige of those organizations,the number of nonprofit boards one serveson, and, finally, the prestige of the universitywhere the executive received his or her edu-cation. Generally speaking, an executive hasgreater prestige power if he or she serves onthe corporate boards of a number of success-ful organizations, also serves on the boardsof nonprofit organizations, and graduatedfrom a prestigious university (e.g., IvyLeague).

Influence Tactics

To this point, we have discussed thepotential of leaders to influence their sub-ordinates. However, to truly understandthe dynamics of power and influence, wemust go beyond the potential to influenceand examine the specific tactics that lead-ers use to influence subordinates. Accord-ing to Yukl and Tracey (1992),nine distinct tactics can be used to influ-ence. These are presented in Table 10.3.As can be seen rational persuasion simplyinvolves providing employees a logicalexplanation of why a given request is

being made. For example, a foreman in afactory may advise a subordinate to wearprotective earphones because chronicexposure to loud noises can lead to grad-ual hearing loss.

When inspirational appeals are used, theleader or person doing the influencingattempts to appeal to the target’s values orideals, and to persuade that person that he orshe will be able to get something done. As anexample of inspirational appeals, a militarycommander might attempt to encourage hisor her troops to continue fighting after theyare fatigued. The commander could explainthe strategic need to carry on, or couldappeal to the troops’ sense of patriotism ormilitary duty. As indicated earlier in thechapter, this type of appeal is used fre-quently by transformational leaders.

In using consultation, the leader influ-ences subordinates by seeking their assist-ance on an activity for which theirparticipation is crucial. This tactic is oftenused when changes are introduced in orga-nizations. For example, if an organizationwants to redesign jobs and must persuadeemployees to accept these changes, a goodway to start is to seek the employees’ assist-ance in the job redesign effort.

By using ingratiation, a leader attempts toinfluence subordinates by putting them in agood mood before making a request. Thiscan be done in a variety of ways such ascomplimenting the subordinate, agreeingwith his or her views or opinions, or doingfavors for this person. A supervisor who isgetting ready to ask a group of subordinatesto work on a weekend may bring the groupdoughnuts before making the request. Ingra-tiation must be used carefully, however; itmay make people less likely to comply with arequest if it is seen as insincere. Some readersmay be familiar with the situation comedy‘‘The Office,’’ where the leader often engages


in blatantly obvious forms of ingratiationthat have little effect on employee behavior.

When exchange is used as an influencetactic, the leader offers subordinates some-thing in return for complying with a request,or perhaps offers them a share of the benefitsthat will accrue when a task is accomplished(see Cialdini, 2001). In some companies,forms of exchange are actually mandated byorganizational policies. For example, whenhourly employees work more than 40 hoursper week, they receive overtime pay for doingso. However, this exchange may be strictlybetween the leader and his or her subordi-nates. For example, if the manager of a fast-

food restaurant wants employees to come foran early morning crew meeting, one way ofgetting employees to be there is to provideanother incentive, such as an extra 30-minutebreak.

When a personal appeal is used as an in-fluence attempt, the leader appeals to a sub-ordinate’s sense of personal loyalty andfriendship before making a request. This in-fluence tactic can only be used if two peopledo in fact share some degree of loyalty andfriendship. Prior to making a request ofa subordinate, the leader may first state:‘‘We’ve been friends for a long time, and havebeen through some tough times together, so I

TABLE 10.3A Summary of Nine Common Influence Tactics Used by Leaders

Tactic Definition

1. Rational persuasion The person uses logical arguments and factual evidence to persuade you that a

proposal or request is viable and likely to result in the attainment of task objectives.

2. Inspirational appeal The person makes a request or proposal that arouses your enthusiasm by appealing

to your values, ideals, or aspirations or by increasing your confidence that you

can do it.

3. Consultation The person seeks your participation in planning a strategy, activity, or change

for which your support and assistance are desired, or the person is willing to

modify the proposal to deal with your concerns and suggestions.

4. Ingratiation The person seeks to get you in a good mood or to think favorably of him or her before

asking you to do something.

5. Exchange The person offers you an exchange of favors, indicates a willingness to reciprocate

at a later time, or promises you a share of the benefits if you help to accomplish

a task.

6. Personal appeal The person appeals to your feelings of loyalty and friendship toward him or her

before asking you to do something.

7. Coalition The person seeks the aid of others to persuade you to do something or uses the

support of others as a reason for you to agree also.

8. Legitimating The person seeks to establish the legitimacy of a request by claiming the authority or

right to make it or by verifying that it is consistent with organizational policies,

rules, practices, or traditions.

9. Pressure The person uses demands, threats, or persistent reminders to influence you to do

what he or she wants.

Source: G. Yukl and J. B. Tracey. (1992). Consequences of influence tactics used with subordinates, peers, and the boss. Journal of

Applied Psychology, 77, 525–535. Copyright # 1992 by the American Psychological Association. Reprinted with permission.


know you’re someone I can really count on.’’After hearing that, most people would findit difficult to turn down the subsequentrequest.

Forming a coalition to influence involvesseeking the aid of others to directly per-suade a subordinate to comply with arequest, or using others as examples ofwhy a request should be honored. A goodexample: Get a subordinate to comply witha requirement to wear safety equipment byhaving other subordinates, who are wearingthe equipment, persuade this individualthat safety equipment is needed. This prin-ciple of influence is called social proof byCialdini (2001), and capitalizes on the ideathat individuals often determine what iscorrect or right by noting what other peopleare doing.

When legitimating is used, the leader seeksto establish the legitimacy of his or her requestby falling back on his or her authority to makethe request or, in some cases, citing organiza-tional policies or rules. In the military, theleader frequently points out that he or sheoutranks the subordinate; in military organi-zations, this form of influence tends to workvery well because of the emphasis on rank. Inother types of organizations, use of legitimat-ing may be less successful and, if used fre-quently, may ultimately engender animosityamong one’s subordinates.

The final influence tactic listed in Table10.3 is pressure. This involves the use ofdemands, threats, or persistent monitoringto make subordinates comply with a request.Suppose a supervisor wants to make sure asubordinate is on time every morning. Oneway to do this would be to check the person’sdesk to see if he or she is present by therequired time. Although pressure may, attimes, get leaders the behavior they desire,this almost always comes in the form ofcompliance on the part of the employee.

Thus, using pressure typically requires agood deal of energy on the part of the leaderbecause subordinates’ behaviors must be fre-quently monitored.

Although research on influence tactics isstill relatively new, there are some reasonablyconsistent research findings. If a leader wishesto obtain behavior change in the form ofprivate acceptance, the most effective way todo so is through inspirational appeals andconsultation (Falbe & Yukl, 1992; Yukl,Kim, & Falbe, 1996; Yukl & Tracey, 1992).Tactics such as coalition formation, legitimat-ing, and pressure are unlikely to lead to pri-vate acceptance, and, in fact, may even lead toresistance. The reason simply may be thatpeople are generally more enthusiastic aboutdoing things when they feel that they havesome freedom of choice in the matter. Brehm(1966) has noted how individuals experienceresistance when they experience a threat totheir personal freedom, which may lead themto do the opposite of what is requested.

Another consistent finding from this lit-erature is that influence tactics may influenceothers’ behaviors in an additive fashion. Forexample, Falbe and Yukl (1992) found thatthe use of combinations of some tactics wasmore effective at facilitating behavior changethan using the tactics alone. For example, aninspirational appeal combined with consul-tation was more effective than using either ofthese tactics alone or using single ‘‘hard’’tactics such as pressure or legitimating. Thissuggests that, in some cases, the influenceprocess takes time, and the leader must beprepared to use multiple tactics to influencesubordinates’ behaviors.

The research on influence tactics is stillrelatively new, but it has produced some veryimportant practical insights for leaders. Per-haps the most important of these is that ifleaders want their subordinates to do thingswillingly, in the long run they are much


better off asking them do it rather than sim-ply relying on their position or using morecoercive techniques. Although asking maytake longer, it will produce more long-lastingbehavioral change than will the use of morecoercive tactics.

Politics in Organizations

The term organizational politics often con-jures up images of very negative forms ofbehavior; therefore, most people want toavoid the politics of an organization. Never-theless, political behavior is a fact of life and,in many cases, represents an important formof influence within organizations. Organiza-tional politics has been defined as influencebehavior, within organizations, that falls out-side of the recognized legitimate power sys-tem (Yoffie & Bergenstein, 1985). Politicalbehavior is often aimed at benefiting an indi-vidual or group at the expense of the orga-nization as a whole and at acquiring morepower.

According to Miles (1980), one of themajor factors motivating political behavior isuncertainty. For example, when employeesare uncertain about the goals of the organi-zation, political behavior often results. An-other factor that strongly contributes topolitical behavior is scarcity of resources.Although technically everyone in the sameorganization is ‘‘on the same team,’’ obtain-ing scarce resources is a highly competitiveprocess in many organizations. Thus, themanager of a department may have toengage in considerable political behavior inorder to obtain even minimally acceptableresources.

Other conditions that motivate politicalbehavior are technological change, ambigu-ity in decision making, and organizationalchange. Often, the introduction of new tech-nologies in organizations creates consider-able uncertainty with respect to work roles

and lines of authority; both conditions areripe for political maneuvering. In many orga-nizations, decisions are made with incom-plete information; thus, it is not clear whichalternative is ‘‘correct.’’ When this is the case,political behavior often results because advo-cates of different positions may attempt toinfluence the decision-making process.Finally, political behavior is very commonduring times of organizational change becausethings are often ‘‘up for grabs’’ and readilyamenable to such forms of influence.

Having defined organizational politics,we now turn to specific tactics that peopleuse when they engage in political behavior.Although many tactics could be used to pro-mote one’s political agenda, some tactics aremore commonly used, and many of these aresimilar to the general influence tactics dis-cussed in the previous section. According toAllen, Madison, Porter, Renwick, and Mayes(1979), six commonly used political tacticsinclude two that were discussed previously(ingratiation and forming coalitions and net-works), and four that are somewhat differentfrom more general influence tactics.

1. Impression management represents behav-iors that are designed to enhance one’svisibility or stature within the organiza-tion. Bolino and Turnley (1999) devel-oped a questionnaire to assess fivedifferent impression-management strat-egies employees use based on a classifi-cation developed by Jones and Pittman(1982). Self-promotion refers to employ-ees discussing their accomplishmentsand abilities with others to come acrossas competent (e.g., ‘‘Talk proudly aboutyour experience or education’’). Ingratia-tion refers to employees doing favors orcomplimenting others to come across aslikeable (e.g., ‘‘Compliment your col-leagues so they will see you as likeable’’).


Exemplification refers to people high-lighting their moral worthiness byappearing to do more or be more ethicalthan other employees (e.g., ‘‘Try toappear like a hard-working, dedicatedemployee’’). Intimidation is where peopleadvertise their power in order to comeacross as threatening within an organiza-tion (e.g., ‘‘Be intimidating with co-workers when it will help you get yourjob done’’). Finally, supplication is whenan employee comes across as weak andneedy so that he or she can be protectedby others (e.g., ‘‘Act like you know lessthan you do so people will help youout’’).

2. Another commonly used political tactic isinformation management. In many organi-zations, ‘‘information is power’’; thus,one way to advance one’s political agendais to control others’ access to information.This may include simply controllingwhether others ever receive informationand the timing of the information’srelease. In political campaigns, for exam-ple, candidates often withhold negativeinformation about their opponent untiljust before the election. By doing so, theyleave the opposition little time to engagein any form of ‘‘damage control’’ thatmight save the election.

3. A political tactic that is somewhat coun-terintuitive, but often highly effective,is promotion of the opposition. This mayinvolve eliminating a political rival byhelping the person become so successfulthat he or she is promoted to a higherposition in the organization and nolonger poses a threat. Using this tactichas a double advantage: The employeeappears to be gracious, and an individ-ual who may be a roadblock en routeto the desired political objectives iseliminated.

4. A final political tactic used in organiza-tions is an employee’s promotion ofhis or her own agenda by pursuingline responsibility—actively seeking aposition within the organization thatmakes it easier to exert one’s influence.In most organizations, some positionsare crucial to the main business of theorganization, and others are consideredperipheral. As a general rule, positionsthat are close to the core technologyof an organization (e.g., production, re-sources acquisition) carry higherlevels of influence than positions indepartments designed to support thattechnology (e.g., research and develop-ment, human resources).

The political tactics described to thispoint are relatively benign, but certain tacticsreflect the ‘‘dark side’’ of political behavior inorganizations. According to DuBrin (1993),more destructive political tactics include theelimination of one’s political rivals, use of a‘‘divide and conquer’’ strategy, and exclusionof one’s political adversaries. Political battlesin organizations can be brutal. In some caseswhen members of organizations are compet-ing with each other, the ‘‘winner’’ is able tofacilitate the exit of rivals by getting themfired or making their lives so difficult thatthey leave voluntarily.

The ‘‘divide and conquer’’ strategy maysurface in situations in which one individualis at odds with a group of other employees. Itis often difficult for an individual to imposehis or her will on such a group because of thenumerical difference. Thus, one way to over-come this situation is to induce conflictwithin the group, making it less likely thatthese individuals will put up a united front.Managers in many types of organizationsoften bemoan the lack of interpersonal har-mony within work groups. However, the


irony is that the existence of interpersonalconflict often makes it much easier formanagers to control their groups and toadvance their personal agendas.

Excluding one’s political rivals simplyinvolves making sure that they are ‘‘out ofthe loop’’ and thus less likely to influenceone’s agenda. As stated earlier, in many orga-nizations, information is power. Thus, oneway to undercut one’s rivals is to make surethat they do not receive crucial informationthat would make it easier for them to exertinfluence. In practice, this form of influencemay involve making sure that one’s rivals arenot invited to important meetings, or perhapsseeing to it that they receive job assignmentsin remote areas of the organization.

Unfortunately, not a great deal of em-pirical research has been devoted to thestudy of organizational politics. The littleresearch that has been done, however, sug-gests that political behavior has a negativeimpact on organizations, particularly whenemployees lack an understanding of thepolitical landscape (e.g., Ferris, Gilmore,& Kacmar, 1990). When one considersthe tactics previously described, this is notsurprising. The atmosphere in an organiza-tion with a great deal of political behavior islikely to be characterized by tension, mis-trust, and, in extreme cases, downright par-anoia.

In addition, Vigoda and Cohen (2002)recently conducted a longitudinal studyinvolving employee influence tactics, metexpectations on the job, and perceptions oforganizational politics. These authors foundthat greater use of influence tactics at Time 1was related to lower met expectations ofemployees (employees feeling the organiza-tion did not live up to what they expected) atTime 2, which was then predictive of per-ception of organizational politics at Time 2.These results illustrate the link between high

levels of influence and perception of organi-zational politics.

It is not realistic to think that politicalbehavior can be (or perhaps even shouldbe) eliminated from organizations. However,there may be ways organizations can decreasethe behavior. Political behavior is often theby-product of uncertainty and ambiguity, sobeing clear about organizational goals andindividual employees’ job assignments is animportant step toward reducing destructivepolitical behavior. Organizations can alsoreduce political behavior by breaking upobvious cliques or coalitions through trans-fers or through job rotation. If individualsconsistently engage in destructive politi-cal behaviors, organizations may be able toreduce these behaviors by confronting theoffenders. Often, employees in organizationswill ‘‘get away with’’ destructive politicalbehaviors simply because they are never con-fronted about it.

Perhaps the most important way thatmanagers can decrease political behavior isby setting a good example for subordinates. Ifa manager is honest and above board in his orher dealings with others in the organization,handles conflicts with others in a construc-tive manner, and conveys to subordinatesthat highly destructive political behavior willnot be tolerated, this sends a powerful mes-sage. Although political behavior in organi-zations may not be eliminated, it may bepossible to decrease it to a nondestructivelevel.

CHAPTER SUMMARY

This chapter focused on leadership and theclosely related topic of influence processes.The study of leadership has been approachedfrom trait, behavioral, and contingency per-spectives. Although most modern theories ofleadership can be considered contingency

Chapter Summary 335


STEPHEN ZACCARO AND THE REEMERGENCE OF RESEARCH ON LEADER

ATTRIBUTES

My interest in leader attributes and leadership

grew from a term paper I wrote for a group

dynamics class while in graduate school. Inthat paper, I had cited an earlier study by

Barnlund (1962) that seemed to provide sup-

port for the prevailing notion that the partic-

ular occupant of the leader role can vary from

situation to situation depending upon the set

of skills and attributes needed in each situa-

tion. David Kenny was my instructor, and he

noted in his grading comments that the resultsof the Barnlund study could be reanalyzed

to partition the variance in leader emergence

more precisely to different sources. He devel-

oped the appropriate statistics for the reanal-

ysis; we applied them to the original study,

and found that a large portion of the variance

could be attributed to attributes of the leader

rather than the situation.We published the results of this reanalysis

in a paper that emphasized trait-based sources

of leadership variance. This was in the early

1980s when much of the zeitgeist in leadership

research was decidedly in a different direction.However, Robert House had already begun to

write about charismatic leadership and about

personality and leadership. Then, Robert

Lord and his colleagues published their

meta-analysis in 1986 supporting a stronger

link between leader attributes and emergence.

These research lines gave some new impetus to

trait-based perspectives of leadership.In our paper, Kenny and I had speculated

that attributes that predicted leader emergence

wouldincludequalities thatpromoteda leader’s

behavioral flexibility to varying situational

demands.IconductedsomeresearchwithRose-

anne Foti and Dave Kenny to test this notion.

We found support linking self-monitoring to

leader emergence across different situations.This study sparked for me an ongoing and

continuingprogramof research to identify lead-

er attributes that promote effectiveness in

dynamic and multi-faceted organizational

domains. We have learned from this work that

leadership can likely be explained by complex

integrations of leader attributes, residing in a

single person or perhaps shared among teammembers, which foster responsiveness to vary-

ing situational demands. We also learned the

value of revisiting old ideas, applying more

sophisticated methods and approaches to gain

new insights.

Stephen J. Zaccaro


George Mason University


theories, the trait and behavioral approachesare by no means dead; they still offer someinsight into leadership processes.

Fiedler’s Contingency Theory proposesthat the effectiveness of a leader hinges onthe match between situational favorabilityand whether the leader is task or relationshiporiented. This theory has received onlymixed support, but it has generated a con-siderable body of leadership research. It alsoserved as the impetus for other contingency-based leadership theories in subsequentyears.

Path–Goal Theory also proposes thatleader effectiveness depends on the leader–situation match. It differs from Fiedler’stheory, however, in the manner in whicheffectiveness is defined, and in proposing thatleaders are able to adapt different forms ofleadership behaviors to different situations.Although Path–Goal Theory still awaits moreempirical scrutiny, it serves as a usefulguide to the understanding of leadershipand may have considerable practical benefitsas well.

The Vroom–Yetton–Jago model of lead-ership is focused on one aspect of leader-ship behavior: decision making. This theoryis somewhat different from the others inthat it is largely prescriptive in nature; thatis, it provides managers with guidelines fordecision making. Support for this modelhas been strong when managers have beenasked to recall decisions, but resultshave beenmore equivocal when other sources of dataare used.

The Leader–Member Exchange (LMX)Theory proposes that leaders develop aunique relationship with each of their sub-ordinates that is largely based on socialexchange. This theory represents a vastdeparture from previous theories that werebased on the rather naıve assumption that

leaders treat all subordinates the same. Re-search on LMX Theory has yielded veryinteresting findings on both the determi-nants and the consequences of differencesin exchange relationship quality. Furtherwork, however, appears to be needed todefine the dimensions of the exchange rela-tionship and to broaden the scope of LMXresearch. Some recent research has usedLMX processes to explain how transforma-tional leaders produce superior performancein their subordinates.

The most recent heavily researchedtheory of leadership described was Trans-formational and Transactional Leadership.To some extent, this approach represents areturn to the trait approach that dominatedleadership research in the early twentiethcentury. Transformational leaders not onlylead others but inspire them as well. Theseindividuals also are capable of facilitatingmeaningful change in organizations. Re-search in this area has been largely descrip-tive. Transactional leaders emphasize thecontingencies necessary for employees toreceive rewards and keep track of employeebehaviors to deliver contingencies. Recentmeta-analyses show that both transforma-tional and transactional leaderships arerelated to employee performance, and thatthe two forms of leadership are actually pos-itively related. We briefly introduced a recenttheory of authentic leadership that empha-sizes leaders acting in ways consistent withtheir self-concept and showing moral char-acter. Future research is necessary to assessthis approach.

Power and influence are at the core ofleadership; therefore, both topics were cov-ered in conjunction with leadership theories.Research has shown that leaders typicallyhave multiple bases from which to exertpower, and, in some cases, these bases may

Chapter Summary 337

be situationally specific. Influence tacticsrepresent the various ways in which leadersexert their power in organizations. Researchhas shown that the most effective tactics arethose that give subordinates some freedom ofchoice, and the least effective tactics arethose that involve pressure and appeals toone’s formal authority.

Organizational politics represents a dis-tinct form of influence that, in many cases,can be destructive. Political behavior mayoccur in any organization, but it is typicallymore prevalent in organizations that have agreat deal of uncertainty and scarce resources.Specific political tactics may take a variety offorms—some more negative than others.Although relatively little research on organi-zational politics exists, there is some evidencethat the impact of political behavior is nega-tive. Although political behavior can neverbe eliminated completely, organizations canreduce it by improving communication and,in some cases, increasing resources. Ulti-mately, the most effective way for managersto reduce political behavior is to set a positiveexample in their dealings with subordinatesand others in the organization.


Graen, G. B., & Uhl-Bien, M. (1995).Development of leader-member exchange(LMX) theory over 25 years: Applying amulti-level multi-domain perspective.Leadership Quarterly, 6, 219--246.

Judge, T. A., & Piccolo, R. F. (2004).Transformational and transactional leader-ship: A meta-analytic test of their relative val-idity. Journal of Applied Psychology, 89, 755--768.

Klein, H. J., & Kim, J. S. (1998). A fieldstudy of the influence of situational con-straints, leader-member exchange, and goalcommitment on performance. Academy ofManagement Journal, 41, 88--95.

Waldman, D. A., & Yammarino, F. J.(1999). CEO charismatic leadership: Levels-of-management and levels-of-analysis effects.Academy of Management Review, 24, 266--285.

Zaccaro, S. J. (2007). Trait-based per-spectives of leadership. American Psycholo-gist, 62, 6--16.


Chapter Eleven

Group Dynamicsand BehaviorM

ost employees belong to someformal work group, and orga-nizations often establish tem-porary or ad hoc groups toaccomplish many important

tasks. Furthermore, different groups of peo-ple within an organization often work to-gether to accomplish objectives. Thus, agreat deal of behavior in organizations takesplace within group situations. It is thereforeessential to examine behavior in groups inorder to obtain a complete understanding ofbehavior in organizational settings. Theprevalence of groups in organizations mir-rors their presence in everyday life. For ex-ample, most people are part of a family andbelong to groups in their communities orchurches, or within their professions. It is agood bet that most readers belong to a varietyof groups, and that membership in thesegroups has an important impact on theirbehaviors and attitudes.

Organizations make frequent use ofgroups for an obvious reason: A group canaccomplish more than an individual. Forexample, a group of five firefighters can obvi-ously bring a fire under control faster thanone firefighter. Groups are also used becausethe output of several people working on atask may be better—or, in some cases, morecreative—than if each person approachedthe task individually. A third reason thatgroups are frequently utilized is simplyconvention. Organizational scholars (e.g.,Hackman, 1992) have noted that behaviorin organizations is often driven by socialinertia, or relying on familiar ways of doingthings. Because groups have been used in the

past, they are sometimes used without muchthought as to whether they are appropriatefor a given task.

Given the importance of groups in orga-nizations, two chapters are devoted to thistopic. The present chapter serves as anintroduction to group behavior through ananalysis of the social psychological andorganizational psychology literatures. Hav-ing a firm foundation of group dynamicsand behavior, Chapter 12 focuses specifi-cally on the effectiveness of teams in orga-nizations, and draws largely (though notexclusively) from the organizational psy-chology literature. After reading these

339

chapters, students should have a firmunderstanding of the impact of groups inorganizational settings.

WHY DO PEOPLE JOINGROUPS?

People join groups for a multitude of rea-sons. A major reason is that group member-ship often results in some form of needsatisfaction on the part of the individual.If one takes an evolutionary perspective(e.g., Buss, 1996), group membership mayappeal to individuals’ basic need for sur-vival. Activities that enhance survival, suchas hunting and defense against predators,are often better accomplished collectivelythan individually. Because of this, somehave argued, the tendency to affiliate andform groups has become an adaptive be-havior and thus has endured over manycenturies.

At the present time, basic survival is notat issue for most people; thus, group mem-bership often allows the fulfillment of othertypes of needs. One that is typically satisfiedby group membership is the need for affili-ation. Baumeister and Leary (1995) haveforcefully argued that individuals have astrong need to belong to social groups, andthat failure to be securely included in desiredgroups prompts actions to restore groupinclusion. Thus, people often join groupssimply to be in the company of other people.In fact, when people do not affiliate withothers for prolonged periods of time, thismay lead to psychological adjustment prob-lems or even more severe forms of psycho-pathology (see Comment 11.1).

Another need that is often satisfiedthrough group membership is the need forpower and control (Schutz, 1958). As wasstated in Chapter 10, social power involvesthe capacity or potential to influence the

behavior of others. If one has a strong needto exert power over others, that person needsto be in the company of other people. (It’spretty hard to boss yourself around! Actu-ally, it can be done, but you might get somefunny looks!) Thus, people often do joingroups so they can hold leadership positionsthat allow them to exert power and influenceover other group members.

Besides providing the opportunity forneed satisfaction, group membership oftengives people a greater opportunity toachieve goals than they would have if theywere acting alone. For example, peopleoften join labor unions because they believethey can achieve higher wages and morefavorable working conditions by acting col-lectively rather than negotiating with theiremployers as individuals. Other examplesof organizations joined for this reason in-clude political parties, criminal organiza-tions, churches, lobbying organizations,and consumer advocacy groups. Theseorganizations seek vastly different goals,but they are similar in their use of the powerof collective action to achieve some goal orhigher purpose.

A final reason that people often joingroups is that being around other peopleoften provides comfort and support (Cohen& Wills, 1985; Smith, Murphy, & Coats,1999). Particularly when people are anxious,or when they are experiencing stressful peri-ods in their lives, being around other peoplecan offer a great deal of support. This is espe-cially true when the other people comprisinga group are experiencing the same difficul-ties (Schacter, 1959). Examples of groupmembership can be seen in the numeroussupport groups for people with certain dis-eases (e.g., cancer), survivors of tragedies(e.g., loss of a spouse), or people who aregoing through other major life transitions(e.g., divorce).

340 Group Dynamics and Behavior

Although organizations typically formgroups in order to accomplish work-relatedtasks, work groups may serve a variety ofother purposes for their members, such asaffiliation and social support. This point isoften overlooked in organizational psychol-ogy, but it is important in understanding thebehavior of work groups. We now addressthe defining characteristics of groups, a taskmore difficult than one might imagine.

DEFINING CHARACTERISTICSOF GROUPS

Psychologists who study group behavior aremuch more precise in their definition of

what constitutes a group, as compared tothe way most people use this term in ev-eryday conversation. Although there is no‘‘universal’’ accepted definition of what con-stitutes a group (Forsyth, 2006), there is ac-tually a good deal of consensus on the mostimportant defining characteristics. A termthat is found in most definitions of a groupis interdependence. Specifically, to be consid-ered a group, a collection of people must, insome way, be interdependent. This simplymeans that the outcomes each member of acollective receives depend, to some degree,on the other members of the collective. Inwork situations, interdependence may beseen when one person may need information

COMMENT 11.1

PEOPLE NEEDING PEOPLE: THE IMPACT OF SOCIAL ISOLATION

DO PEOPLE HAVE a strong need to be around

other people? That is, do people suffer when

they are deprived of contact with others? The

answer to this question is rather complex, butthere are many reasons to believe that social

isolation may be detrimental to both mental

and physical health. Research in developmen-

tal psychology, for example, has shown that

children often have severe developmental

delays and other long-term difficulties when

they are deprived of social contact as infants.

Also, studies of individuals such as explorers,scientists working in seclusion, and prisoners

document the psychological suffering that

often accompanies seclusion from others. This

suffering is often greater then the physical

hardships these individuals face.

Another indication of the need for social

contact (and hence the pain associated with

isolation) is the comfort people often find whenthey are with other people, particularly in times

of distress or turmoil. Often, when something

traumatic happens in someone’s life (e.g., loss

of a spouse), the first words of concerned

friends or relatives are: ‘‘He/She shouldn’t be

alone.’’ The soothing effect of social contact

also explains, at least partially, why many

people regularly go to churches, synagogues,and mosques to practice their religious faith. It

is certainly possible to practice one’s religious

faith in isolation; however, doing so with

others often provides a great deal of comfort.

Despite the well-documented negative

effects, isolation from others can provide peo-

ple an opportunity for reflection and personal

growth. Writers, artists, and others engagedin creative endeavors often find inspiration

through solitude and isolation. It is important

to note, however, that, for these individuals,

the isolation is self-imposed and they are typ-

ically able to break their isolation if they want

to. This element of choice appears to be a key

element in whether social isolation is detri-

mental.

Source: D. R. Forsyth. (2006). Group dynamics (4th ed.).

Belmont, CA: Wadsworth/Thompson.

Defining Characteristics of Groups 341

from other employees in order to do his orher job. Interdependence may also exist insocial situations; that is, people may dependon each other for having fun.

Another key defining characteristic of agroup is social interaction. To be considereda group, people must interact with eachother in some way. This typically takes theform of verbal and nonverbal communica-tion. If people are not in the same physicallocation, this interaction may take otherforms (e.g., phone, e-mail). On the otherhand, people who do not interact are typi-cally not considered to be a group. Consider,for example, five people standing in an ele-vator. The people in this situation are essen-tially ignoring each other and would not beconsidered a group.

A third defining characteristic of a groupis the perception of being a group on the partof the actual group members and those exter-nal to the group. There may be instanceswhere people interact with each other, andthey may even be somewhat interdependent,but do not perceive themselves as a group.Consider, for example, the members of awedding party. These individuals certainlyembody the first two characteristics of agroup: They interact and their behavior issomewhat interdependent (e.g., there is usu-ally some predetermined order in which theymust walk down the aisle). In most cases,though, these individuals probably perceivethemselves as a collection of individualsrather than a group. Furthermore, even ifsome level of group identity does developamong these individuals, it is very shortlived.

Psychologists coined the term entitativityto refer to the extent that a collection of indi-viduals is perceived as a whole (Campbell,1958). Lickel and his colleagues (2000)asked participants to distinguish between anumber of different groups in terms of those

that seemed more entitative (groupy) thanothers (Lickel, Hamilton, Wieczorkowska,Lewis, Sherman, & Uhles, 2000). Theauthors found that intimacy groups such asfamilies and friends were seen as most enti-tative, followed by work groups, aggrega-tions of people at a location, and broadsocial categories (e.g., women, lawyers).Groups that perceive themselves as moreentitative are likely to have higher levels ofcohesion than other groups (e.g., Mullen &Cooper, 1994).

A final key defining characteristic ofgroups is commonality of purpose. For a col-lection of people to be a group, they musthave some common goal or other reasonfor existence. Common goals may be quiteformal—for example, for a work group—orquite informal, as would be the case for agroup of friends who get together simplybecause they enjoy each other’s company.The major point is that for a collection ofpeople to be a group, they must have some-thing they are trying to accomplish collec-tively.

Based on these defining characteristics ofgroups, two important points are worthmentioning. First, clearly dividing collec-tions of people into groups and nongroups isoften difficult to do. Some collections ofpeople are more ‘‘grouplike’’ than others.Thus, whether a collection of people consti-tutes a group is more a matter of degree thanit is an absolute judgment. Second, withinorganizational settings, we often use the termgroup incorrectly. A formal work group maysimply be a collection of people who arelinked for administrative purposes butexhibit few of the defining characteristics ofa group. In universities, for example, aca-demic departments are typically consideredgroups, even though the work of the mem-bers (professors) is not usually interdepend-ent. They may interact very infrequently,


may not perceive themselves as a group, andmay disagree vehemently about departmen-tal goals.

This issue has been addressed somewhatin the organizational psychology literatureby using the term team to discuss workgroups that are configured to accomplishspecific tasks (Kozlowski & Bell, 2003).The next chapter deals primary with workgroups in an organizational context, and istherefore titled team (rather than group)effectiveness. The other side of this, how-ever, is that, within organizations, informalgroups may develop, and the impact of thesegroups may be powerful. For example, peo-ple may develop friendship groups based ontheir level of seniority within the organiza-tion or their common interests (e.g., run-ning, playing golf). These informal groupsare important because they may impact em-ployees’ attitudes and can ultimately influ-ence whether employees stay with anorganization. They are certainly as real asformal groups created by organizationalstructures.

GROUP STRUCTURE

To understand groups, it is useful to considergroup structure, which represents a set ofdimensions along which any group can bedescribed. These dimensions also help whenwe are describing differences between groups.The most important elements of group struc-ture are roles, norms, values, communicationpatterns, and status differentials. Each ofthese is discussed next.

Roles

Roles represent prescribed patterns of behav-ior that are specific to a particular individual,or to the position the individual occupies(King & King, 1990). In organizations,

role-related behaviors are often communi-cated to employees through formal docu-ments such as job descriptions, but theymay also be communicated by more informalmeans. As an element of group structure,roles are quite relevant. When groups firstform, there is a great deal of uncertainty andambiguity surrounding what individualgroup members are supposed to do (Levine,Moreland, & Choi, 2001; Tuckman, 1965).Over time, individual group members’ rolesare defined through the process of role dif-ferentiation (Hackman, 1992; Levine et al.,2001). Role differentiation simply representsthe process by which (1) role-related expec-tations are communicated to group mem-bers, and (2) the various roles within thegroup take shape. Groups tend to form rolestructures primarily because this method ismuch more efficient than having ‘‘everyonedo everything.’’

Although the specific roles played bygroup members may be highly specific andtask dependent, more general roles areenacted in all groups. For example, in mostgroups, a member or members plays whatmight be described as the task role. Behaviorsassociated with the task role might includeclarifying task requirements, providingperformance-related coaching and assistanceto group members, and, at times, keeping thegroup focused on the task at hand. In many(but not all) cases, behaviors associated withthe task role are performed by a formallydesignated group leader. It is not unusual,for example, for senior members of a workgroup to share their expertise with newermembers, even when the groups have for-mally designated leaders.

A second role that is commonly seen inmost groups is the socio-emotional role. Be-haviors associated with the socio-emotionalrole are aimed at maintaining the ‘‘socialfabric’’ of the group. At various points in

Group Structure 343

the life of any group, there is a need forsomeone to encourage others, to lightenthe atmosphere with a joke, or to diffuseconflicts among group members. Socio-emotional behaviors could obviously be per-formed by any member of a group, but theyoften are not performed by a formally desig-nated leader when one is present. Exhibitingsocio-emotional behaviors often puts leadersin an awkward position, given that they arealso typically responsible for judging the per-formance of the members of their groups. Inprofessional sports, teams often acquire vet-eran players because they are able to performaspects of the socio-emotional role, such askeeping morale high and diffusing person-ality conflicts among other players.

The third role that is common to mostgroups is labeled the individual role. In allgroups, individual members have individ-ual needs and desires that may or may notbe compatible with those of the group asa whole. Individual group members pursuethese needs to varying degrees, but (hope-fully) not to the detriment of the largergroup goals. However, these needs couldbe problematic when they conflict withthe goals of the group. In many organiza-tions, for example, employees have incen-tives that are based purely on individualperformance.

A final role that is common in mostgroups is that of leader. As discussed in ourlast chapter, leadership comes in many formsand varieties. We bring up the role of leaderin our discussion of groups because evensmall groups will likely designate a leaderwhose responsibility is to coordinate the ac-tivities of the group members and motivatethem to perform well. Leaders within groupsmay be formally designated, or they mayemerge as group members determinewho can most effectively fill that role (For-syth, 2006).

Norms

The normative standards (norms) that areadopted by the group represent a secondimportant dimension of group structure.Norms are explicit or implicit standards thatgovern behavior. According to Hackman(1992), groups adopt norms primarily toincrease the predictability of group mem-bers’ behaviors and, more generally, to keepthings running smoothly within the group.Groups may adopt norms governing a num-ber of aspects of group behavior, such as theformat of meetings, the openness of commu-nication, and even the way people dress.Typically, though, groups adopt norms onlyfor behaviors that are deemed important forthe functioning of the group.

Once norms are adopted, group mem-bers typically ‘‘fall in line’’ and behave inaccordance with those normative stand-ards. In some cases, however, a groupmember may behave in direct violation ofthose norms. If such a violation is uninten-tional, politely bringing it to the person’sattention may be all that’s needed to correctthe behavior. For example, if a new groupmember arrives at a meeting 15 minuteslate because he or she didn’t realize thestarting time, this type of norm violationcan be dealt with simply by bringing it tothe person’s attention or making light ofthe infraction.

What happens when norm violation ismore intentional and occurs repeatedly? Re-search suggests that, in these types of situa-tions, group members should try to modifythe norm violator’s behavior—but only to apoint. If repeated attempts fail to bring thenorm violator’s behavior in line with theprevailing group norms, one of three thingscan happen. First, the group may eliminatethe norm violator. Evidence suggests, how-ever, that this is a fairly drastic step; groups


typically expel norm violators only as a lastresort. There may even be instances in whicha group simply does not have this option.Encouraging group members to assimilatethe importance of group norms is a criticalcomponent of group socialization (Levineet al., 2001), which we turn to in the nextmajor section of the chapter.

A second option is to allow the normviolator to remain a member of thegroup—but a very marginal member. Stateddifferently, the norm-violating group mem-ber may become an institutionalized deviantwho is essentially ignored by the other mem-bers of the group (Dentler & Erikson, 1959).When this happens, the group essentially‘‘gives up’’ on the norm violator and ignoreshis or her behavior. In some cases (but notall), being relegated to the role of institution-alized deviant may be enough to prompta person to leave a group. The negative partof having a person in the role of institution-alized deviant is that the group has essen-tially lost the contributions of one of itsmembers.

A final possibility is that the norm viola-tor may ultimately change the relevant groupnorm. One of the positive things about normviolation is that it may force a group to take acritical look at its prevailing norms. Al-though, in many cases, norms help groupsto function more effectively, they may alsohave negative effects. For example, a groupthat adopts a norm of never acknowledginginternal conflict may function well on thesurface but have problems in the long run.Thus, taking a critical look at prevailinggroup norms may reveal some that are eitheroutdated or dysfunctional for the currentgroup. As Hackman (1992) aptly points out,‘‘Just as it has been said that the unexaminedlife is not worth living, so it may also be saidthat an unexamined norm is not worth enforc-ing’’ (p. 248).

For a norm violator to actually changeprevailing group norms, however, a numberof conditions must be present (Moscovici,1994). Specifically, normative change ismuch more likely if there is at least one othergroup member violating the prevailingnorm; one group member going against thegrain will typically not be enough to change agroup norm. It has also been found that thenorm violator must be consistent in his orher behavior over time, and that the othermembers of the group must see this consis-tency as evidence of the strength of his or herconvictions. Normative change is also morelikely when the difference between the pre-vailing norm and the behavior of the normviolator is reasonably consistent with pre-vailing cultural values. Finally, members ofthe minority must avoid being cast into therole of institutionalized deviants. All in all,one may conclude that prevailing groupnorms can be changed by a norm violator,but this is not an easy process.

Values

Schwartz (1994) has defined values as ‘‘desir-able trans-situational goals, varying in impor-tance, that serve as guiding principles in thelife of a person or other social entity’’ (p. 21).In the context of groups, values representsuch goals or ideas that the group deems asimportant (Rokeach, 1973). When newmembers are socialized into any collectivebody, the values of that collective body arecommunicated either explicitly or implicitly.For example, a group producing a productmay strongly value quality, a group that dealswith customers may strongly value customersatisfaction, and a group that forms for socialreasons may strongly value group members’enjoyment.

Regardless of whether a group commu-nicates its values explicitly through written

Group Structure 345

documents or on a more ad hoc basis, newmembers must accept them in order toremain part of the group. For membershipin most groups, new members are not re-quired to become completely committed toall the values of the group. At some level,however, a new group member must acceptsome of the group’s values in order to remainpart of the group.

Like norms, values may be functionalbecause they often serve as ‘‘rallying points’’for a good deal of group members’ behaviors.For example, the fact that a group stronglyvalues customer service may motivate groupmembers to ‘‘go the extra mile’’ for custom-ers. Values, however, can also be dysfunc-tional. A group that values conformity andagreement over all else may, at times, makevery poor decisions and ultimately be inef-fective (Janis, 1982). Thus, like norms, thevalues of a group must be periodically madeexplicit and critically examined.

Communication Patterns

Another way that groups can be described isby the characteristic patterns group mem-bers use to communicate with each other.In the group dynamics literature, the mosttypical distinction regarding communicationis between centralized and decentralizedcommunication networks (Leavitt, 1951;Shaw, 1964, 1978). When a centralized com-munication network is adopted in a group,communication tends to flow from onesource to all group members. A typical formof centralized communication, termed thehub and spokes model, is depicted in Figure11.1. Notice that in this type of communica-tion network, one individual (often thegroup leader) is the focal point. This indi-vidual takes in information and disseminatesit to the other members of the group. Theadvantage of this type of centralization is that

it allows information to be standardized; thatis, all group members can obtain the sameinformation. This is particularly importantwhen groups are working on tasks that arefairly standardized and routine.

As one looks at Figure 11.1, an obviousdisadvantage is that centralized communica-tion tends to restrict and discourage the flowof communication among group members.Depending on the nature of the task, thismay ultimately have a negative impact onthe performance of individual group mem-bers and the group as a whole. For example,it has been shown that performance onhighly complex tasks is hindered by highlycentralized communication flow (Hackman,1990; Leavitt, 1951). For example, if a groupof researchers is working on developing anew product, communicating only througha supervisor would be highly inefficient.

Highly centralized communication alsohas a less obvious problem: It puts a greatdeal of reliance on one person; that is, the‘‘hub’’ of any centralized communication net-work must take in and disseminate a greatdeal of information and may simply overloadthe person with too much information. Thisform of communication network is also

FIGURE 11.1The Hub-and-Spokes Communication Network

The lines with no arrows indicate that communication

flows in a bidirectional manner.

Adapted from: M. E. Shaw. (1964). Communication

networks. Advances in Experimental Social Psychology, 1,

111–147.


based on the premise that the informationthat is disseminated by the person occupyingthe focal point will be accurate and consis-tent. In a highly centralized communicationnetwork, there is a great deal of potentialfor the person occupying the focal point todisseminate information in a way that enhan-ces his or her political agenda without en-hancing the productivity of the group.

In direct contrast to centralized commu-nication networks, in a decentralized net-work communication flows freely withinthe group. The network depicted in Figure11.2, which is referred to as a Comcon,depicts decentralized communication quiteclearly. Note that everyone in this five-person group communicates with everyoneelse. Compared to a centralized network, aclear advantage of decentralization is that itallows communication to flow wider andfaster. If two members of a group need tocommunicate with each other about animportant issue, they can do so without hav-ing to go though an immediate supervisor.Research has shown that this type of free-flowing communication is an asset, especiallywhen a group is working on a highly complexinterdependent task (Leavitt, 1951). The pri-mary disadvantage of decentralization is thesheer amount of information that is commu-nicated. This is particularly true with thewidespread use of electronic communication.Groups using highly decentralized commu-nication must develop ways of managing theflow of information that do not restrict it, butthat do, at the same time, introduce somelevel of consistency.

Other than their degree of centralization,groups can often be distinguished accordingto whether group members communicatewith each other (Moreno, 1953). In somegroups, members communicate a great dealof information. However, in other groups,communication is very restricted and group

members often find themselves ‘‘out of theloop’’ on important issues. A low level ofcommunication among group members isoften (but not always) symptomatic of aninability or unwillingness of a leader to grap-ple with the issue of communication. It mayalso be symptomatic of more serious prob-lems within a group (Hackman, 1990).

Another aspect of communication can beseen in groups: the differences between for-mal and informal communication patterns.Formal communication methods includethings such as written memos and directives,formal statements of policy, and informationprovided at group meetings. Virtually allengage in some formal communication, al-though the degree to which they stay intouch varies considerably. Formal commu-nication is often necessary for disseminationof important information and policies togroup members.

Communication within groups may alsobe quite informal. Group members may en-gage in small talk before beginning theirworkday, or may discuss work-related mat-ters while socializing after work. Organiza-tions obviously have less control overinformal communication than they do over

FIGURE 11.2The ‘‘Comcon’’ Communication Network

The lines with no arrows indicate that communication flows

in a bidirectional manner.

Adapted from: M. E. Shaw. (1964). Communication networks.

Advances in Experimental Social Psychology, 1, 111–147.

Group Structure 347

more formal modes of communication. In-deed, the ‘‘grapevine’’ can be a very impor-tant source of communication for groupmembers. A key issue, however, is the accu-racy of information that is communicatedwithin informal communication networks.When the information is accurate, informalcommunication networks can be a highlyefficient way to communicate information.On the other hand, when it is inaccurate,this can create big problems within groupsor even within entire organizations (seeComment 11.2).

Status Differentials

As groups develop, there are typically differ-ences in status among the various groupmembers. The reasons for such status differ-

ences are generally classified into two generalcategories: diffuse status characteristics andtask-specific status characteristics. Diffusestatus characteristics are those that are notdirectly related to the task the group is per-forming but are still seen as legitimate baseson which to attribute status. In many soci-eties, status is based on one’s occupation.In American society, those employed inprofessional occupations, such as medicine,law, and engineering, are typically accordedhigher status than individuals in occupationssuch as sales or manual labor. Many societiesalso attribute status based on demographicvariables such as age, gender, and ethnicorigin. Evidence of the impact of diffusestatus characteristics can clearly be seen instudies of jury deliberations (e.g., Schneider& Cook, 1995).

COMMENT 11.2

INFORMAL COMMUNICATION NETWORKS: THE ‘ ‘GRAPEVINE’ ’

MOST OF THE research on communication in

groups and organizations deals with formal

systems of communication, but we know that

much of the communication that takes place

in groups and organizations is informal. Theterm grapevine is often used to describe such

informal communication networks. Anyone

who has worked in an organization has prob-

ably acquired or distributed information

through the grapevine.

Managers in organizations are often leery

of grapevine communication; they fear that

such information is often inaccurate. Are man-agers justified in their distrust of the grapevine?

Not a great deal of research has been done on

this form of communication, so the evidence is

somewhat mixed. Some research has shown

that managers’ distrust of the grapevine is justi-

fied; information communicated through the

grapevine is often inaccurate. Other research,

however, has shown that the accuracy of grape-

vine communication is quite high, especially

when the information being communicated is

not controversial.

We know one thing for sure about thegrapevine: It is much quicker than more formal

communication channels. News often travels

fast in organizations. Thus, rather than

completely avoid it, managers can sometimes

use the grapevine as a way to quickly commu-

nicate information. This obviously must be

done carefully, but it may be a very useful

way to work around more cumbersome formalcommunication channels when information

must be communicated rapidly.

Sources: R. Hershey. (1966). The grapevine—Here to stay

but not beyond control. Personnel, 20, 64; and B. Smith.

(1996). Care and feeding of the office grapevine. Manage-

ment Review, 85, 6.


Groups may also attribute status based ondifferences in group members’ relative contri-butions to the task the group is performing.As one would expect, an individual who hascontributed a great deal to the group’s taskperformance, or who is capable of doing so inthe future, is typically accorded higher statusthan a group member whose contributionsare more limited (Strodtbeck & Lipinski,1985). For example, in a military combatteam, prior combat experience would be seenas a relevant basis on which to attribute sta-tus.

In work situations, one obvious and verytangible consequence of status differentials isthat higher-status group members are typi-cally paid more than those with a lowerstatus. Whether such differentials are basedon diffuse or task-specific status character-istics, they are typically reflected in employ-ees’ paychecks and, in some cases, thesedifferences are quite dramatic. For example,in professional sports, star players are oftenpaid several million dollars more than play-ers of lower status.

A somewhat less tangible consequence ofstatus differentials is that they typicallyimpact a group’s level of tolerance for normviolation. Put simply, compared to lower-status group members, high-status groupmembers are given greater latitude in violat-ing group norms. Hollander (1971) intro-duced the concept of idiosyncrasy creditsto explain this process. Idiosyncrasy creditsare akin to financial currency that groupmembers ‘‘bank’’ and may use in the eventthat they must violate a group norm. Themore idiosyncrasy credits one has accrued,the more latitude one is given in violating agroup norm. In most group situations, high-status members have considerably more idi-osyncrasy credits than lower-status membersdo and thus are able to violate group normswith fewer repercussions.

One important point to note about idio-syncrasy credits is that even high-statusgroup members do not have an infinite sup-ply. Each time a high-status group memberviolates a norm, his or her idiosyncrasy cred-its are reduced by an amount proportionateto the importance of the norm. Thus, evenhigh-status group members may eventuallyrun out of idiosyncrasy credits and risk beingsanctioned by the group for norm violations.This often occurs in professional sports,when teams initially accept unconventionalbehavior on the part of talented players, butultimately reach a point where such behavioris not tolerated and the player is released ortraded.

In addition to running the risk of deplet-ing their supply of idiosyncrasy credits,high-status group members must be carefulthat their norm violations do not negativelyinfluence the group. Research has shownthat when a norm violation on the part ofa high-status group member negativelyimpacts a group, this behavior is viewedmore negatively than similar behavior onthe part of a low-status group member (For-syth, 2006). This is most likely due to the factthat high-status group members have muchgreater potential than lower-status membersto positively impact a group. Thus, a normviolation on the part of a high-status groupmember that negatively impacts the group ismuch more visible and salient than similarbehavior coming from low-status members.

STAGES OF GROUPDEVELOPMENT

Now that the basic dimensions of groupstructure have been described, we turn tothe issue of how groups develop and changeover time. All groups are somewhat uniquein the way they are formed and the manner


in which they may change over time. Despitethis uniqueness, group dynamics researchersand theorists have identified a great deal ofcommonality in the way group behaviorunfolds over time. Three of the most populartheoretical models describing the process ofgroup development are described in the fol-lowing paragraphs.

Tuckman’s (1965) Stage Model

Tuckman (1965) reviewed 50 articles dealingwith development processes in a variety ofgroups (e.g., therapy groups, sensitivity train-ing groups, naturally occurring groups, labo-ratory groups) and concluded that there was agood deal of commonality in the processes bywhich these groups developed over time.Based on these findings, he proposed a stagemodel of group development that ultimatelybecame quite popular and has endured verywell over time (see Forsyth, 2006; Kozlowski& Bell, 2003). The stages in this model arepresented in Figure 11.3, and reflect themajor issues that a group must grapple withat various points in its development.

As can be seen, the first stage in themodel is forming. This is the beginning pointin the life of a group and is typically charac-terized by a great deal of uncertainty, or evenanxiety, on the part of group members. Thisoccurs because members of a group may beunfamiliar with each other and may havevastly different expectations about what toexpect from membership in the group. Atthis point in the life of a group, members maydeal with such uncertainties by dependingheavily on the group leader for informationand direction. Ultimately, though, the uncer-tainties that accompany membership in anew group gradually dissipate over time asgroup members acquire information and feelmore comfortable being part of the group.

After issues associated with being in anew group are resolved, the next stage ingroup development is labeled storming. Asone might guess from its name, this stageis characterized by conflict over a numberof issues. For example, group members maydisagree over important group norms, orperhaps over who should assume leadershipresponsibilities. This stage may be ratherunpleasant, but it is also necessary if thegroups hope to ultimately function effec-tively. If group members never acknowledgetheir disagreements, these may ultimatelycome out in subtler ways and prevent thegroup from ever performing effectively. Itshould be noted, however, that it is alsopossible for group members to be too vigor-ous in airing their differences. If conflicts aretoo intense and personal, group membersmay simply be unable to work together andnever move past this stage. It is perhaps notsurprising that groups containing peoplewith agreeable personalities experience lessconflict than groups with people prone to be

FIGURE 11.3Tuckman’s (1965) Model of Group Development

Forming

Storming

Norming

Performing

Adjourning


disagreeable (Graziano, Jenson-Campbell, &Hair, 1996).

Assuming that the conflicts identifiedduring the storming stage can be resolved,the group next moves into the norming stage.In a very real sense, this is the point wherea ‘‘collection of people’’ becomes a ‘‘group.’’Some level of role differentiation in groupmembers’ behavior occurs, and the behaviorof the group develops some consistent pat-terns. For example, different group mem-bers may serve different functions, and thegroup may develop norms with regard tomeetings, modes of communication, and,perhaps, the way group members areexpected to dress. Once the norming stageis reached, the group is capable of function-ing as a collective body, rather than simplyas a collection of individuals.

After a group has reached the normingstageand iscapableofworkingasanintegratedunit, the next stage in group development isperforming. This is the point at which a groupaccomplishes the major taskor tasks for whichit was formed. For example, if a group wasformed to develop a strategic plan for an orga-nization, this would be the point at which thegroup would actually come up with that plan.As Tuckman (1965) and others (e.g., Hack-man, 1990, 1992) point out, not all groupsreach this stage in group development. Prob-lems during the earlier stages of group devel-opment (e.g., unresolved conflicts) mayprevent a group from accomplishing its majortasks. All groups, however, have the potentialto reach this stage.

What happens when a group ultimatelyperforms the task for which it was formed? Agroup may simply keep on performing thissame task or move on to another one. Inmany other cases, a group is disbanded andthe individuals move on to other activities. Inrecognition of this fact, Tuckman and Jensen(1977) added a fifth stage, adjourning, to the

original model. In some cases, particularlywhen groups are formed for a very shortduration, the adjourning phase is relativelymundane. That is, group members simplymove on. However, when group membersare together for a long period or the groupexperience is very intense, this can be adifficult time. Group members may genu-inely miss each other and have feelings ofloss or abandonment. Over time, groupmembers will usually overcome these feel-ings, but initially it may be very difficult.

During the adjourning stage, the groupmembers may reflect on their experiences inthe group. Did they feel that the group wassuccessful? Was working in the group arewarding experience? Did they enjoy work-ing with the other members of the group?These types of reflections are important, fortwo reasons. First, they may influence mem-bers’ general views about working in groups.People who have negative experiences maybe hesitant to work in groups in the future.Second, such reflections may strongly influ-ence whether the members of a particulargroup can work together in the future. Inorganizations, ad hoc groups often have toform and adjourn several times. If the mem-bers of a group have very negative reflectionsof being in a group, they may be unable toreconvene the group and function effectivelyin the future.

Tuckman’s (1965) model is certainly use-ful in describing developmental processes ina great many groups. However, there areinstances in which groups do not strictlyadhere to the sequence described in themodel. For example, a group of individualsmay have to come together and performimmediately, and address other issues later.There may also be cases in which a group ofindividuals immediately adheres to a strongset of norms, and deals with other issueslater. The important point for readers to


understand is that this model explains groupdevelopment in general.

A second and related point is that Tuck-man’s (1965) model is best thought of ascyclical. For example, a group may progressall the way to the performing stage, but mayhave to digress to storming if important con-flicts among group members arise. Also, inmany instances, the composition of a groupmay change over time. Each time a newmember joins a group, certain elements ofthe forming stage are replayed. For the newmember, this period may be fraught withuncertainty and anxiety. However, longer-tenured group members may also haveapprehensions about the new group mem-ber(s), and thus may experience a good dealof uncertainty of their own. The importantpoint here is that real groups in organizationsdo not develop in a lock-step fashion. Overtime, they cycle back and forth between dif-ferent stages.

Moreland and Levine’s (1982) Model

Group development can also be viewed interms of the manner in which new groupmembers are socialized into—and ultimat-ely out of—a group. Moreland and Levine(1982) proposed a model to explain thissocialization process (see also Levine etal., 2001). Although not a model of groupdevelopment per se, this model is useful inhelping to understand many of the impor-tant transitions that occur in groups overtime.

The major premise underlying Morelandand Levine’s (1982) model is that the sociali-zation of group members occurs in five uniquestages. These stages are demarcated by thetransitions an individual makes progressingfrom first being an outsider, then a groupmember, and, ultimately, an ex-member.The other important portion of Moreland

and Levine’s model, especially with respectto group development, is that each stage inthe model is characterized by unique proc-esses on the part of the individual seekinggroup membership and the group itself.

The full model is depicted in Figure 11.4.As can be seen, the first stage in the model isinvestigation. At this point in time, an indi-vidual is only a prospective member of agroup. From the individual’s point of view,this period is focused on gathering informa-tion about the group because information ofthis sort will help the individual decidewhether he or she will ultimately seek mem-bership in the group. The group, at this point,is trying to recruit or attract prospectivemembers. To do this, the group may provideprospective members with information and,in most cases, emphasize the positive aspectsof group membership and downplay thenegative ones. The recruitment process wasdescribed in some detail in Chapter 3.

FIGURE 11.4Moreland and Levine’s (1982) Model of GroupSocialization

Investigation Stage:Recruitment/Reconnaissance

Socialization Stage:Accommodation/Assimilation

Maintenance Stage:Role Negotiation

Resocialization Stage:Accommodation/Assimilation

Remembrance Stage:Tradition/Reminiscence


If a prospective member decides thatmembership in a group is attractive, he orshe may decide to seek membership. Assum-ing that the individual is allowed by thegroup to do so, the individual moves on tothe next stage in the model, which is social-ization. At this point, the individual entersthe group and makes the important transi-tion from being an outsider to being a newmember. The major task for the new memberat this point is trying to fit in with the groupand assimilate its norms and values. Themajor task for the group at this point is tofacilitate this assimilation process by educat-ing the new group member. This may also bea time of adjustment for the group, however,because they may have to accommodate oradjust to some of the unique characteristicsof new group members. New members willnot give up all of their individuality. Thus,the major issue at this point is balancingconformity to group norms with new mem-bers’ individuality.

Assuming that such a balance can beachieved, the individual is accepted intothe group as a full member. This is repre-sented by the maintenance stage in the mod-el. The major issue at this stage is for thegroup member to carve out his or her nichethrough the process of role negotiation.Through this give-and-take process, theindividual ultimately settles into a comfort-able pattern of behavior as a group member.Depending on the nature of the group, thisstage may last for a short or long period oftime. An individual may remain a memberof a group for several years, although thespecific roles he or she plays may vary overtime, depending on this role-negotiationprocess.

Eventually, an individual’s views maydiverge from those of the group, and theindividual may become a marginal groupmember. This is represented by the resocial-

ization stage. This is a critical period becauseit represents a point at which the individualdiverges from the group in important ways.If such divergence cannot be resolvedthrough the same assimilation and accom-modation processes that were operating dur-ing the initial socialization, the individualmay ultimately leave the group. On the otherhand, if the individual can be successfullyresocialized, he or she will again become afull member of the group.

If resocialization is not successful and theindividual ultimately decides to leave the gro-up, this leads to the final stage in themodel: remembrance. At this point, the indi-vidual assumes the role of an ex-memberand reflects on his or her experience in thegroup. Such reflections may be positive ornegative, but they will undoubtedly have animpact on the individual’s subsequent behav-ior in group settings. From a group’s perspec-tive, the major task at this point is to maintainsome level of stability or tradition, even tho-ugh the exit of a member represents an imp-ortant change (see Comment 11.3).

Although Moreland and Levine’s (1982)model is designed primarily to explain theprocess of group member socialization, it isalso useful in helping to understand howgroups develop over time. Most groups gothrough this cyclical process of bringing innew members, socializing them, and ulti-mately seeing them leave. Thus, this modelprovides useful insight into the processesthat groups engage in, at least relative tothe entry and exit of specific group members.

Gersick’s Punctuated EquilibriumModel

In the group dynamics literature, the ideathat groups strive to maintain an equilibriumpoint between task accomplishment and theinterpersonal needs of group members has


been proposed for some time (e.g., Bales,1965). Building on this general principle,Gersick (1988) proposed that groups mightgo through periods of relative inertia versusrapid change depending on group members’awareness of time and deadlines. For exam-ple, if a task force has 2 months to develop astrategic plan for an organization, Gersick’s(1988) model proposes that this group mayspend a good portion of the first month

‘‘spinning its wheels’’ in an attempt to definethe task, decide on how best to approachthe task, and possibly deal with internalconflicts. However, once the halfway pointis reached, the model predicts that groupmembers will recognize that their time islimited, and a great deal of progress will bemade in a relatively short period of time.

Gersick’s (1988) model is well suited toorganizational settings because groups in

COMMENT 11.3

GROUPS THAT LIVE ON: DAVE, SAMMY, GARY, . . . AND DAVE?

STUDENTS ARE OFTEN surprised to find that

college professors enjoy listening to music

other than Mozart, Bach, or some other titan

of refined classical music. Both authors enjoyVan Halen, and one (SMJ) even listens to the

group when writing manuscripts. The other

author’s (TWB) claim to fame is seeing Van

Halen perform in support of the album titled

‘‘1984,’’ which was the group’s last concert

with lead singer David Lee Roth.

In terms of group dynamics, one of the

things that is interesting about Van Halen ishow they have been able to remain highly

successful, despite having to replace a key

member of the group three times. When the

group started in the 1970s, David Lee Roth was

the lead vocalist. He remained with the group

until the mid-1980s, when he left to pursue a

solo career. After Roth’s exit, the group hired

Sammy Hagar, who remained with the groupuntil the mid-1990s. Like Roth, Hagar left the

group to pursue a solo career. Following

Hagar’s departure, the group turned to Gary

Cherone. Cherone had the shortest tenure with

the group; he recorded only one album. Since

Cherone has left, the group remains without a

lead singer. However, in a somewhat odd twist

of fate, Van Halen has recently announced thatDavid Lee Roth will once again be fronting the

group for a tour in the fall of 2007.

Throughout these key personnel changes,

the group has still managed to be highly suc-

cessful. Why? One of the reasons is that,

despite having three lead vocalists, the othermembers of the group have remained constant.

Eddie Van Halen, Alex Valen, and Mike

Anthony have been members of the group from

the start (although note that Mike Anthony has

been replaced in the recent incarnation of Van

Halen by Eddie Van Halen’s 15-year-old son,

Wolfgang). They have provided the group with

stability and a distinctive sound that isindependent of the lead vocalist. Another

key to the group’s success is that, along with

stability, they have had the ability to adapt. The

style and lyrics of the group’s music have

changed somewhat with each of the changes

in the lead vocalist.

Perhaps the lesson here is that, in order for

a group to be successful over time, its membersneed to have some level of stability in their

membership. This serves to maintain continu-

ity and perhaps reinforces important norms

and values. However, when groups have

changes in personnel, they must be flexible

and willing to utilize the unique talents of these

new group members.

Source: http://www.van-halen.com.


organizations are typically formed to ac-complish meaningful tasks and often haveto do so within a specific time frame. Incontrast, groups in laboratory settings oftendo not have meaningful tasks to perform,and deadlines have little relevance becausethe duration of their activity is very limited.Furthermore, this model has received sup-port using a variety of methodological ap-proaches (e.g., Gersick, 1989; Hackman,1990; Page, Davis, Berkow, & O’Leary,1989).

The major practical implication of Ger-sick’s (1988) Punctuated Equilibrium Modelis that managers should be patient withgroups when they are beginning to workon a task. This model also suggests that it isdesirable for the members of task-performinggroups to be aware of deadlines. Based onGersick’s model, one would assume that ifgroup members are not made aware of dead-lines, this will increase the probability thatthe group will flounder indefinitely and, ulti-mately, will never become productive.

Thus far, the models and issues we havediscussed have emphasized intragroup dy-namics. The focus has been on processes thatoccur within groups and between groupmembers. The remainder of this chapterfocuses on intergroup dynamics, where theemphasis is on how groups interact with oneanother. We will see that groups can worktogether to accomplish joint tasks or cancompete with one another in escalating spi-rals of conflict. Understanding group behav-ior in organizations requires an analysis ofboth intragroup and intergroup processes.

INTERGROUP DYNAMICSAND BEHAVIOR

To highlight the importance of intergroupbehavior, consider an example of a fairlytypical organizational activity: hiring a new

employee. In most organizations, a decisionto hire a new employee typically begins inthe department or unit that is doing thehiring. For example, in a university setting,if a psychology department wants to hire anew organizational psychologist, that de-partment would typically send a hiringrequest to a university or college committeethat is set up to handle faculty staffing issues.Assuming that the request to hire is ap-proved, the psychology department wouldthen work with the human resources depart-ment, and perhaps the college dean’s office,throughout the hiring process. Notice thatthroughout this process a great deal of inter-group cooperation is needed to successfullycomplete the task of hiring a new facultymember.

In discussing the determinants and con-sequences of intergroup behavior, we firstdescribe the various forms of intergrouptransactions that typically occur in organiza-tions. We then address the factors that influ-ence both the frequency and the quality ofintergroup interactions. From there, weaddress the issue of intergroup conflict andstrategies that can be used to keep intergroupconflict to manageable levels.

TYPES OF INTERGROUPINTERACTIONS

Given the vast array of organizational tasksthat may require interactions among two ormore groups, intergroup behavior mighttake a variety of forms. According to Thomas(1976), the nature of intergroup interactionsdepends largely on the degree to whichgroups must interact in order to achieve theirgoals (i.e., their interdependence), and thedegree of compatibility between the goals ofdifferent groups. Crossing these two factorsyields the matrix in Figure 11.5.


As can be seen at the bottom of thematrix, there are some instances in whichgroups simply do not have to interact inorder to achieve their goals. Groups canaccomplish some goals by acting more orless autonomously. If this is the case, and ifthe goals of two groups are reasonably com-patible, interaction between these twogroups can best be described as accommo-dation. This means that there is some cordialgive and take between the two groups. Anexample of accommodation might be twobaseball teams deciding the order in whichthey are going to use the field for pregamepractice. This is obviously not a crucial issue,so the coaches of the two teams might have acordial discussion and make a decision.

Where interaction is not crucial and thegoals of two groups are incompatible, thetypical interaction is best described as avoid-ance; the groups can simply avoid interactingwith each other. In many organizations,there are groups or departments that arepursuing vastly different and, to a largeextent, conflicting goals. A group of re-searchers working on new product develop-ment is pursuing a goal that is far different

from that of a group of engineers attemptingto streamline current production processes.Such groups obviously have different goals,and these goals are largely independent.Thus, in many organizations, these types ofgroups can avoid interacting with each other.

Given the level of complexity in today’sorganizations, and the interdependence ofmany tasks, the reality is that groups oftenmust interact with other groups in order toaccomplish their goals. This state of affairs isreflected in the top portion of the matrix inFigure 11.5. As can be seen, when interac-tion is crucial to goal accomplishment andthe goals of two groups are largely compat-ible, the interactions between groups canbest be described as collaboration. An exam-ple of this might be two research groups thatare researching the same phenomenon. Onegroup may possess expertise that the otherdoes not have. Thus, by working togethercollaboratively, these two research groupsaccomplish more than they would if eachwas to work independently.

Where interaction is crucial to goalattainment, yet the goals of groups areincompatible, the interactions are typicallycharacterized by competition. Two top colle-giate football teams that are striving for anational championship are interdependentin the sense that they must play each otherin order to achieve their goals. However,their goals are obviously diametricallyopposed—there can only be one nationalchampion. Thus, when these two teams play,there is a very high level of competition.There may also be instances in organizationswhen groups compete with each other. Salesgroups may compete for awards, differentacademic departments may compete forscarce resources, and different political par-ties may compete for the right to govern. Onenegative aspect of competition is that it maylead to destructive conflict among groups.

FIGURE 11.5Conceptual Map of the Various forms of intergroupBehavior

Interdependence

Goal Compatibility

High

Low

Low High

Avoidance Accommodation

Compromise

Competition Collaboration


On the other hand, if approached in a pos-itive manner, competition has the potentialto bring out the best in a group (see Com-ment 11.4).

The final form of intergroup interactiondepicted in Figure 11.5 is compromise. Thisform occurs when interaction is moderatelyimportant to goal attainment and there is alsoa moderate degree of goal compatibility be-tween groups. Probably the best example ofthis form of interaction is that which occursbetween unions and management withinunionized organizations. It is not crucialfor these groups to interact frequently inorder for each to achieve their goals. How-ever, there obviously must be a certain

degree of interaction in order to produce acollective bargaining agreement and to peri-odically resolve disputes. With regard to goalcompatibility, there is probably what couldbe described as a modest level. A union’s goalof obtaining higher wages and benefits for itsmembership would appear to be incompat-ible with an organization’s goal of controllingpersonnel costs. However, a union has avested interest in an organization’s stayingin business and, for that reason, may at timescompromise on issues. Conversely, an orga-nization may make concessions to a union attimes because its employees are needed toproduce a product or perform a service forcustomers.

COMMENT 11.4

THE PROS AND CONS OF INTERGROUP COMPETITION

AS WITH INDIVIDUALS, groups in an organization

often find themselves in competition. This is

often due to a scarcity of organizational re-

sources, but it may also be due to incompat-

ible goals between the groups. For example, ifone work group gains approval from upper

management, this often means that others do

not. Thus, groups in organizations are often in

a position of competing against each other

and, typically, are keenly aware of this com-

petition.

Is competition between groups a bad

thing? Not necessarily. When groups are com-peting, this may serve as an important source of

motivation to group members. They may work

harder, come up with more innovative ideas,

and behave more cooperatively toward each

other when they are competing with other

groups. These types of behaviors, if widely

prevalent, have the potential for enhancing

organizational effectiveness. There are, how-ever, some negative aspects of intergroup com-

petition. For example, members of competing

groups may communicate very little and

develop distorted perceptions of each other

(Forsyth, 2006). It is also not unusual for

groups in competition to actively try to thwart

each others’ efforts in order to ‘‘win.’’Can organizations foster some level of

intergroup competition, but not let it escalate

to destructive levels? This is a difficult question

to answer (and one that has not been addressed

empirically). The most likely way to accom-

plish this is by focusing competition primarily

on the tasks that groups are performing, rather

than on the personalities of individual groupmembers. People should be able to compete

without having to dislike each other. Compe-

tition can also be more positive when organi-

zations move away from a ‘‘winner takes all’’

mentality. When groups perform better by

competing with each other, there may in fact

be many ‘‘winners.’’ Ultimately, the question

really revolves around how organizations man-age intergroup competition.


PREDICTORS OF INTERGROUPINTERACTION PATTERNS

Although interactions between groups inorganizations are obviously influenced bythe goals being pursued, many other factorsinfluence the quality and frequency of inter-group interactions. In the intergroup litera-ture, those factors shown to have the mostconsistent effects are interdependence, orga-nizational culture, past history, and thesocial networks that evolve within organiza-tions. Each of these is reviewed in the follow-ing paragraphs.

Interdependence

A relatively consistent theme in reviews ofthe intergroup literature (see Alderfer &Smith, 1982, and Brett & Rognes, 1986) isthat a major factor influencing the interac-tions between groups is interdependence.The degree to which groups depend on eachother increases the frequency of interactionand, in some cases, heightens the potentialfor conflict. There are, however, differentforms of interdependence and these mayhave very different consequences for groupinteractions. According to Thomas (1976),pooled interdependence exists when groupsare relatively independent of each other buttheir combined outputs contribute to theorganization as a whole (see also Mathieu& Day, 1997). A good example of pooledinterdependence is the organizational struc-ture of automotive manufacturers such asGeneral Motors. Divisions such as Chevrolet,Buick, and Pontiac operate relativelyindependent of each other, yet they combineto determine the profitability of the corpo-ration as a whole. To some extent, it is amisnomer to call this arrangement interde-pendence because these divisions run fairlyautonomously. Keep in mind, however, that

even when there is pooled interdependence,groups are not completely independent ofeach other. For example, if the PontiacDivision of General Motors is performingpoorly, the other divisions may be negativelyimpacted.

A second common form of interdepend-ence in organizations is referred to as sequen-tial interdependence. In this form, the outputsof one group serve as the inputs to another.In many organizations, a good example ofsequential interdependence can be seenwhen new products are developed. Forinstance, the market research departmentmay conduct a study and determine thatconsumers desire a product that the organi-zation currently does not offer. This infor-mation may then be passed on to theresearch and development department,which is responsible for the actual develop-ment of this new product. Once the newproduct has been developed and tested, theinformation from research and developmentmay be passed on to the production engi-neers who will determine how this newproduct will be manufactured.

Reciprocal interdependence is commonlyfound among groups in organizations. Thisform of interdependence involves a series ofcontinuous mutual exchanges amonggroups. As one might guess, reciprocalinterdependence is highly complex andrequires the most interaction amonggroups. In many organizations, reciprocalinterdependence commonly exists betweengroups involved in manufacturing andmaintenance. A manufacturing departmentdepends on maintenance to repair equip-ment in order to minimize productiondelays and machine downtime. Conversely,a maintenance department relies heavily onthe cooperation of those on the manufac-turing side when repairs must be made andwhen preventative maintenance measures


are taken. Note that one of the hallmarks ofreciprocal interdependence is that there is arelatively equal status among the groups.Thus, with this form of interdependence,it is less likely that one group can gain theupper hand on another. The dependence ismutual.

What determines the type of interde-pendence that exists between any twogroups? One obvious determinant is thenature of the task(s) that different groupsare performing. Different tasks, by their verynature, require different levels and forms ofinteraction between groups. Generallyspeaking, groups that perform relativelycomplex tasks will often require higher levelsof interaction compared to groups perform-ing simpler tasks. Complex tasks may havegreater information requirements that canonly be satisfied through interaction withother groups.

Another important determinant of theform of interdependence between groups isorganizational structure. As will be shown inChapter 13, an organization can be struc-tured in a variety of ways, ranging from veryhierarchical and bureaucratic to structuresthat revolve primarily around importantorganizational projects. Often, when organi-zations are very formal and there are rigidboundaries around groups, the need anddesire for intergroup interaction may be de-creased. Hence, groups can operate fairlyautonomously. Conversely, when suchboundaries are more permeable, a great dealmore intergroup interaction may be re-quired, and, consequently, higher levels ofinterdependence will result.

A final potential determinant of interde-pendence is the authority system that pre-dominates within an organization. Meyerand Allen (1997) draw a distinction betweencontrol and commitment organizations. Incontrol organizations, the primary authority

mechanisms center on formal rules andmonitoring mechanisms. In such an environ-ment, interdependence may be discouragedbecause it makes control more difficult. Incommitment organizations, much moreauthority rests with individual employees,and there is less emphasis on formal rulesand monitoring. In this type of environment,communication may flow very freely amonggroups, and there is a great deal more poten-tial for interdependence.

Organizational Culture

Organizational culture essentially repre-sents the set of implicitly shared normsand values that predominate within a givenorganization (Schein, 1990). With respectto intergroup relations, some organizationsmay develop cultures that encourage andvalue positive intergroup interactions. Forexample, some universities have culturesthat encourage and facilitate interdepart-mental cooperation. Individuals in such aculture may engage in a great deal of inter-disciplinary work without even thinkingabout it, because it is so ingrained in theculture of the institution. Conversely, inother organizations, the culture may dis-courage most forms of intergroup interac-tion. For example, in some universities, theculture makes academic departments very‘‘compartmentalized.’’ Communication withother academic departments is infrequent,and they are viewed with suspicion. In thistype of culture, it is doubtful that produc-tive cross-departmental collaboration willoccur.

The effects of culture may be particularlysalient during the socialization of new em-ployees. The newcomers are informed veryquickly about group customs regardingexpected behaviors toward other groups. Ifa new employee mistakenly behaves the


wrong way in this regard, he or she is quickly‘‘brought into line’’ and reminded of theprevailing norms regarding interaction withother groups. Given the power of norms toshape behavior (e.g., Hackman, 1992), suchmessages will typically have a very importanteffect on the general quality of intergrouprelations.

Past History with IntergroupRelations

Another important factor that influences thequality of relations between pairs of groups ispast history. If the members of two groupshave generally had positive experiencesworking together, they are likely to approachfuture interactions in a positive manner.Conversely, if interactions have been conten-tious, groups will tend to approach futureinteractions with a great deal of apprehen-

sion and suspicion. Unfortunately, appre-hension and suspicion on the part of onegroup tend to evoke similar feelings in mem-bers of other groups (see Comment 11.5).

Readers may wonder whether the influ-ence of history tends to decay over time asnew members, who do not have the ‘‘baggage’’associated with past interactions, are broughtinto groups. As noted in Chapter 3, one of themajor dimensions of socialization into organi-zations is being ‘‘educated’’ about the past his-tory of the organization and, in many cases, ofthe specific work group that a new employeejoins (VanMaanen&Schein, 1979).Thisedu-cation often includes a detailed history ofevents that have involved other organizationalgroups. Thus, in many cases, new employeesare expected to ‘‘carry the baggage’’ associatedwith past conflicts and adjust their behavioraccordingly. Employees who resist such pres-sures may encounter very negative sanctions

COMMENT 11.5

THE IMPACT OF PAST HISTORY: A NEGLECTED TOPIC

BECAUSE A GREAT deal of research on intergroup

behavior has been carried out in laboratorysettings, there is very little empirical evidence

of the impact of past history on intergroup

relations. However, anyone who has joined an

established group in an organization would

likely attest to the powerful impact that past

history has on intergroup relations. As with

individuals, groups in organizations often

have histories of dealing with each other thatmay even transcend the tenure of individual

group members.

If groups have had a positive history of

working together to get things done, interac-

tions will be approached positively by the

members of both groups. Furthermore, these

positive interactions will likely be mutually

reinforcing and the positive relations between

groups will continue. In contrast, when two

groups have a negative history, it may be diffi-cult for them to work productively in the

future. Members of each group may be taught

to view each other with a great deal of suspicion

and mistrust. Unfortunately, when people are

treated with suspicion and mistrust, this

evokes similar feelings and a negative spiral

may begin.

Future research is needed on the mecha-nisms that are involved in communicating the

‘‘past history’’ of intergroup relations to new

group members. The greatest need, however, is

future research that examines how groups that

have negative histories can reverse this nega-

tive spiral and develop positive working rela-

tions.


from their more experienced coworkers, or,perhaps, may never become fully integratedinto the group.

Social Networks in Organizations

When all is said and done, much of thequality of intergroup interaction dependson the nature of the relationships (i.e., thesocial network) that develop between indi-vidual members of groups. To the extent thatindividual members of a group develop cor-dial relationships with individual membersof another group, this should enhance thequality of intergroup relationships. Evidenceof this principle can often be seen in the areaof international relations. For example, whenleaders of nations get along well and have

rewarding interpersonal relationships, thiscompatibility often positively influences dip-lomatic relations (see Comment 11.6).

Although not a great deal of research haslinked personal relationships within organi-zations to the quality of intergroup relations,there is some evidence that the two variablesare related. For example, LaBianca, Brass,and Gray (1998) examined the social net-works within a large health care facilityand found that negative relations amongindividual members of different groups hadan impact on the quality of intergroup rela-tions. When one member of a group had anegative encounter with a member of anothergroup, this negatively impacted the qualityof the relations between the two groups.Interestingly, positive social ties between

COMMENT 11.6

THE IMPACT OF PERSONAL RELATIONSHIPS ON INTERGROUP RELATIONS

PEOPLE OFTEN THINK of international relations

as being governed by highly formal me-

chanisms such as treaties and collective

alliances such as NATO. Such formal mech-

anisms do exist and do influence the trans-

actions between nations. We also know,however, that international relations are influ-

enced by highly personal relationships that

develop between national leaders.

A good example of this is the relationship

that ultimately developed between U.S. Pres-

ident George H. W. Bush and Soviet leader

Mikhail Gorbachev during the first term of

Bush’s presidency. Interestingly, the first meet-ing between these two men, which occurred in

1988 when Bush was president-elect, did not

go very well. Despite getting off on the wrong

foot initially, Bush and Gorbachev ultimately

developed a genuine friendship. Furthermore,

the friendship that developed between these

two world leaders may have been one of the

reasons that the United States supported Gor-

bachev, even when it became obvious that he

was losing his grip on power. Complementing

this strong relationship between Bush and

Gorbachev was an equal, if not stronger, rela-

tionship between U.S. Secretary of State JamesBaker and Soviet Foreign Minister Eduard

Shevardnadze.

Although events in most organizations are

not as dramatic as those that occurred between

the United States and the former Soviet Union,

it is true that relations between groups in any

setting are strongly impacted by the interper-

sonal relations that develop between groupleaders or members. When all is said and done,

intergroup relations, like most forms of social

behavior, are a ‘‘relationship business.’’

Source: G. H. W. Bush and B. Skowcroft. (1999). A world

transformed. New York: Knopf.


individuals from different groups had noimpact on intergroup relations. These find-ings suggest that when group members formimpressions of other groups, they utilize in-formation that is derived from the directexperience of their own group members.This may be much easier than approachingencounters with other group members objec-tively and, on that basis, forming an impres-sion of the other group. These findings alsosuggest that individuals weigh negative infor-mation more heavily than positive informa-tion in forming their impressions of othergroups. One reason for this may simply bethat negative information is more distinctiveand thus is more ‘‘attention getting’’ thanpositive information (Fiske, 2004; Taylor,1991). Another related reason may be thatmost people generally approach interperso-nal interactions in a positive manner. Thus,negative interactions run counter to our ex-pectations and thus receive more weight inthe impression-formation process.

INTERGROUP CONFLICT

Even a cursory perusal of the intergroupliterature reveals that much of it is focusedon conflict between groups. That is not to saythat all intergroup encounters are negative.However, as we will see in this section, manyconditions surrounding intergroup transac-tions in organizations greatly increase thepotential for relations between groups to behighly contentious. In this section, we focuson the conditions that lead to intergroupconflict, describe the potential effects ofintergroup conflict, and finally explore waysto diffuse or reduce it.

Causes of Intergroup Conflict

An impressive program of research in socialpsychology has shown that groups may be

inherently more prone to conflict and com-petition in comparison to individuals, afinding termed the interpersonal-intergroupdiscontinuity effect (Schopler & Insko, 1992).Studies demonstrating this effect typically usemixed-motive games, where either indi-viduals or groups have to choose betweencooperative moves where both parties obtaina small reward, versus competitive moveswhere one party obtains a large reward andthe other party receives nothing. In multiplestudies, groups playing the game are morelikely to make competitive choices than whenindividuals are playing against each other(Schopler et al., 2001; Wildschut, Insko, &Gaertner, 2002).

As one might imagine, conflict betweengroups in organizations may be caused by anumber of different factors. Thus, pinningdown the causes of conflict to a manageablenumber is no easy task. However, withinorganizations, a small number of factorsseem to contribute disproportionately tointergroup conflict. These include competi-tion for resources, goal incompatibility, timeincompatibility, and contentious influencetactics. Each is discussed briefly in the fol-lowing paragraphs.

One of the realities of life in most orga-nizations is that resources are scarce. Mostorganizations cannot provide groups withlavish support in the form of budgets, per-sonnel, or even physical space. As a result,groups are often in the position of competingvery hard with each other for these scarceresources. Like any competitive situation,there are often ‘‘winners’’ and ‘‘losers’’ inthe resource game. Furthermore, the factthat there are winners and losers in thecompetition for resources has the potentialto create very intense and long-standingconflicts between groups.

To a certain extent, the fact that groupsmust compete for resources is unavoidable.


If members of groups understand and acceptthis, conflicts may be kept from escalating todestructive levels. However, another waythat organizations can help to prevent suchcompetition from escalating to destructivelevels is to make sure that the resource-allocation process is fair (Colquitt et al.,2001). When resource-allocation proce-dures are fair, groups will not always bepleased with the resources allocated to them;however, at least they will be satisfied withthe procedures used to allocate the re-sources. Ways of doing this may includemaking sure that all groups have equal accessto the resources-allocation process, and thatpolitical considerations are minimized asmuch as possible. However, the reality isthat, in most organizations, resources willnever be in abundance, and competitionfor them will always be an issue that organi-zations must face.

A second common cause of intergroupconflict is goal incompatibility. At the mostgeneral level, goal incompatibility existswhen the goals of two or more groups arein direct opposition—that is, one groupaccomplishes its goals at the direct expenseof another group’s achieving its goals(Brown, Maras, Massar, Vivian, & Hewstone,2001). For example, in a democratic systemof government, the goal of each politicalparty is to assume control over variousbranches of government, but this can onlybe done by thwarting the same goal of otherpolitical parties. In an organizational set-ting, a common example of goal incom-patibility often exists between marketingand production. The primary goal of themarketing function is to satisfy customersby giving them what they want, when theywant it. The production function, how-ever, attempts to achieve efficiency andeconomies of scale—both of which aretough to do if products are tailored to indi-

vidual customers and delivered veryquickly.

One important point to consider aboutgoal incompatibility is the distinction be-tween real goal incompatibility versus per-ceived goal incompatibility (Brown et al.,2001). In many instances, such as the onepreviously described, there are real incom-patibilities in the goals of different groups. Inmany other cases, however, the extent ofincompatibility is much more a matter ofperception. For example, in colleges anduniversities, many people believe that edu-cating students and encouraging faculty re-search are completely incompatible goals. Incontrast, a good number of people in highereducation believe that educating studentsand encouraging faculty research mayactually be highly compatible goals. It isobviously beyond the scope of this chapterto decide which of these two positions iscorrect. The more important point is thatthis, along with many other forms of goalincompatibility, is largely a matter of percep-tions and opinions.

A third cause of intergroup conflict,which is actually closely related to goalincompatibility, is time incompatibility.Due to the pursuit of different goals andthe performance of different tasks, workgroups may work under very different timeframes. In a research and developmentdepartment, for example, current productdevelopment projects ultimately may notcome to fruition for a decade or longer.As a result, those who work in researchand development may take their time andaccumulate considerable data before mak-ing important decisions.

In contrast to research and develop-ment, those whose jobs revolve aroundthe production of products typically workunder a much different time frame. Produc-tion schedules are typically made in a very


short time span, and the major concern iswhat is happening at the moment ratherthan what may occur 10 years hence.Because of this short time frame, those inproduction may need information veryquickly and may not have the luxury oftaking the time needed to consider everypossible alternative before they make adecision.

The issue of time incompatibility is oftendealt with by making sure that there is littleinterdependence between groups that oper-ate under very different time frames. Thiscan be done in a number of ways, but it ismost typically accomplished through phys-ical separation. For example, the Xerox re-search and development center in Palo Alto,California, is credited with groundbreakingadvances in computer technology and islocated far from the company’s East Coastcorporate headquarters.

In some cases, it is not possible for orga-nizations to create such ‘‘buffer zones’’ be-tween groups with differing time frames. Forexample, a sales group must deal frequentlywith a market research group to gauge cus-tomers’ needs, or a production group mayneed crucial information from a research anddevelopment group to plan future produc-tion processes. Where such groups dependon each other, the potential for conflict isquite evident. Groups that operate on a veryshort time frame may see the other group asbeing indecisive and overanalyzing prob-lems. In contrast, groups operating under along-term perspective may see the othergroup as reckless and not caring about thequality of the information it is issuing.

Disputes like these are often difficult toresolve because they may have been going onfor many years and, in some cases, may evenbe part of an individual’s professional social-ization before entering the organization. Forexample, people trained in production man-

agement may be taught to be leery of ‘‘R & Dtypes.’’ Conversely, those likely to enter re-search and development fields may be social-ized to assume that those in areas such asmarketing do not appreciate the scientificvalue of their work.

A final factor that may lead to conflictbetween groups is the use of contentiousinfluence tactics by members of one grouptoward members of another. The term con-tentious influence tactics simply means tryingto influence another individual or other indi-viduals through the use of threats, demands,or other negative methods.

Recall from Chapter 5 that the use of con-tentious influence tactics is a general deter-minant of conflict in organizations, and thefact that they may lead to intergroup conflictmay come as no surprise.

One obvious reason that contentious in-fluence tactics heighten the potential forintergroup conflict is that they tend to inviteretaliation. When someone requests some-thing in a rude or abrasive manner, the nat-ural response of the person on the receivingend is to respond to the request in a rude orabrasive manner, or to simply refuse to com-ply (Falbe & Yukl, 1992). This can lead to anegative cycle of escalation and, ultimately,bitter conflict. In some cases, status differ-ences may prevent direct retaliation (e.g., asubordinate cannot directly retaliate when asupervisor is rude), but retaliation may occurnonetheless. One of the authors (SMJ) canrecall an instance in which a former facultycolleague was very demanding (and, attimes, rude) about his clerical work requests.The clerical person assigned to him acceptedsuch requests without being rude, but hisclerical requests typically went to ‘‘the bot-tom of the pile’’ when she was prioritizingher work.

Another reason that such influence tac-tics lead to intergroup conflict has to do with


the social networks that exist in groups(LaBianca et al., 1998). Employees in orga-nizations obviously do not work in a socialvacuum. Thus, when a person from onegroup attempts to influence a member ofanother group in a contentious manner, thisindividual will often relay news of a badexperience to members of his or her owngroup. As a result, some individuals maydevelop ‘‘bad reputations’’ before ever havingany direct contact with members of anothergroup. Unfortunately, when this is the case,the potential for conflict is greatly height-ened.

Consequences of Intergroup Conflict

Most organizations would prefer to have dif-ferent groups work well together because thealternative is unpleasant. However, to merelysay that intergroup conflict is ‘‘unpleasant’’begs the question: What really happenswithin a group when it is in conflict withanother group? Fortunately, a fair amount ofsocial psychological research has examinedthis very issue, although some of the researchhas not been conducted in formal organiza-tional settings (e.g., Sherif, 1966).

One thing we know very clearly is thatconflict changes group members’ perceptionsof each other (Roccas & Schwartz, 1993). Asone might expect, members of groups inconflict view each other in a negative man-ner. It is also clear from the intergroup liter-ature that members of groups in conflicttend to adopt an ‘‘us versus them’’ mentality(Castano, Yzerbyt, Paldini, & Sacchi, 2002;Depret & Fiske, 1999). Members of the othergroup are viewed as being very different, butvery similar to each other (e.g., ‘‘They’re allalike in accounting’’). Conversely, membersof one’s own group are viewed as positiveand as having a much greater degree ofindividuality than the ‘‘out-group.’’ Making

this ‘‘in-group’’ versus ‘‘out-group’’ distinc-tion may not actually reflect the complexityof situations, but it is cognitively mucheasier for the members of groups embroiledin conflict (Linville, Fischer, & Salovey,1989).

Another thing that typically happenswhen groups are in conflict is that, withingroups, members often become morecohesive—that is, when groups see a threatfrom an outside group, they may pull togeth-er and become highly cohesive (Castanoet al., 2002). A good example of this is theintense social bond that often develops be-tween soldiers who are involved in combat.It is not unusual for such individuals todevelop friendships that far outlast theirmilitary service (see Siebold, 2006). This‘‘common enemy’’ effect also occurs in orga-nizations; the members of a group bandtogether when faced with an outside threat.

Beyond the perceptual changes that havebeen discussed to this point, intergroup con-flict may also have tangible negative effects inorganizations. If groups are completelyindependent and have little need to interact,conflict will have little tangible impact otherthan perhaps a bit of discomfort here andthere. However, in most organizations, it isunlikely that any one group could becompletely independent of other groups.Thus, in most cases where there is inter-group conflict, the groups involved in suchconflict must interact. What, then, doesintergroup conflict do to the quality of inter-group interactions?

One point most of us know from experi-ence is that when two groups are embroiledin a conflict, the interaction between mem-bers of these groups is certainly strained andoften unpleasant. Because of this, members ofgroups in conflict may actively try to avoideach other, even if they are highly interde-pendent. The quality of work may then be


compromised. For example, it is not unusualfor students in universities to ask faculty frommore than one academic department to serveon thesis or dissertation committees. If differ-ent academic departments represented in acommittee are in conflict, this may greatlydecrease the communication between facultyand ultimately make it more difficult forstudents. Put differently, the student maybe caught in the middle of such a conflict.

The other outcome that may occur whengroups are in conflict is that members of eachgroup may go out of their way to make thingsas difficult as possible for the other group.Phone calls from members of the other groupmay be returned late, or requests for infor-mation may be ignored. The most troublingaspect of this is that contentious behaviortends to breed contentious behavior. As aresult, groups (like individuals) may getcaught up in a vicious cycle of conflict thatis difficult to break. Thus, the conflict maytake on a life of its own, often to the detri-ment of customers or the quality of the workin general.

Another troubling aspect of intergroupconflict is that groups often pass on theseconflicts to newcomers. That is, when a newmember enters a group, he or she is ‘‘edu-cated’’ about the history of the conflict and isexpected to adjust his or her behavioraccordingly. If the newcomer is seen frater-nizing with members of the ‘‘other’’ group, orexpresses positive views toward them, his orher own group members may react verysternly in order to bring the newcomer ‘‘backinto line.’’ In very extreme cases, a newcomerwho refuses to embrace the group’s conflictsand act accordingly may be forced to leavethe group or may do so voluntarily.

A final problem caused by intergroupconflict is that, when groups are in conflictwith each other, their goals focus primarilyinward and less toward the organization as a

whole (Sherif, 1966). Put differently, whathappens in the group may become moreimportant than what happens in the organi-zation as a whole. This may not be a problemif the goals of a group are perfectly alignedwith the goals of the organization. However,it could be a major problem if the goals of agroup are unrelated to or are in conflict withthe major goals of the organization. For ex-ample, a group that has a primary goal ofmaintaining intergroup harmony and stabil-ity may be a liability to an organization thatplaces a strong emphasis on quickly adaptingto change in the environment.

IMPROVING THE QUALITYOF INTERGROUP RELATIONS

Given the potentially dysfunctional effectsof intergroup conflict, organizations have avested interest in keeping it to a manageablelevel. In fact, organizations would ideally liketo create conditions that foster positive inter-group relations so that destructive conflictsdo not occur in the first place. In this section,we examine some of the most commonlyused methods for improving intergrouprelations. These include superordinate goals,negotiation, group member exchanges, inter-group team development, and, finally, de-creasing the interdependence among groups.

Superordinate Goals

One of the dangers of intergroup conflict isthat groups will focus inward and becomemore focused on their own goals than onthose of the organization as a whole. Thus,one way of improving relations betweengroups is to introduce superordinate goals—goals that both groups endorse and that oftenrequire cooperative intergroup behavior tobe achieved. The notion of superordinategoals comes from the Robbers Cave study,


which was conducted over 40 years ago byMuzafer and Carol Sherif and their col-leagues at the University of Oklahoma(Sherif, Harvey, White, Hood, & Sherif,1961). In this study, groups of boys believedthey were participating in a summer camp atRobbers Cave in Oklahoma, but they wereactually participating in a study of intergroupconflict. The researchers assigned these boysto two groups and created all the conditionsthat would likely foster intergroup conflict.For example, the two groups had differentnames (‘‘Eagles’’ and ‘‘Rattlers’’), were physi-cally separated, and competed against eachother in a variety of athletic contests. Thesemanipulations were largely successful at cre-ating a good deal of intergroup conflict.

After creating this state of intergroupconflict, the researchers then set about test-ing various methods of reducing it. One ap-proach, which was largely unsuccessful, wassimply to get the two groups together for ameal. The groups’ reaction to this interven-tion was rather messy: They engaged in afood fight! This finding has also been sup-ported in subsequent research (e.g., Brown,Condor, Matthews, Wade, & Williams, 1986)and thus suggests that simply getting groupstogether will not diffuse conflict.

The other approach employed by theresearchers, however, met with much greatersuccess. They staged a series of ‘‘incidents’’that required the cooperation of both groupsin order to solve a problem. For example,when both groups were being transported toa certain area of the camp, the bus they wereriding in broke down and required the com-bined efforts of both groups to fix it. Thegroups were also forced to cooperate to fix astaged plumbing problem in the cabin inwhich they both ate. In both incidents, thegroups were forced to work together to solvea problem that negatively impacted bothgroups.

What lessons can organizations learnabout managing intergroup conflict fromgroups of boys at a summer camp? One isthat simply getting conflicting groups tointeract with each other will probably notreduce the conflict. In fact, the findings fromthe Robbers Cave study, and subsequentresearch, suggest that such efforts mayactually intensify conflicts (although it’sunlikely, in organizations, that food fightswould break out!). The more practical impli-cation, for example, is that simply locatingconflicting groups in close physical proxim-ity will probably not successfully diffuseconflicts (Allport, 1954).

The other important lesson from thisstudy is that conflict can be reduced if mem-bers of groups must cooperate in order toachieve superordinate goals. Thus, for exam-ple, if two groups must cooperate to producea product or prepare an important report,this may go a long way toward diffusingconflict. Furthermore, what makes this veryfeasible in organizational settings is that op-portunities to introduce superordinate goalstypically abound. For example, an organiza-tion can always point out the fact that the‘‘superordinate goal’’ is the success of theorganization.

It is also possible in many organizationsto evoke the concept of superordinate goalsthrough the introduction of a ‘‘commonenemy’’ that the groups must band togetherto fight. In professional sports, coaches oftenbuild team unity by focusing on opponentsor, in some cases, on criticism of the team bythe news media. In organizations, this ‘‘com-mon enemy’’ approach could obviously targeta competitor, but could also focus on the pooreconomic environment, or perhaps on gov-ernment regulators. Organizations, however,need to be careful in using this approach.If an organization becomes too internally fo-cused and adopts an extreme ‘‘us versus


them’’ mentality, important external cues maybe missed, which may ultimately reduce theeffectiveness of the organization as a whole.

Negotiation

In any type of conflict, one of the problems isthat the parties involved avoid each otherand, as a result, do not communicate. Theirony, of course, is that conflicts can often beresolved if the parties sit down and commu-nicate rationally about their differences. Oneway of doing this is through some form ofnegotiation, typically between the leadersof the groups in conflict (Brett & Rognes,1986).

Negotiation involves a discussion of theissues causing conflict between the groupsand, ultimately, a resolution of those issuesin a manner that is acceptable to bothgroups. Why is negotiation a potential mech-anism for reducing intergroup conflict? First,negotiation facilitates communication. Nego-tiation is a bilateral influence technique; itinvolves a mutual exchange between parties(Kipnis, 1984). During this mutual ex-change, it is possible for misunderstandingsand misperceptions to be corrected.

Another reason that negotiation can besuccessful is that it is generally seen as a fairmethod of dispute resolution, especiallywhen all parties abide by the resolution(Fiske, 2004). Often, when groups are inconflict in organizations, the managers ofthe conflicting groups will make their casesto someone at a higher level, and a decisionwill be made that favors one group or theother. In another common way of handlingdisputes between groups, a higher-levelmanager serves as a mediator, much inthe way mediators help to resolve labordisputes. In the first method, the disputeis resolved but there is often a ‘‘winner’’ anda ‘‘loser.’’ Thus, members of one of the two

groups involved in the dispute may feel thatthey have been treated unfairly. In the sec-ond scenario, where a higher-level managermediates disputes, the problem that causedthe dispute may never be resolved. There isalso a chance, in this scenario, that ahigher-level manager may favor one groupover the other and effectively ‘‘force’’ a res-olution that is unfavorable to one of thegroups.

Although negotiation can be used effec-tively to resolve intergroup disputes, not allforms of negotiation are equally effective.Fischer and Ury (1981) coined the term prin-cipled negotiation to refer to a style of negotiat-ing whereby parties strive to achieveresolution to problems, rather than to stakeout positions. This ‘‘problem-solving’’ ap-proach to negotiating is likely to result increative resolutions to problems that mayin fact be quite palatable to both groups. Incontrast, when groups in conflict approach anegotiation with a mentality that states ‘‘Thisis my position, and I’m not going to compro-mise,’’ they are doing little to resolve the dis-pute and may even be heightening theconflict. Unfortunately, according to Brettand Rognes (1986), many managers of groupsare not trained in principled negotiation tech-niques and may lack more general negotiationskills. Thus, organizations that want toencourage managers to resolve intergroupconflict through negotiation should providemanagers with training in negotiation.

Member Exchanges

As we’ve seen, one of the things that hap-pens when groups are in conflict is thatgroup members distort their perceptions.Members of their own group are seen pos-itively, and members of the other group areseen negatively (Tajfel, 1981). One way tocombat such biased perceptions is for


groups to periodically engage in memberexchanges: For a stated period of time,members of the conflicting groups willswitch and reverse roles. This type of inter-vention may initially be uncomfortable forthose working in the ‘‘enemy camp,’’ but itcan also be very instrumental in reducingintergroup conflict. As an example, at auniversity where one of the authors (SMJ)once worked, there was a program wherebyfaculty could become ‘‘administrativeinterns’’ for 1 year. These individuals essen-tially worked as administrators for 1 year,although they were still responsible for asmall portion of their other faculty duties.The program was designed primarily forfaculty considering administrative careers,but it also helped to educate the faculty‘‘interns’’ about some of the challengesfaced by administrators. Although not amember-exchange program, per se, it didhelp the participants to increase theirunderstanding and therefore may havehad the effect of reducing conflict betweenfaculty and the university’s administration.

There are obviously some situations inwhich member exchanges cannot occur. Ifthe tasks performed by groups require vastlydifferent skills, or if large geographic differ-ences exist, exchanges may be impractical.There may also be some cases where conflictis so bad that the individual being exchangedmay have a very difficult time or may even beharassed. However, when they are possible,member exchanges can be an excellent wayof correcting misperceptions and, ultimately,diffusing conflict.

Intergroup Team Development

Team development (an approach discussedin more detail in Chapter 12) is an inter-vention that has been used to improve per-formance within groups. Team-development

approaches have also been designed to facil-itate or improve relations between differentgroups (Dyer, 1987; French & Bell, 1995).Team development will also be described ingreater detail in Chapter 16, but two specificintergroup team-development interventionsare relevant here.

One intervention, which was developedby Blake, Shephard, and Mouton (1964), is astructured way of highlighting possible areasof misperception and encouraging groups towork through their misperceptions. Figure11.6 provides a summary of the basic stepsinvolved in this intervention. The first step isto have the members of both groups generatetwo lists. The first list is designed to reflecthow each group views the other group. Thisincludes perceptions of the other group,what the other group does that ‘‘gets in ourway,’’ and so on. The second list reflects eachgroup’s prediction of what the other groupwill say about them; that is, each group istrying to anticipate what the other groupdislikes about them, or how they are per-ceived.

FIGURE 11.6Summary of the Steps Involved in a TypicalIntergroup Team Development Intervention

The two groups then reconvene todiscuss the lists generated in Step 1.

Each group meets separately andgenerates the following lists: • How they view the other group • What they think the other group will say about them

Development of action plans toimprove intergroup relations


After each group generates these two listsseparately, they reconvene to ‘‘comparenotes’’ on their perceptions. This can be aquite revealing exercise because it highlightsimportant issues and problems that are pre-venting the groups from working together,and it can serve as the basis for concreteactions designed to improve relations be-tween the groups. As an example, let’s saythat the members of one group see the othergroup as unresponsive and unwilling to pro-vide service to them, and the other group hassimilar perceptions. Discussing these per-ceptions may allow both groups to developa greater understanding of the pressuresfaced by the other group. It may also leadto concrete action plans designed to improvethe situation.

A variation of this intervention, theorganizational mirror technique, can also beused very effectively for correcting the mis-perceptions of conflicting groups. With thistechnique, one group forms a circle aroundthe perimeter of the other group. The group‘‘inside the fishbowl’’ is instructed to dis-cuss its self-perceptions and its perceptionsof the other group. During this time, thegroup observing is instructed not to inter-rupt, even if they feel that the perceptions ofthe group are not correct. The groups thenmay reverse roles; the group seeking feed-back may discuss what they have heard andmay possibly seek clarification. Followingthis step, the groups reconvene and ulti-mately generate a master list of action plansfor improving the relations between the twogroups.

As with the previous intervention, theorganizational mirror may provide uniqueopportunities for members of groups in con-flict to correct misperceptions and resolvemisunderstandings. The advantages of theorganizational mirror technique, however,are that it may be less time consuming and

more adaptable. A consultant may be able toemploy this technique at several points dur-ing a day-long intergroup team-developmentintervention. A potential disadvantage of thefishbowl technique is that it is potentiallyvery confrontational. Therefore, this methodshould only be used in conjunction with aconsultant who is well trained in group facil-itation and conflict management. The tech-nique probably should not be used whengroups are embroiled in highly intense, per-sonalized conflicts.

Reducing the Need for IntergroupInteraction

Despite well-intentioned efforts to improveintergroup relations, there may be timeswhen groups simply cannot work well to-gether. This may be due to irreconcilablepersonality conflicts or the nature of thework that groups perform being too differ-ent. One way of dealing with intergroupconflict is creating conditions in which twogroups have little or no need to interact. Thiscan generally be done in one of two ways.The first approach is to change the structureof an organization, or the flow of work, sothat the level of interdependence betweentwo groups is greatly reduced. A way ofdoing this is to create a ‘‘coordinating group’’to manage the interactions between groups.In this type of structure, each group needs tocommunicate only with the coordinatinggroup. This eliminates the need to interactwith the other group. To decrease interde-pendence, organizations frequently outsourceproducts or services that are currently beingperformed internally. The auto industry hasrelied heavily on outsourcing for many of theparts that they used to produce internally.This has proved to be cost effective andundoubtedly has reduced the level of inter-dependence among internal groups.


A second general way to reduce interde-pendence among groups, particularly insequential production systems, is to buildsome level of ‘‘slack’’ into the system. If onegroup supplies parts to a second group, forexample, maintaining a reserve inventory ofthese parts will decrease the interdependenceof the groups. Creating slack resources of thistype results in somewhat less sequentialinterdependence. This decrease in interde-pendence may ultimately decrease tensionsbetween the two groups, because one is notso dependent on the other.

The Resource-Allocation Process

A final way that organizations can potentiallyreduce conflict between groups is throughthe resource-allocation process. As was notedearlier in the chapter, a very common causeof intergroup conflict is that groups are oftenin competition for scarce organizational reso-urces. Thus, one way to reduce conflict is toreexamine the resource-allocation process tomake sure that groups have the resourcesthey need to be effective.

At first glance, the suggestion that organi-zations ‘‘give groups more resources’’ mayseem to be a rather simplistic and naıve wayof decreasing intergroup conflict. After all, thevast majority of organizations operate underconditions whereby resources are scarce. Thismay be true, but organizations do have con-trol over where those scarce resources areallocated and the process by whichresource-allocation decisions are made. Thus,organizations can choose to allocate resourcesin a manner that is commensurate with groupneeds, or they can do so on the basis ofpolitical gamesmanship and ‘‘back-roomdealing.’’ Similarly, organizations can makethe choice to allocate resources in a fair,rational manner, or they can do so in a highlysecretive, clandestine fashion.

When organizations allocate resources ina fair, rational, and open manner, this will notnecessarily eliminate the problem of resourcescarcity. However, based on what is knownabout procedural justice (e.g., Folger &Cropanzano, 1998), it is certainly possiblethat even when groups come up short onresources, they will still see the resource-allocation process as fair. Another moregeneral benefit of fair resource-allocation pro-cedures is that these may simply foster a morepositive organizational climate. When peoplefeel they are being treated fairly by the orga-nization, this will tend to foster a climate ofopenness and trust as opposed to mistrustand paranoia. Such a climate certainly doesnot guarantee positive intergroup relations,but it certainly makes them more likely.

CHAPTER SUMMARY

The purpose of this chapter was to providean overview of group dynamics of behaviorby drawing on contributions from bothsocial and organizational psychology. Webegan by examining the defining character-istics of a group. Despite the numerous def-initions proposed over the years, mostmaintain that the key features are interde-pendence, interaction, perception of being agroup, and some commonality of purposeamong group members.

Group structure represents a number ofdimensions that can be described. The mostcommon dimensions of group structure in-clude roles, norms, values, status differen-tials, and communication patterns. Groupmembers’ behaviors come to be differenti-ated; normative behavior patterns develop;behaviors that are consistent with certainvalues are emphasized; status hierarchiesemerge; and, finally, consistent modes ofcommunication are adopted.

Chapter Summary 371


DONELSON FORSYTH AND THE DYNAMICS OF GROUPS

My interest in groups and their processes can

be traced—as it should be—to the many influ-

ential groups that I have joined over the years.

Some of these groups, such as high school

cliques that sharply differentiated between

those thought to have social value and thoseconsidered unworthy, teams that failed mis-

erably or succeeded unexpectedly, peer

groups that contributed to my delinquency

more than my rectitude, and work groups that

rarely did any work, piqued my interest in

social, rather than purely psychological,

causes of behavior. Growing up in an Ameri-

can society that stressed the values of indi-vidualism and self-reliance over team-work

and interdependence, I wondered how people

manage to combine their individual motiva-

tions and actions in the pursuit of shared

goals.

I continue to be intrigued by all aspects of

groups and interpersonal relations: why net-

works of relationships form; how groups sus-tain their members; when people conform to

the standards set by others; why and when

groups perform to their full potential or fall

short of their goals and fail; how leaders can

motivate those around them. When I examine

any one of these topics, I often find that group-

level processes offer members the means to

reach goals that would have eluded them as

individuals. Yet, my own experiences in

groups—and the research data as well—

remind us that many groups fail to reach theirgoals; for every high performing team we find

ones that blunder through their tasks and

never search for ways to improve. As I’m

obsessed with avoiding failure on a personal

level, and have therefore found many ways to

bolster my sense of self-worth each time I do

misstep, I continue to be intrigued by the

intricacies of responsibility allocation ingroups. In a number of studies I have examined

how people working in groups react to group-

level setbacks and failures as well as achieve-

ments and successes. That research suggests

that groups provide members with the perfect

habitat to mask their own inadequacies, for

rarely is a single member of a group held

responsible for any group’s failure. Whereasthe lone worker who performs poorly will be

easily evaluated as inadequate, when sub-

merged in a group the loafing, distracted,

or the incompetent member may escape

detection—except when appraisals are pro-

vided by co-workers whose own outcomes

are undermined by an inferior other. More-

over, because the relative contributions of eachperson to a group activity are difficult to pin-

point precisely, the group microcosm provides

members with the opportunity to take more

credit for their group’s work than they actually

deserve. Hence, when I total the responsibility

claims of each member of a group, I usually

find that members feel they are more respon-

sible than the average other group member—particularly if they are the leader of the group.

These findings suggest that groups in orga-

nizational settings, to maximize their produc-

tivity, should monitor their responsibility

allocation mechanisms, but they also suggest

that the natural tendency for individuals to use

the group context to maximize or minimize


A number of models have been devel-oped to describe the manner in whichgroups develop over time. The best knownof these is Tuckman’s (1965) model, whichis based on the issues that must be dealtwith at various points in the life of a group.Moreland and Levine’s (1982) model ap-proaches group development from the per-spective of the socialization of new groupmembers. Gersick’s (1988) PunctuatedEquilibrium Model focuses specifically ontask groups. Each of these provides some-what different and unique insights intogroup development.

After discussing within-group dynamicsand behavior, we then examined a relativelyneglected topic in the group literature: inter-group relations. In most organizationsgroups must work together collaborativelyto accomplish many important tasks. Wediscussed the various types of interactionsthat may occur between groups, rangingfrom avoidance to collaboration dependingon the compatibility of goals and the level ofinterdependence that exists between twogroups. Other factors that determine thenature of intergroup interactions includethe type of interdependence, task-relateddemands, the culture of the organization,past events that have transpired betweengroups, and the nature of the individualsthat comprise each of the groups.

Given that different groups in organiza-tions may have different goals, operate ondifferent time frames, and employ differenttypes of people, the potential for intergroup

conflict is often high. Thus, it is not surpris-ing that a great deal of the intergroup liter-ature has focused on conflict. Conflictbetween groups—due to biases in percep-tions and a tendency to adopt an ‘‘us versusthem’’ mentality—clearly has a negativeimpact on relations between groups. Whilenegative in and of themselves, biases mayultimately have a negative impact on organi-zational effectiveness. Thus, organizationshave a vested interest in keeping intergroupconflict at manageable levels.

From the social psychological literature,a method that has been shown to reduceintergroup conflict is the adoption of super-ordinate goals. These are goals that transcendthe interests of individual groups and oftenrequire collaborative efforts in order toachieve them. Among other methods ofreducing intergroup conflict, the chapterdiscussed member exchanges, intergroupteam-development interventions, decreasinginterdependence among groups, and im-provement of the resource-allocation pro-cess. Given the realities of organizationallife, it is doubtful that intergroup conflictcan ever be completely eliminated. However,it is certainly possible for organizations tokeep it to manageable levels.


Brett, J. M., Goldberg, S. B., & Ury, W. L.(1990). Designing systems for resolving

their personal responsibility may be func-

tional: Some undertakings may seem so risky

that people would not undertake them if they

knew they would bear the brunt of the blame

should they fail, and individuals may be less

motivated to expend effort in pursuit of shared

goals if they could not take a bit of extra credit

when the group succeeds.

Donelson R. Forsyth

Jepson School of Leadership Studies

University of Richmond


disputes in organizations. American Psy-chologist, 45, 162--170.

Forsyth, D. R. (2006). Group dynamics (4thEd.). Belmont, CA: Wadsworth/Thompson.

Johnson, D. W., & Johnson, F. P. (2000).Joining together: Group theory and groupskills (7th ed.). Boston: Allyn and Bacon.

Levine, J. M., & Moreland, R. L. (1998).Small groups. In D. T. Gilbert, S. T. Fiske,

& G. Lindzey (Eds.), The handbook of socialpsychology (Vol. 2, pp. 415--469). NewYork: McGraw-Hill.

Nelson, R. E. (1989). The strength of str-ong ties: Social networks and intergroupconflict in organizations. Academy of Man-agement Journal, 32, 377--401.


Chapter Twelve

TeamEffectivenessI

n Chapter 11, we provided an over-view of research on group and inter-group behavior, focusing primarily onhow groups come to exist anddevelop and how they get along with

each other. These are certainly importantprocesses in organizations. Thus, the fieldof organizational psychology has benefitedgreatly from the work of social psychologistswho were the early pioneers of group dy-namics research. However, two characteris-tics of early social psychological groupdynamics research place limitations on itsusefulness to organizational psychologists.First, very little social psychological groupdynamics research directly examined thedeterminants of group effectiveness. Rather,the dependent variables examined in thisresearch have typically been attributes ofindividuals rather than of performancegroups. Research that directly examines teameffectiveness is inherently more useful toorganizational psychologists.

Second, the overwhelming majority ofsocial psychological group studies have beenconducted in laboratory settings. As waspointed out in Chapter 2, organizationalphenomena are often replicated quite wellin laboratory settings; thus, generalizabilityis often robust (Locke, 1986). Team effec-tiveness, however, may be one organiza-tional variable for which generalizing fromlaboratory to field settings may be problem-atic because the environmental context inwhich a team performs is an important keyto understanding team effectiveness. In lab-oratory settings, groups are formed ran-domly, and they perform tasks in relativelystandardized settings. In fact, proper exper-

imental methodology dictates that the envi-ronmental conditions under which differentgroups perform must be as similar as possi-ble (Cook & Campbell, 1979).

In real organizations, however, there areimportant differences in the environmentalcontext in which various teams perform. Forexample, even within the same organization,there are important differences among workteams in staffing levels, the manner in whichrewards are distributed, the way leaders setgoals with team members, and the wayteams interface with other teams. Also, inreal organizations, members often haveknown each other prior to their presentteam assignments, and those prior relation-ships may promote collegiality or lead to

375

conflict. These contextual factors are diffi-cult to simulate in laboratory research, yetthere is little denying that they influence theeffectiveness of teams.

Due to the importance of organizationalcontext, what has been described as an orga-nizational psychology team literature hasdeveloped relatively recently. In fact, manyhave come to the conclusion that when itcomes to the study of team effectiveness,‘‘the torch has been passed’’ from social toorganizational psychology. Levine andMoreland (1990)—who, incidentally, areboth social psychologists—make the pointby stating: ‘‘Despite all the excellent researchon small groups within social psychology,that discipline has already lost its domi-nance in this field. The torch has effectivelybeen passed to (or, more accurately, pickedup by) colleagues in other disciplines, par-ticularly organizational psychology. Theyhave no doubts about the importance ofsmall groups and are often in the forefrontof group research. So, rather than lamentingthe decline of interest in group, we shouldall be celebrating the resurgence, albeit in adifferent locale’’ (p. 620). This point has alsobeen made by other researchers (see Ilgen,Hollenbeck, Johnson, & Jundt, 2005).

The purpose of this chapter is to discussthe voluminous amount of recent work, pri-marily by organizational psychologists, onthe factors that contribute to team effective-ness. We purposefully use the term teamrather than group because so much of thework done by organizational psychologistsuses teams within organizations performingtheir normal work or being placed in realisticsimulations to provide a more controlledassessment of team dynamics and perfor-mance. Kozlowski and Bell (2003) providea detailed definition of what constitutes ateam. They define work teams as ‘‘composedof two or more individuals who (a) exist to

perform organizationally relevant tasks, (b)share one or more common goals, (c) interactsocially, (d) exhibit task interdependencies,(e) maintain and manage boundaries, and (f)are embedded in an organizational contextthat sets boundaries, constrains the team,and influences exchanges with other unitsin the broader entity’’ (p. 334).

We begin the chapter by examining abasic question: What constitutes team effec-tiveness? The focus then shifts to examiningseveral influential models of team effective-ness. These models have helped to guidemuch of the organizational research on teameffectiveness and have served as a foundationfor many organizational efforts to improvethe performance of teams.

Across these models, several variablesseem to pop up repeatedly as determinantsof team effectiveness. These include teamcomposition, task design, organizationalresources, organizational rewards, teamgoals, and team processes (e.g., sharedmental models, transactive memory). Eachof these will be examined individually, andsome of the most common methods orga-nizations use to improve the performanceof teams will be described. The chapterconcludes with a brief discussion of therole that teams are likely to play in organi-zations in the future.

DEFINING TEAMEFFECTIVENESS

To say that teams now represent an impor-tant part of organizational life is certainly anunderstatement. Nearly all organizations usesmall teams to accomplish at least sometasks, and it has become increasingly popu-lar for small teams to be the basic founda-tions upon which organizations are built(Guzzo & Shea, 1992; Kozlowski & Bell,2003; Peters & Waterman, 1982).

376 Team Effectiveness

Despite the importance of team effec-tiveness, it is not necessarily easy to definewhat is meant by an effective team. Steiner(1972) proposed one of the earliest theo-retical propositions bearing on team effec-tiveness:

Actual productivity¼ Potential productivity� Process losses

The potential productivity of a team rep-resents the highest level of performance thatis attainable by a team. Consider the follow-ing example. If each of the five starting play-ers on a basketball team is capable of scoring20 points per game, those players collectivelyshould be able to score at least 100 points pergame. The term process losses represents less-than-optimal ways of combining the inputsof team members into a team product. Proc-ess losses generally occur because of a lack ofcoordination among team members, orbecause the motivation of individuals maychange when they are performing in a teamsetting (Latane, Williams, & Harkins, 1979).The team previously described may not reachits scoring potential because the styles of theindividual players do not mesh well, orbecause individual players may not feel agreat deal of personal responsibility for theteam’s performance and, as a result, mayreduce their effort.

Steiner’s (1972) basic model is certainlyuseful in helping us to understand, in a verygeneral sense, what determines team perfor-mance, but it also has some serious limita-tions. The most serious flaw is that thismodel doesn’t specify which aspect(s) ofthe team, or the organizational context, canbe used to enhance the coordination of teammembers or to prevent process losses. Theother problem with this model is that it isbased on the rather naıve assumption thatorganizational goals and team goals are per-fectly aligned (Hackman, 1992). For exam-

ple, if a team’s goal is to produce a high-quality product, and the organization’s goalis for the team to produce a large quantity,the team may appear to suffer from processloss, but in fact it may be quite successful;the team is simply succeeding on its ownterms.

The definition of team effectiveness thathas been used most often by researchers wasput forth by Hackman (1987) in a compre-hensive review of the work-team design lit-erature. According to Hackman, teameffectiveness is a multidimensional constructconsisting of three interrelated dimensions.The first of these dimensions is related to theoutput of the team, the second has to do withthe long-term viability of the team as a per-forming unit, and the third has to do with theimpact of the team experience on individualteam members. Each of these will be dis-cussed in the following paragraphs.

The vast majority of teams in organiza-tions are formed to fulfill some specificcharge or purpose; that is, they are formedto do something. For example, a top manage-ment team may be formed to create a long-range strategic plan, a surgical team may beassembled to perform a delicate cardiacbypass operation, and a rescue team maybe formed to evacuate hurricane victimsfrom their homes. Thus, one way to measurethe effectiveness of a team is to judge wheth-er its output is satisfactory. As Hackman(1987) states, for a team to be judged suc-cessful, ‘‘The productive output of the workteam should meet or exceed the performancestandards of the people who receive and/orreview the output’’ (p. 323). In the previousexamples, the top management team wouldhardly be considered a success if it failed toproduce a viable strategic plan, the surgicalteam would be unsuccessful if the patient’sarteries remain blocked, and the rescue teamwould be unsuccessful if the hurricane

Defining Team Effectiveness 377

victims are not evacuated. Task accomplish-ment is thus an important component ofteam effectiveness.

Vince Lombardi, the legendary footballcoach of the Green Bay Packers, told histeam: ‘‘Winning isn’t everything, it’s the onlything.’’ In many organizations, a variant ofthis attitude prevails in that the quality of ateam’s output is seen as the only indicator ofits effectiveness. According to Hackman(1987), a second indication of the effective-ness of a team is whether the social processesused in performing the task maintain orenhance the capability of the team membersto work together in the future. In some cases,teams accomplish their assigned tasks, butthey do so in such a contentious manner thatthey are incapable of working together in thefuture. According to Hackman, a team wouldhardly be considered effective if it essentiallyburns itself up while performing its assignedtask, even if that task is performed success-fully. In some cases, a team comes togetheronly once to perform a task (e.g., a searchcommittee is selecting a new college presi-dent). In most organizations, however, this ismore the exception than the rule.

Working as part of a team can be anextremely rewarding experience. Often, agreat deal of satisfaction goes with collec-tive accomplishment, and the socialrelationships that often develop in teamscan be extremely rewarding. Unfortu-nately, there is also a downside to workingin teams. Depending on others to get thingsdone can be extremely frustrating, high-performing individuals may not receivetheir full share of the credit for a team’sperformance, and relationships with otherteam members may go sour. According toHackman (1987), if the experiences ofteam members are largely negative andfrustrating, this is an indication that theteam is unsuccessful. Thus, the third

dimension of team effectiveness is teammember satisfaction. This is really relatedto the second point about viability, butgoes somewhat further. Specifically, thosewho are largely negative about teamworkmay be loath to work in any team withinthe organization in the future. When ateam performs its task well, whilecompletely ignoring the needs of teammembers, it can hardly be consideredeffective (see Comment 12.1).

MODELS OF TEAMEFFECTIVENESS

As with many of the topics covered in thisbook, there have been a number of attemptsto model the factors that contribute to teameffectiveness. Like any theoretical models, themodels of team effectiveness are incomplete;that is, they do not include every possiblefactor that contributes to team effectiveness.They are useful, however, in highlighting thegeneral factors that differentiate successfuland unsuccessful teams. They also help usto understand the processes by which thesegeneral factors work together to influenceteam effectiveness. This aids in organizingresearch findings and may also guide organi-zational efforts toward improving the perfor-mance of teams. In this section, we examinefive of the most influential models in theteam-effectiveness literature.

McGrath’s (1964) Model

McGrath (1964) proposed that team effec-tiveness is determined by a basic input–process–output sequence; that is, certaininputs lead to differences in team process,which eventually lead to differences in teamoutput. This basic idea is expanded in Fig-ure 12.1 as a model of team effectiveness.Notice that McGrath included Individual-


Level Factors, Team-Level Factors, andEnvironment-Level Factors in the input col-umn. Individual-Level Factors include char-acteristics of team members that may have animpact on team effectiveness, such as theirlevel and mix of skills, attitudes, and person-ality characteristics.

Team-Level Factors are essentially struc-tural properties of teams themselves. Thesemay include elements of team structure (e.g.,roles, authority structure, norms), the levelof cohesiveness within the team, and thenumber of individuals within the team.McGrath (1964) emphasized that all teams

are not created equal, and some are designedin a more optimal fashion than others. Thus,some teams are more ‘‘primed’’ for successthan others.

Environment-Level Factors representaspects of the organizational context underwhich the team works. Perhaps the mostimportant of these is the task that the teamis performing. The nature of the task willdictate, to a large degree, the most appropri-ate strategies to be used by the team. The taskmay also be important for motivational rea-sons. One of the problems inherent in team-work is social loafing, where team members

COMMENT 12.1

WINNING ISN’T EVERYTHING: A BROADER VIEW OF GROUP

EFFECTIVENESS

DESCRIBING HACKMAN’S (1987) multidimen-

sional view of group effectiveness frequently

generates a great deal of discussion and, attimes, vigorous debate. Typically, these dis-

cussions center around one question: ‘‘If a

group performs its task well, does it really

matter if people can work together in the

future, or whether they have enjoyed working

as part of the group?’’ After all, most organi-

zations exist primarily to make a profit, not to

provide psychological gratification to theiremployees.

This point of view is certainly compelling,

but it is also rather shortsighted, and it ignores

some of the realities of organizational life.

There may be crisis situations in which indi-

viduals must come together, perform very

quickly, and disband immediately after their

task is finished. For example, an internationalcrisis may require that a number of national

security officials come together to decide

whether military intervention is warranted.

In such a case, one could argue that making

the correct decision is far more important than

the satisfaction of the group members or the

viability of the group.

In most organizations, however, groupmembership is more stable; thus, long-term

viability and member satisfaction are impor-

tant issues. Groups that repeatedly accomplish

their primary tasks, but create a great deal of

dissatisfaction among their members, often

cannot survive over the long haul. Further-

more, when employees have bad experiences

in a work group, these incidents will oftencarry over to other situations. The employees

may even be reluctant to work in groups in the

future. Thus, a group leader who is concerned

about the psychological needs of the group

members may appear to be compromising task

performance, at least in the short term. Ulti-

mately, though, groups under this type of

direction are often much more successful thanthose governed by a ‘‘win-at-all-costs’’ philos-

ophy.

Source: J. R. Hackman. (1987). The design of work teams.

In J. W. Lorsch (Ed.), Handbook of organizational behavior

(pp. 315–342). Englewood Cliffs, NJ: Prentice-Hall.


do not exert as much effort in a group as theywould alone (Latane et al., 1979). Socialloafing is probably less likely to occur whena team is performing a complex and stimu-lating task. More will be said about taskcharacteristics later in the chapter.

Two other important aspects of the envi-ronment are the Reward Structure and thelevel of Environmental Stress under whichthe team performs. Rewards are important inshaping individual performance, so it makessense that they would also be a determinantof team performance as well. The mostimportant consideration is whether the per-formance of the team is rewarded, ratherthan simply the performance of individualteam members. The level of environmental

stress may be determined by a number offactors, such as the criticality of work theteam is performing and the time pressureunder which it must be performed. A highlevel of environmental stress may lead toproblems in a team’s decision-making proc-esses (e.g., Janis, 1982) or to conflicts sur-rounding the distribution of authority withinthe team (e.g., Driskell & Salas, 1991). Fur-thermore, individuals become less effectiveteam members when under environmentalstressors such as heavy workload or timepressure (Driskell, Salas, & Johnston, 1999).

Given all of these input factors, the mod-el proposes that they combine to determineTeam Interaction Process, which representsthe manner in which a team performs its

FIGURE 12.1McGrath’s Model of Group Effectiveness

Individual-LevelFactors

• Pattern of Member Skills• Attitudes• Personality Characteristics

Group-LevelFactors

• Structure• Level of “Cohesiveness”• Group Size

GroupInteraction

Process

Process OutputInput

Time 2Time 1

Environment-LevelFactors

• Group Task Characteristics• Reward Structure• Level of Environmental Stress

Performance Outcomes

• Performance Quality• Speed to Solution• Number of Errors

Other Outcomes

• Member Satisfaction• Group “Cohesiveness”• Attitude Change• Sociometric Structure

Adapted from: J. E. McGrath. (1964). Social psychology: A brief introduction. New York: Holt. Reprinted by permission of the

author.


task. Obviously, this may cover a good dealof territory, but the most crucial aspects ofthe team process are likely to be things suchas the performance strategies adopted by theteam, the level of interpersonal harmonywithin the team, and the ability of the teamto have a shared understanding of the taskand coordinate the necessary responses(Salas, Sims, & Burke, 2005). Recent researchon team effectiveness has emphasized theimportance of team processes for team effec-tiveness. These processes will be detailed aswe describe other models and determinantsof team effectiveness.

McGrath’s model then proposes thatTeam Interaction Process will have a directeffect on the output of teams. The Output of ateam can be viewed in two ways. First andforemost, a team’s output can be viewed interms of the level of performance. Thiswould include factors such as the judgedquality of the team’s output, the time it takesfor the team to make a decision or develop asolution to a problem, or, possibly, the num-ber of errors that are made in performing thetask. The category titled Other Outcomes rep-resents social aspects of the team experience,such as the satisfaction of team members, thelevel of cohesiveness within the team afterthe task is performed, the attitudes of teammembers, and, finally, the pattern of rela-tionships following the team’s performance.Although these do not represent team per-formance per se (Hackman [1987] wouldlater argue that they are part of it), they arestill important outcomes.

It is difficult to overemphasize the impactthat McGrath’s model had on team research.Ilgen et al., (2005) recently noted that mostteam research has been either explicitly orimplicitly guided by the input–process–outcome model. The likely reason for thesuccess of the model is its breadth and sim-plicity. Most of the factors influencing teams

can be categorized either as Inputs, Proc-esses, or Outcomes. However, other organi-zational researchers have attempted to add tothese basic concepts, including additionalcomponents in the prediction of team effec-tiveness.

Gladstein’s (1984) Model

Gladstein (1984) proposed a model of teameffectiveness that, in many ways, is quitesimilar to McGrath’s (1964). As can be seenin Figure 12.2, this model follows the samegeneral input–process–output sequence.The inputs in this model are also very similarto those proposed by McGrath. Notice that,at the team level, these include character-istics of individual team members as well aselements of team structure. At the organiza-tional level, however, the model differssomewhat from McGrath’s. The critical fac-tors proposed by Gladstein include the re-sources available to teams (training andtechnical consultation are available to teamsand to the markets of customers they serve)as well as aspects of organizational structure(rewards and supervisory control).

Gladstein (1984) proposed that all ofthese inputs contribute directly to the teamprocess. The critical elements of the teamprocess include communication, level ofsupport, handling of conflict, discussion ofperformance strategies, weighing of indi-viduals’ outputs, and boundary management,or the manner in which the team interfaceswith other units both inside and outside ofthe organization. Team process is then pro-posed to lead to team effectiveness (indicatedby performance and satisfaction), as it doesin McGrath’s model.

This model differs from McGrath’s in twoimportant ways. First, Gladstein proposesthat inputs have a direct impact on teameffectiveness, in addition to the effect that


is mediated by team process. For example, ateam with extraordinary levels of skill may beeffective, regardless of the quality of teamprocess. A second important difference isthat Gladstein proposed that a team’s taskmoderates the relationship between teamprocess and team effectiveness. For example,a very unstructured, freewheeling style ofinteraction may be useful for a team creatinga new product, but inappropriate for a teamthat is merely required to follow directions.This suggests that some forms of team proc-

ess will be more or less effective, dependingon the complexity, uncertainty, and interde-pendence present in the team’s task. Recallthat McGrath (1964) proposed that thenature of the task represented an input thatcontributed directly to team process and,subsequently, to effectiveness.

When Gladstein (1984) tested her modelof team effectiveness, she used a sample ofsales teams from the marketing division of anorganization in the communications indus-try and found mixed support. The model

FIGURE 12.2Gladstein’s Model of Group Effectiveness

Group Composition• Adequate Skills• Heterogeneity• Organizational Tenure• Job Tenure

Group Structure• Role and Goal Clarity• Specific Work Norms• Task Control• Size• Formal Leadership

Process

Indicates a moderated relationship

OutputInput

Organizational Level

Group Level

Resources Available

•Training and Technical•Consulting•Markets Served

OrganizationalStructure

• Rewards for Group Performance• Supervisory Control

Group Task• Task Complexity• Environmental Uncertainty• Interdependence

Group ProcessGroup Effectiveness

• Performance• Satisfaction

• Open Communication• Supportiveness• Conflict• Discussion of Strategy• Weighting Individual Inputs• Boundary Management

Adapted from: D. Gladstein. (1984). Groups in context: A model of task group effectiveness. Administrative Science Quarterly, 29,

499–517. Copyright # 1984, Cornell University Press. Reprinted by permission.


received the strongest support for the pre-diction of team members’ perceptions ofeffectiveness, but support was much weakerwhen predicting actual sales revenue. Thissuggests that perhaps the model is morereflective of individuals’ ‘‘implicit theories’’of the determinants of team effectivenessthan of actual team effectiveness. Anotherimportant finding from this study was thattask characteristics did not moderate therelations between team process and effective-ness (perceived or sales revenue). This find-ing is difficult to interpret, given that thestudy was cross-sectional and there was nota great deal of variation in the nature of thetasks that these teams performed.

Hackman’s (1987) Model

Building again on the general input–process–output framework, Hackman (1987) devel-oped what he termed a normative model ofteam effectiveness. The term normative wasused because Hackman’s intent was clearly todevelop a model that explicitly revealed themost important performance levers that organi-zations could use to enhance team effective-ness. Stated differently, Hackman’s purpose indeveloping this model was more to serve as aguide to improving team performance ratherthan merely facilitating an understanding ofwhy a team fails.

As can be seen in Figure 12.3, the two‘‘input’’ factors in the model are organiza-tional context and team design. Underorganizational context, the most importantfactor proposed by Hackman is the rewardsystem under which teams work. Oneimportant aspect of an organizationalreward system, with respect to teams, iswhether challenging and specific perfor-mance objectives exist. The failure to pro-vide performance objectives often plaguesindividual-level reward systems but is partic-

ularly problematic with teams simplybecause most organizational performance–management systems are designed with indi-viduals, not teams, in mind.

A second important aspect of organiza-tional reward systems is that they must bedesigned so that teams receive positiveconsequences for excellent performance.The reinforcement value of positive conse-quences on individual behavior is wellknown (e.g., Luthans & Kreitner, 1985;Weiss, 1990). According to Hackman(1987), this same principle applies toteams, even though individual team mem-bers must ultimately see the connectionbetween the performance of the team andthe potential rewards it may obtain. If thereare few positive consequences for excellentteam performance, this will have anadverse motivational effect on teams.

The third aspect of organizationalreward systems that is important is thatrewards and objectives must focus onteam, not individual, behavior. This isadmittedly a tricky route for organizationsto navigate because most organizationalreward systems are designed for indi-viduals, and many employees may beaverse to completely giving up the possi-bility of having their performancerewarded. However, it is also true that, insome cases, individual and team-levelreward systems work at cross-purposes.For example, in professional sports, it isbecoming increasingly popular for teams toput incentive clauses, based purely onindividual performance, into players’ con-tracts. Unfortunately, in some cases, thebehaviors reinforced by these incentiveclauses do not support team performance.

In the organizational context, two otherfactors that contribute to team effectivenessare the educational and informational sys-tems present in the organization. People do


not automatically know how to work as partof a team; thus, teams often need educationaland training assistance. In addition, teamsneed valid information so that they can com-petently make decisions and carry out tasks.To the extent that either of these is lacking oris of poor quality, it will undermine teameffectiveness.

The team-design component has to dolargely with the structural features of thework team. Chief among these factors is thestructure of the task that the team is perfor-ming. According to Hackman (1987), thedesign of a team’s task has important impli-

cations for motivation (e.g., Hackman &Oldham, 1980), but it may also be importantfor other reasons. Some tasks simply do notlend themselves well to teamwork (e.g., solv-ing a highly technical problem); thus, an orga-nization may have to face the fact that a giventask may not be appropriate for teams.

The other features of the team-designcomponent are the composition of the teamand the team’s performance-related norms.Teams obviously need individuals who pos-sess the skills and abilities necessary to do thework. They also need individuals who pos-sess at least a minimal level of compatibility

FIGURE 12.3Hackman’s Normative Model of Group Effectiveness

OrganizationalContext

A context that supportsand reinforces competenttask work, via:• Reward System• Education System• Information System

Group DesignA design that promptsand facilitates competentwork on the task, via:• Structure of the Task• Composition of the Group• Norms about Performance Processes

Material Resources• Sufficiency of material resources required to accomplish the task well and on time

Group SynergyAssistance to the group byinteracting in ways that:• Reduce Process Losses• Create Synergistic Process Gains

Process Criteria ofEffectiveness

Group Effectiveness• Task output acceptable to those who receive or review it• Capability of members to work together in the future is maintained or strengthened• Members’ needs are more satisfied than frustrated by the group experience

• Level of effort brought to bear on the group task• Amount of knowledge and skill applied to task work• Appropriateness of task performance strategies used by the group

Adapted from: J. R. Hackman. (1987). The design of work teams. In J. W. Lorsch (Ed.), Handbook of organizational behavior

(pp. 315–342). Englewood Cliffs, NJ: Prentice-Hall.


in terms of personality and temperament. Aswas shown in Chapter 11, norms can have apowerful impact on behavior within teams.Most critical is whether teams develop normsthat favor a high level of performance andeffort.

As can be seen in Figure 12.3, the modelproposes that both organizational contextand team design contribute to what is termedthe process criteria of effectiveness. These cri-teria are represented by the level of effort theteam members exert, the amount of knowl-edge and skill they apply to the task, and theappropriateness of task-performance strat-egies. Hackman (1987), in effect, turnedwhat is team process in other models intoan intermediate or proximal criterion ofeffectiveness. For example, the fact that teammembers are working hard on a task is anintermediate indicator that they will be suc-cessful.

Notice, however, that the impact of bothorganizational context and team design onthe process criteria of effectiveness dependson team synergy. According to Hackman(1987), team synergy relates to the extentto which a team avoids process losses (e.g.,wasting or misdirecting time), or whetherthe team takes the initiative to create inno-vative strategic plans. Specifically, the teammay create synergistic process gains bybuilding on each other’s ideas. By puttingthis concept into his theory, Hackman(1987) reminds us that teams may squandera favorable performance situation, or, con-versely, they may do great things with onlyaverage conditions.

The next step in the model is from proc-ess criteria of effectiveness to actual teameffectiveness, which was defined previously.This represents the familiar link betweenteam process and team effectiveness, whichis part of the previous two models. Hackman(1987) adds a somewhat different wrinkle,

however, by proposing that material re-sources moderate the relation between theprocess criteria of effectiveness and teameffectiveness. A team can operate in a veryfavorable organizational context, have favor-able design features, translate these favorableconditions into positive team processes, and,ultimately, not be successful. For example,a production team will not be successful ifit does not have the tools necessary to pro-duce the required products (see also Peters& O-Conner, 1985).

Shea and Guzzo’s (1987) Model

Shea and Guzzo (1987) proposed a model ofteam effectiveness that is much less extensivethan those previously described but none-theless highlights some important determi-nants of team effectiveness. According totheir model, team effectiveness is a conse-quence of three key factors: (1) outcomeinterdependence, (2) task interdependence,and (3) potency. Outcome interdependencereflects the extent to which members of ateam share a common fate. An example ofhigh outcome interdependence would occurif all of the members of a team were to receivea cash bonus because the team performedwell. As one might expect, outcome interde-pendence is influenced, to a large extent, byorganizational compensation practices. Ac-cording to Shea and Guzzo (1987), a highdegree of outcome interdependence will fos-ter many of the behaviors necessary for teamsto be effective—for example, cooperation,workload sharing, and so on. On the otherhand, if team members’ outcomes are largelyindependent, the chances that team mem-bers will act cooperatively are reduced,thereby reducing overall team effectiveness.

Task interdependence involves thedegree to which members of a team have todepend on each other to get their work done.


Interdependence is typically cited as one ofthe crucial characteristics that define teams.Thus, task interdependence can be seenmore as a test of the appropriateness of teamsthan as a direct determinant of effectivenessper se. As one might expect, teams will gen-erally be more effective when the tasks theyare performing require a certain amount ofinterdependence. If there is very little inter-dependence, this is a sign that a team maynot be appropriate or that the task may needto be redesigned (see Comment 12.2).

Potency reflects the collective belief amongteam members that the team can be effective(Guzzo, Yost, Campbell, & Shea, 1993). It isanalogous to the individual-level construct of

self-efficacy, which represents individuals’beliefs that they can carry out various coursesof action (Bandura, 1997). According to Sheaand Guzzo (1987), potency is the most prox-imal determinant of team effectiveness—aproposition that has received some empiricalsupport (e.g., Riggs & Knight, 1994). Presum-ably, potency positively impacts team effec-tiveness because it fosters persistence and mayalso lead to greater cooperation and cohesionamong team members (see Jex & Bliese,1999). A recent meta-analysis found a rela-tively strong relationship between collectiveefficacy and team performance (Gully, Joshi,Incalcaterra, & Beaubein, 2002). Team mem-bers who have a collective belief in their ability

COMMENT 12.2

TO TEAM OR NOT TO TEAM?: A CRUCIAL QUESTION

ALTHOUGH ORGANIZATIONS ARE often quite

rational, they are also subject to fads and

fashion. After all, decisions in organizations

are made by people, and people are certainly

persuaded by fad and fashion to make a

variety of personal choices. (What else could

possibly explain the popularity of leisure

suits in the 1970s?!)The use of teams in organizations has liter-

ally skyrocketed in the past 20 years, and there

is no sign that this trend will be reversed. A

positive aspect of the trend is that teams are

highly appropriate for many organizational

tasks. That is, much of the work performed

in today’s organizations requires a high degree

of coordination and interdependence. Further-more, the speed at which the competitive en-

vironment changes often makes it unlikely that

one person can keep up with everything; thus,

knowledge sharing among team members is

often critical.

Despite all of the positive aspects of teams,

they are certainly not appropriate for every

task, nor do they fit well with the culture of

every organization. Generally speaking, teams

tend not to work well when tasks require

highly independent work, or in organizations

that do not encourage and value participation

among their employees. The author (SMJ) has

seen many cases, both as an employee and an

external consultant, in which teams are formedwith very little consideration as to whether they

are appropriate. It is as if managers often form

teams as a ‘‘reflex action,’’ which suggests that,

in some instances, the use of teams is due to fad

or fashion (‘‘Everybody uses teams’’) rather

than serious task-related considerations. This

is unfortunate because, in such instances,

teams may actually create more problems thanthey solve.

Does this mean that organizations should

not use teams? Not at all. It does mean, how-

ever, that organizations need to think very

carefully about task-related demands, and real-

istically appraise their culture before they cre-

ate teams to perform tasks.


to perform their mission will likely bettercoordinate their actions and communicateduring episodes of team performance.

Shea and Guzzo (1987) proposed thatthree key variables contribute to teampotency: (1) resources, (2) rewards, and (3)goals. As one would expect, when a team hasaccess to numerous resources, their availabil-ity will reinforce team members’ perceptionsthat they are capable of performing any taskand responding to any challenge. Rewardsare an important determinant of potencybecause they signal that performance meanssomething, and this signal provides a teamwith an incentive to develop the internalcapabilities necessary to perform well. Tothe extent that teams have challenging, spe-cific goals, effort and persistence areenhanced, as well as performance-strategydevelopment. All of these, in turn, shouldenhance potency. The contribution ofrewards, resources, and goals to potency, andultimately to effectiveness,has been supportedempirically (Guzzo & Campbell, 1990), al-though clearly more work needs to be done.

Campion’s Synthesisof Team-Effectiveness Models

Campion, Medsker, and Higgs (1993) pro-vided an extensive review of the team-effectiveness literature and, based on thatreview, a model of team effectiveness thatrepresents a synthesis of all of the modelscovered up to this point (as well as someothers). Thus, what is presented in thissection is not a new model, but rather ameta model of team effectiveness thatemphasizes the common features underlyingother models.

This synthesis of team-effectiveness mod-els is presented in Figure 12.4. Notice that thismodel is clearly simpler than the first threethat were described. Only factors that directly

influence team effectiveness are proposed;thus, many of the intermediate linkages andmoderator variables described in previousmodels are not included. The column labeled‘‘Themes/Characteristics’’ indicates that thereare essentially five general direct determinants

FIGURE 12.4Campion’s Synthesis Model of Group Effectiveness

Job Design • Self-Management • Participation • Task Variety • Task Significance • Task Identity

Interdependence • Task Interdependence • Goal Interdependence • Interdependent Feedback and Rewards

Composition • Heterogeneity • Flexibility • Relative Size • Preference for Group Work

Context • Training • Managerial Support • Communication/ Cooperation between Groups

Process • Potency • Social Support • Workload Sharing • Communication/ Cooperation within Groups

Productivity

Satisfaction

Manager Judgments

Effectiveness CriteriaThemes/Characteristics

Source: M. A. Campion, G. J. Medsker, and A. C. Higgs.

(1993). Relations between work group characteristics and

effectiveness: Implications for designing effective work

groups. Personnel Psychology, 46, 823–850. Reprinted by

permission.


of team effectiveness. The first of these, jobdesign, centers around the nature of the taskthat the team is performing. According toCampion et al. (1993), the key aspects of ateam’s task environment include the degree ofself-management and participation amongteam members, as well as the level of variety,significance, identity, and opportunity forfeedback that is built into a team’s task. Highlevels of effectiveness are associated with highdegrees of all of these factors, although fordifferent reasons. Self-management and par-ticipation enhance the team’s sense of owner-ship of their work, and the task characteristicsare likely to have their effects throughenhanced intrinsic motivation.

The second determinant of effectivenessis labeled interdependence. This representsthe degree to which team members are inter-dependent in terms of the tasks that theyperform, the goals that the team adopts, thefeedback they receive, and the rewards theystrive for. As with Shea and Guzzo’s (1987)model, Campion et al. (1993) proposed thathigher levels of interdependence in all ofthese areas should enhance effectiveness.This is proposed primarily because a highlevel of interdependence will tend to fostercooperation and will also lead team membersto coordinate their efforts.

The third determinant, labeled composi-tion, largely involves the characteristics ofthe team members themselves. One keyaspect of this is the degree to which teamsare composed of members who possess het-erogeneous skills and are flexible enough tocover for each other when required. Anotherimportant factor here is the relative size ofthe team. Although there is no magic numberthat is recognized as the ‘‘correct’’ team size,the general rule of thumb is that teamsshould be large enough to do the work, butnot too large (i.e., more is not better).Finally, the variable preference for teamwork

is somewhat unique because it is not con-tained in the previous models. Some peoplesimply like or dislike working in some teamsmore than others.

The third category, labeled context, iscomprised of factors in the organizationalenvironment in which the team performs.These include the training available to teams,the degree to which managers supportteams, and the extent to which there is coop-eration and communication among differentteams. Teams will be most effective when theorganizational context provides plentifultraining opportunities, when managers sup-port teams, and when there is a high degreeof cooperation and communication amongteams.

The last determinant of team effective-ness in this model is process. This representspotency, or team members’ collective per-ceptions of the team’s capabilities, as wellas the levels of social support, workloadsharing, and communication/cooperationwithin the team. As with previous modelsof team effectiveness, Campion et al. (1993)proposed that a positive team processdirectly facilitates team effectiveness. Thatis, teams are more successful when thereare high levels of potency (i.e., membersbelieve in the team’s capabilities), social sup-port, workload sharing, and communica-tion/cooperation among members.

The final notable feature of the Campionet al. (1993) model is the category of ‘‘effec-tiveness criteria.’’ These criteria are largelybased on Hackman’s (1987) definition ofteam effectiveness; however, notice that theone component of Hackman’s definition thatis missing is viability, or the likelihood thatmembers of a team will work together in thefuture.

The Campion et al. (1993) model pro-vides a comprehensive account of the majorfactors that influence team effectiveness and


has received empirical support. Specifically,Campion et al. and Campion, Papper, andMedsker (1995) showed that many of thecharacteristics proposed to be related toteam effectiveness were related to effective-ness criteria among employees in a largeinsurance company. Thus, like Gladstein(1984), Campion et al. not only proposed amodel but tested it as well. The most obviousweakness in the Campion et al. model is thatit proposed direct relations only betweenthemes/characteristics and effectiveness cri-teria.

A Summary of theTeam-Effectiveness Models

In this section, we examined four of the mostwidely cited models of team effectiveness andone attempt to synthesize these models. Thecommon characteristic in all of these modelsis they follow the familiar input–process–output sequence—that is, they propose thataspects of the organizational context (e.g.,rewards, interdependence, task design)directly impact the way a team works, andthis, in turn, impacts effectiveness. Thesemodels are valuable because they highlighta number of organizational and team factorsthat influence team performance. Further-more, there appears to be a good deal ofconsensus on what these contextual factorsare. For example, most models highlight thecomposition of the team, the reward systemsunder which teams work, the design of thetask that the team is working on, the re-sources available to the team, the goals thatare set for a team (or those that it sets foritself), and internal processes such as cohe-siveness, communication, and conflict man-agement. Organizations have some degree ofcontrol over all of these variables and thusmay be able to change them in order toimprove team effectiveness. Given their

importance, each of these factors is examinedin greater detail in the following section.

Although models based on the input–process–output approach have served teamresearchers well over the years, authors haverecently recognized some limitations in thesemodels. Ilgen et al. (2005) point out thatmany of the team-level variables found toinfluence the relationships between inputsand outputs are not necessarily processes,but rather emergent states such as a sense ofcollective efficacy. In addition, the authorsnote ‘‘the I-P-O framework limits researchby implying a single-cycle linear path frominputs through outcomes’’ (p. 520). Rigidadherence to this type of framework preventsresearchers from examining the consequencesof various outputs (e.g., performance) forcharacteristics of the teams or processesengaged in by the team members. Finally,Marks, Mathieu, and Zaccaro (2001) notedthat existing input–process–output modelsfail to consider the importance of temporalfactors in how teams function. These authorsnoted that teams go through different phases(planning their performance, implementinggoal-directed activities) as they do their work,and these phases need to be incorporated intomodels of team effectiveness. Therefore,future modeling of team effectiveness willneed to consider bidirectional influences be-tween components of input–process–outputmodels and the importance of temporalphases in team performance.

DETERMINANTS OF TEAMEFFECTIVENESS

Although models of team effectiveness areuseful in highlighting important variablesthat contribute to team effectiveness and inmapping out the processes by which theyhave such an impact, they are often shorton detail. In this section, we take a closer


look at several of the most common deter-minants of team effectiveness.

Team Composition

Perhaps one of the most robust findings inthe team-effectiveness literature is that teamsstaffed with more highly skilled members aremore effective than teams possessing lowerabsolute skill levels (Guzzo & Shea, 1992;Hackman, 1987). The average cognitive abil-ity of a team has been found to be a predictorof performance in different types of teams,including military tank crews (Tziner &Eden, 1985). This finding is consistent withthe literature on individual-level perfor-mance that has shown a very robust relationbetween cognitive ability and performance(Schmidt & Hunter, 1988). Devine and Phil-lips (2000) conducted a meta-analysis acrossall of the studies investigating the relation-ship between cognitive ability and team per-formance, and found the relationship wasstrongest when the teams were performingan unfamiliar task.

Given that team members’ skill/abilityis positively related to team effectiveness,does this mean that organizations needonly hire talented people in order to maketeams effective? Probably not. Another rel-atively consistent finding in the literatureis that, although the absolute level of skill/ability in a team is important, organiza-tions must also consider the mix of skills/abilities within teams (Campion et al.,1995). For example, if a basketball teamis to be effective, it is certainly necessary tohave players who are capable of scoring alot of points. On the other hand, it is alsoimportant to have players who reboundwell and others who are defensive special-ists. A team composed only of scorerswould probably not do very well. In fact,it has been shown that diversity of team

skills is positively related to team perfor-mance (Guzzo & Shea, 1992).

Despite the importance of team mem-bers’ skill/ability, there are instances inwhich even teams composed of highly tal-ented members are ineffective. In addition tocognitive ability, team personality has alsobeen found to predict team performance.The most reliable personality predictor isteam conscientiousness, which at the indi-vidual level reflects someone who is reliable,orderly, and follows through with differenttypes of tasks (Barrick & Mount, 2005). VanVianen and De Dreu (2001) found that team-level agreeableness and conscientiousnesswere both related to team performance. Inaddition, a recent meta-analysis revealed thatteam agreeableness and conscientiousnesswere both related to superior team perfor-mance across a number of published studies(Peeters, Van Tuijl, Rutte, & Reymen, 2006).Another interesting finding in the meta-analysis was that variability among teammembers in both agreeableness and consci-entiousness was negatively related to teamperformance. That is, controlling for theoverall level of agreeableness or conscien-tiousness, teams that varied widely on agree-ableness or conscientiousness tended toperform poorer.

Researchers have also examined whetherthe relationship between personality andteam performance may depend on thetype of team task being performed. LePine,Colquitt, and Erez (2000) found that consci-entiousness was negatively related to teamdecision-making effectiveness when the de-cision rules presented to the team requiredadaptability. The need for order possessedby conscientious individuals may makeadapting to changing circumstances moredifficult.

Another reason that team members’ per-sonalities may be important is that they may


influence the climate within the team.George (1990), for example, found thatteams could be distinguished in terms oftheir affective tone, which was determinedby the personalities of individual team mem-bers. Teams with more positive affectivetones were found to provide higher levelsof service to customers—an important aspectof performance for these teams, who wereemployed in a retail environment. The factthat personality influences the climate ofteams should not be surprising, given thatthe climates of entire organizations are influ-enced by the personalities of individual em-ployees (Schneider, 1987).

In addition to the direct impact of teammembers’ personalities, as well as theimpact on team climate, the particular mixof personalities within a team may also havean important effect on team effectiveness.Within team settings, individual teammembers may have personalities that‘‘clash,’’ thus leading to negative conflict.It is also possible that one team memberwho possesses a very negative personalitytrait can have a negative impact on theprocesses within a team and may ultimatelyhave a negative impact on performance. Forexample, in a study by Barrick et al. (1998),teams that had at least one member with avery low level of emotional stabilityreported lower levels of social cohesion,flexibility, communication, and workloadsharing, and higher levels of conflict com-pared to teams that did not have such amember. Interestingly, though, having atleast one team member with low emotionalstability did not have a negative impact onteam performance.

A final aspect of team composition thatmay contribute to team effectiveness is theattitudes of team members. The impact ofteam members’ attitudes can be seen in oneof two ways. The most direct way is what

Campion et al. (1993) termed ‘‘preferencefor team work (p. 828),’’ which reflectswhether people like working in team set-tings. Despite the proliferation of teams inorganizations, it is still likely that not every-one enjoys working in team settings.Despite the camaraderie that is often foundin collective work, teamwork can be frus-trating because of social loafing, conflict,and other difficulties. Campion et al.(1993) found that where the average levelof preference for teamwork was low, teamsperformed lower on several performancecriteria (see also Driskell, Goodwin, Salas,& O’Shea, 2006).

Another way that attitudes may comeinto play involves the similarity of teammembers’ attitudes. Social psychological re-search has shown conclusively that peopletend to like people who are perceived asbeing similar to themselves (Byrne, 1971).Thus, teams are likely to function moreeffectively when there is at least a moderatelevel of similarity in their attitudes towardimportant issues such as the effectiveness ofthe team leader or the supportiveness of theorganization (Bliese & Britt, 2001). Despitethe potential value of agreement, it is alsopossible that agreement may be too high.When the members of a team agree oneverything, they may suppress neededdebate and cause the team to becomeextremely resistant to change. Janis (1982)referred to such a dynamic as groupthink,which he defined as ‘‘ . . . a mode of think-ing that people engage in when they aredeeply involved in a cohesive in-group,when the members’ striving for unanimityoverrides their motivation to realisticallyappraise alternative courses of action. . . .Groupthink refers to a deterioration ofmental efficiency, reality testing, and moraljudgment that results from in-group pres-sures’’ (p. 9; see also Whyte, 1998).


Task Design

Another consistent theme among the modelsreviewed is that the design of a team’s task is akey variable that influences team effectiveness.The exact manner in which task design isproposed to influence team effectivenessvaries considerably between models, however.At the most basic level, task design is impor-tant because it speaks to the issue of whetherthe task a team is performing is appropriate forteams. Generally speaking, tasks involving ahigh level of interdependence are best accom-plished by teams. Conversely, tasks thatrequire primarily independent work are bestperformed by individuals. The zeal to useteams may blind organizations to the factthat some tasks are best accomplished byindividuals.

Assuming for the moment that the task ateam is performing is appropriate for teams,in what other ways does task design influ-ence team effectiveness? One way is throughits motivational impact. Individuals’ motiva-tion can be increased by redesigning jobs toenhance features such as autonomy, feed-back, skill variety, task significance, and taskidentity (Hackman & Oldham, 1980). Thissame basic logic applies to teams; thus,members of teams that are performing tasksthat are challenging, interesting, and engag-ing are likely to be more motivated thanmembers of teams performing more mun-dane tasks.

The design of a team’s task is also impor-tant. Even tasks that are well suited for teamscan be approached in a variety of ways; thus,some task-related strategies will be moreeffective than others. Therefore, an impor-tant determinant of team effectiveness is theextent to which strategies and approachesare appropriate for the task(s) it is perfor-ming. It has been shown that teams (a) tendnot to allocate much of their time to discus-

sions of task-related strategies (Hackman &Morris, 1975), and (b) tend to be moreeffective when they do, especially when theyare performing complex tasks (Hackman,Brousseau, & Weiss, 1976; Mathieu etal., 2000).

Beyond the important impact it has oneffectiveness, the nature of a team’s task isoften a key to understanding team interac-tions and processes. Many of the behavioraldynamics that occur within teams can beunderstood only within the context of thework that a team performs. For example,Denison and Sutton (1990) describe manyof the dynamics (e.g., loud music, joking)that occur in an operating room. To a largeextent, these behaviors represent team mem-bers’ mechanism for reducing tensionbecause the work they do often has life-and-death consequences.

Organizational Resources

Just as individuals need organizational re-sources to perform well, so do teams. Giventhis need for organizational resources, animportant question is: ‘‘What specific orga-nizational resources do teams need in orderto be effective?’’ To a certain degree, teamsneed many of the same resources that indi-viduals need. Consider, for example, Petersand O’Connor’s (1988) classification of orga-nizational conditions that constrain individ-ual performance. Many of these conditionsalso apply to teams. For example, teamsoften need equipment, information, budget-ary resources, and time in order to accom-plish their assigned tasks.

Teams may also have important andunique resource needs. Chief among theseare training and consultation on how to workas a team. Members of teams may need train-ing and assistance in learning how to workcooperatively with others, understanding


how to coordinate their efforts with otherteam members, and, perhaps, trainingin how to resolve task-related disputes(Driskell, Salas, & Johnston, 2001). In theauthors’ experience, many organizationsform work teams and instruct team membersto ‘‘work out the details’’ with respect to taskcompletion, but the organizations ignoreissues associated with the internal dynamicsof the team. Such an approach is based onthe rather naıve assumption that teamworkcomes naturally to people and they will beable to adapt to it naturally. Even a collectionof highly intelligent, reasonable individualsmay perform very poorly as a team.

Another resource need that is unique toteams is meeting space/time. For a team to besuccessful, the members need to be morethan a collection of individuals working ontheir own. To be a collective entity, teamsneed to come together face-to-face, commu-nicate with each other about task-relatedmatters, and develop some level of cohesion.If an organization provides no meetingspace, or very limited time, a team may neverfully develop into a mature performing unit.

A final resource to consider is leadership.Just as individuals often rely on leaders inorder to be successful, teams need leadershipin order to be effective. Leaders of teams,however, must strike a delicate balance be-tween being too detached and unavailableand being too directive, particularly in‘‘self-managed’’ or autonomous work teams.Hackman (1990) likens the leadership ofteams to being on a balance beam to showhow leaders must often strike this delicatebalance. What, then, is the appropriate bal-ance? The answer depends, to a large extent,on the nature of the team one is leading andthe organization in which the team resides.For teams that are largely self-directed,Hackman suggests that leaders essentiallyneed to provide the team with an engaging

sense of direction,’’ provide task-relatedassistance and consultation as needed, andwork within the organization to obtain re-sources for the team. For anyone who hasfilled this role, these are demanding activitiesthat are quite different from more traditionalforms of supervision. DeChurch and Marks(2006) recently found that quality of leadercoordination and strategizing were positivelyrelated to team performance in an F-22 battlesimulation task.

Rewards

One of the most consistent findings withrespect to individual-level performance isthat rewards tend to enhance effort and, asa result, often lead to high levels of perfor-mance (Luthans & Kreitner, 1985). Rewardsare also an important determinant of teamperformance, although the issues organiza-tions face in designing team-level rewardsystems are vastly different from those facedwhen designing rewards for individuals.Many organizations make the mistake ofespousing the value of ‘‘teamwork’’ whilerewarding employees solely on the basis oftheir individual accomplishments. Undersuch conditions, it is unlikely that teams willperform very effectively. Recognizing this,more and more organizations have beendeveloping team-based compensation plans(Lawler, Mohrman, & Ledford, 1995), andthere is some empirical evidence to suggestthat team-based rewards are associatedwith team effectiveness (Hertel, Konradt, &Orlikowski, 2004).

One of the most fundamental issues orga-nizations face in designing reward systemsfor teams is the extent to which the workthey do is truly interdependent. If the workthat teams do is truly interdependent, thenrewarding teams rather than individuals isappropriate (Wageman & Baker, 1997). But


what if a team performs a task that is nothighly interdependent? Team-level rewardsystems are then less appropriate and, in fact,may be downright unfair. Consider, for ex-ample, a team of five individuals who performindependently. If rewards are based on teamperformance and the team performs well, lesscompetent team members are rewardedbecause they are lucky enough to be in a teamwith others who are highly competent. Onthe other hand, if the team performs poorly,highly competent individuals are penalizedbecause they happen to be in a team withfellow employees who are much less compe-tent than they are.

Another important issue that organiza-tions must consider in the design of team-level reward systems is the interplay betweenindividual- and team-level rewards. It is rareto find organizations in which employees arerewarded solely on the basis of team-levelperformance. In most organizations thathave team-level reward systems, these mustcoexist with individual-level reward systems(DeMatteo, Eby, & Sundstrom, 1998). Fur-thermore, in most cases, individual-levelreward systems are better established thanthose based on team performance. In theauthors’ opinion, it is unrealistic to thinkthat individual-level reward systems can beeliminated, even in organizations that de-pend heavily on the performance of teams.

It is important to not have individual-and team-level reward systems work againsteach other. If a team-level reward system isgoing to be effective, it is important to nothave behaviors encouraged by the individ-ual-level reward system undermine teamperformance, and vice versa. This may seemto be a rather obvious proposition, but it isactually quite common in organizations. Forexample, in sales organizations, individualsare typically rewarded largely on the basis ofthe dollar value of their own sales. Thus,

individuals are highly motivated to make asmany sales as possible, and will likely allo-cate their time accordingly.

On the other hand, when one considersthe performance of a sales team, each indi-vidual’s effort to maximize his or her salesvolume may not necessarily maximize theteam effort. In the long run, the team maybe better served if individual salespeopledevote at least some of their time to servicingexisting accounts, providing training andguidance to less-experienced salespeople,and making sure they keep current onemerging product lines. Unfortunately, theprospect of earning high sales commissionsmay induce a form of ‘‘tunnel vision’’ inemployees that ultimately undermines teamperformance. Furthermore, organizationsare often quite naıve about this conflict be-tween employees’ maximizing their ownrewards and contributing to the performanceof the team or of the organization as a whole.

A third consideration in the design ofteam-level reward systems is the degree towhich teams have control over their ownperformance. This is an issue in the designof individual-level reward systems, but it isespecially salient with team-level reward sys-tems because the performance of teams maybe highly dependent on the efficiency andreliability of the technologies they employ(Goodman, 1986), or it may be limited byorganizational resource constraints (Shea &Guzzo, 1987). Holding teams to high per-formance standards while, at the same time,providing them with faulty technology orvery limited resources is unfair and counter-productive. Teams may be able to overcomesuch performance barriers (Tesluk &Mathieu, 1999), but it is questionablewhether this can be done in the long runwithout having adverse motivational effectson team members and ultimately undermin-ing the team’s viability (Hackman, 1987).


One final issue to consider in examiningthe consequences of team reward systems isthe attitudes of employees toward team-based rewards. Haines and Taggar (2006)recently examined the predictors of positiveattitudes toward team reward structures. Theauthors found that the most direct predictorsof a positive attitude were belief in the valueof teamwork and the employee’s job perfor-mance. Interestingly, believing in the valueof teamwork was positively related to a favor-able attitude to team-based reward struc-tures, but individual performance wasnegatively related to such a favorable atti-tude. Individuals who performed well asindividuals were less inclined to wantteam-based reward systems, which mayreflect a concern that their individual effortswill go unnoticed by their supervisors.

Team Goals

In many ways, the impact of goals on teamperformance mirrors the impact of goals onindividual-level performance. For example,O’Leary-Kelly, Martocchio, and Frink (1994)conducted a meta-analysis of the team goal-setting literature and found strong supportfor the impact of goals on team performance.This finding has been reinforced by individ-ual studies that have been conducted subse-quent to this review (e.g., Whitney, 1994).Given such findings, it is tempting to con-clude that there is a perfect correspondencebetween team and individual goal setting.

There are, however, some important dif-ferences in the dynamics associated withindividual- and team-level goal setting. Forexample, an important consideration in set-ting goals for teams is the interplay betweenteam goals and individual-level goals. This isanalogous to the dilemma organizations facein developing team-level reward systems. Ithas been found, for example, that team-level

goals are much more effective when indi-viduals either do not have goals or havegoals that are compatible with team goals(Mitchell & Silver, 1990). Thus, organiza-tions must be cognizant of the overall goalsystem when they are developing perfor-mance-related goals for teams (DeShonet al., 2004).

Another important divergence betweenthe individual and team goal-setting litera-tures has to do with the intervening mecha-nisms between goals and performance.Because the impact of goals is so well estab-lished, goal-setting researchers have recentlyfocused the majority of their attention onexplaining why goal setting works (seeChapter 8). Most explanations for the effectsof goal setting have centered on the factthat goals serve as an important focal pointfor self-regulation (Klein, 1989), and goalcommitment is a crucial intervening stepbetween goals and performance (Ambrose& Kulik, 1999).

Although comparatively less research hasexamined the processes responsible for teamgoal-setting effects, there is some evidence thatthe mechanisms responsible for these effectsare quite different from those responsible atthe individual level. For example, there issome evidence that collective efficacy may bean important intervening mechanism betweenteam goals and performance. Whitney (1994)found that teams that were assigned difficultgoals reported higher levels of collective effi-cacy, compared to teams with easier or ‘‘doyour best’’ goals. Thus, teams given difficultgoals may take this as a sign that the organi-zation considers them more capable thanother teams (e.g., ‘‘We have a difficult goal,therefore we must be good.’’). These enhancedfeelings of competence may then subsequentlylead to a high level of performance throughmany of the same mechanisms by which goalsenhance individual-level performance (e.g.,


increased effort, commitment, strategy devel-opment).

Another important difference betweenindividual- and team-level goal processes isthat difficult goals may enhance members’attraction to the team or team cohesion(Whitney, 1994); that is, a difficult goalmay serve to ‘‘draw team members together’’in a way that enhances attraction to the teamand ultimately enhances performance. Sup-porting this idea, Aube and Rousseau (2005)examined the relationship between teamgoal commitment and multiple indicatorsof team effectiveness (e.g., team perfor-mance, team viability) among intact teamsworking in Canadian organizations. Theauthors found a positive relationship be-tween team goal commitment and indexesof effectiveness. The authors also found thatthis relationship was completely a functionof supportive team behaviors. That is, teamscommitted to attaining a group goal weremore likely to help each other out whenneeded, which resulted in superior teamperformance. It is important to point out,however, that for enhanced cohesivenessand attraction to facilitate team performance,a team must have norms that support highlevels of performance (Seashore, 1954).

Within-Team Processes

Organizational psychologists have been veryactive in identifying key team processes thatconnect inputs to indicators of effectiveness.However, although team processes are seen ascentral to understanding the effectiveness ofteams, Kozlowski and Bell (2003) note‘‘ . . . there is little convergence on a coreset of processes’’ (p. 346). Numerous taxon-omies of team processes have been offeredby different researchers (Dickinson &McIntyre, 1997; Hallam & Campbell, 1997;Mathieu & Day, 1997; Kozlowski & Bell,

2003; Salas et al., 2005). Despite lack ofcomplete agreement in the processes that in-fluence team performance, most of these tax-onomies highlight a handful of processes thathave been empirically found to predict teamperformance in both laboratory and appliedsettings. The purpose of this section is tofocus on certain elements of team process thatseem to be more important than others.

In the most general sense, team processrepresents the way in which teams accom-plish their tasks. Team process is not con-cerned with what a team produces, butrather how a team does its work. Given thisgeneral definition, a number of different ele-ments of team process may be important toteam effectiveness. For the present chapterwe highlight the team processes presented inTable 12.1. Table 12.1 also divides theseprocesses into those that are behavioral andcognitive/affective in nature.

BEHAVIORAL PROCESSES

Many of these processes are self-explanatoryand make intuitive sense. Considering thebehavioral team processes, communicationamong team members would appear tobe a necessary requirement for effectiveperformance, and research supports thisprediction (Ancona & Caldwell, 1992a,1992b). Communication within a teamreflects both the amount of informationflow and the manner in which informationis disseminated. Teams in which membersshare little information with each other andcommunicate very infrequently tend to per-form more poorly than teams with morefree-flowing communication (Hackman,1987). The size of this difference, however,depends to a large extent on the nature ofthe task that a team is performing. Highlyinterdependent and complex tasks havemuch greater information requirements


than tasks that are simple or require verylow levels of interdependence. In addition,researchers have also distinguished be-tween task-related and task-unrelated com-munication, with the latter type ofcommunication being related to poor teamperformance (Toquam, Macauley, Westra,Fujita, & Murphy, 1997).

Another aspect of communication,which was discussed in the previous chapter,is the extent to which a team’s style of com-munication is either highly centralized orhighly decentralized. Recall that centralizedforms of communication have been found tobe effective when teams are performing tasksthat are relatively simple and have few coor-dination requirements. Decentralization,however, tends to facilitate performancewhen a team is working on highly complextasks or tasks that have high coordinationrequirements.

A manifestation of this issue (centraliza-tion versus decentralization) that the authorshave observed over the years when conduct-ing team-development activities is the extentto which all members of a team participate.Although no data can be presented to backthis up, one of the most common attributesof teams that appeared to be performing wellhas been that all members tended to partic-ipate and voice their opinions. On the otherhand, in more dysfunctional teams, it wascommon to find participation very unevenamong team members. Typically, one or twohighly vocal team members dominated agreat deal of the team’s deliberations. Thismay lead a team to become dysfunctionalbecause those who are the most vociferousmay not always have the best ideas (in fact, inthe author’s experience, the opposite is oftenthe case). Thus, many potentially good ideasnever see the light of day.

TABLE 12.1Behavioral and Cognitive/Affective Team Processes

Category Description

Behavioral

Communication: ‘‘Exchange of information between two or more team members’’ (p. 25; Dickinson

& McIntyre, 1997)

Coordination: ‘‘Team members executing their activities in a timely and integrated manner’’

Performance of team members influenced by others (Dickinson & McIntyre, 1997)

Cooperation: Team members helping one another to complete interdependent tasks (Wagner,

1995). Seen as the opposite of conflict (Kozlowski & Bell, 2003)

Cognitive/Affective

Team Cohesion

Task: ‘‘The group’s shared commitment or attraction to the group task or goal’’ (p. 349,

Kozlowski & Bell, 2003)

Interpersonal: The extent to which group members like each other and are attracted to each

other (Siebold, 2006)

Team Efficacy: ‘‘A team’s belief that it can successfully perform a specific task’’ (p.820, Gully et al.,

2002)

Team Empower.: Task motivation that is due to team members’ collective, positive assessments of

their organizational tasks (Kirkman et al., 2004)

Shared Mental Mod.: Shared knowledge structures about team tasks, goals, and activities (Salas et al.,

2005)

Team Conflict: Conflict over tasks or emotional issues within a team (Jehn, 1994)


With respect to communication, a finalissue that has only recently become impor-tant is teams’ use of electronic communica-tion (e.g., e-mail). Employees in manyorganizations rely heavily on electronic com-munication, and there has been some re-search on this form of communication. Forexample, although electronic communica-tion is highly efficient for transmitting factualinformation (Carey & Kacmar, 1997), it ismuch less effective when a team’s task ishighly interdependent, or the informationto be communicated is emotive in nature(Straus & McGrath, 1994). Also, becauseusers of electronic communication may feela certain degree of anonymity that is notpossible in face-to-face encounters, elec-tronic communications may be more bluntand less tactful (Dubrovsky, Keisler, &Sethna, 1991).

Research on electronic communicationwithin teams is still relatively new, but someconsistent findings are beginning to emerge.For example, it has been found that membersof teams generally prefer not to use electroniccommunication media when important,task-related information must be communi-cated (Straus & McGrath, 1994). The reason:Electronic communication is a rather limitedcommunication medium. It is not possible touse physical gestures and other forms ofnonverbal communication, so it may takelonger for a person to get a point across(McGrath, 1990).

Despite this somewhat negative assess-ment of electronic communication, thereare some instances when it may be superiorto have over face-to-face communication.For example, if the purpose of the commu-nication is simply to disseminate informa-tion, electronic communication is muchmore efficient than a face-to-face meeting.Some readers have probably had the experi-ence of attending a meeting that was merely

an ‘‘information dump’’ and could have beenaccomplished by other means (e.g., elec-tronic transmission of information).

Another instance in which electroniccommunication may be effective is whenteams are working on brainstorming tasks.Brainstorming is often used when teams needto come up with creative or novel ideas. Themajor ground rules of brainstorming are thatparticipants should be encouraged to gener-ate as many ideas as possible (e.g., quantity ismore important than quality), ideas shouldnot be evaluated as they are generated, andparticipants should try to build on the ideasof others. Certainly, under the right condi-tions, brainstorming can be done very wellface-to-face. However, in many cases, partic-ipants may experience evaluation apprehen-sion or become hesitant at the prospect ofsharing novel ideas in front of others. Thereis some evidence that when brainstorming isdone through electronic communication,participants may overcome such apprehen-sion (Gallupe et al., 1992). This is due largelyto the anonymity that electronic communi-cation allows.

The research on electronic communica-tion, although still fairly new, is important.As more and more organizations make use ofteams in which members are geographicallydispersed (e.g., ‘‘virtual’’ teams; Kirkman &Mathieu, 2005), it will become crucial tounderstand the limitations of such arrange-ments. The little research done so far sug-gests that teams may lose something if theyare allowed to communicate only electroni-cally. Clearly, more research is needed tostrengthen these findings and communicateother limitations of various communicationmedia.

Coordination has been defined as ‘‘activi-ties required to manage interdependencieswith the team work flow’’ (p. 352; Kozlowski& Bell, 2003, and coordination has been


measured through observer ratings of teammember interactions or through the actualmonitoring of exchanges between team mem-bers as a task is completed (Brannick, Roach,& Salas, 1993; Stout, Salas, & Carson, 1994).Cooperation, also referred to as ‘‘backup-behavior,’’ refers to the extent to which em-ployees help each other and give each otherneeded information in completing tasks(Wagner, 1995). Cooperation can beassessed either by surveying team members(Hallam & Campbell, 1997) or by examiningthe number of exchanges that occur betweenteam members (Seers, Petty, & Cashman,1995). Communication, coordination, andcooperation all represent team processes thatare conducive to effective team performance.

Affective/Cognitive Processes

The remainder of team processes describedin Table 12.1 deal with either cognitiveissues related to how team members thinkabout one another and their tasks or affec-tive variables related to how team membersfeel about each other and the work they aredoing as a team. Team cohesiveness is an-other variable that is measured in mostteam-effectiveness models, but not all mod-els consider it to be an aspect of team pro-cess. Although cohesiveness has beendefined in a variety of ways, most teamresearchers see it as the degree to whichthe members of a team are attracted to theteam and place a high value on team mem-bership (Mudrack, 1989). In everyday lan-guage, cohesiveness is often described asthe level of ‘‘team spirit’’ or ‘‘espirit decorps’’ that we see within teams.

It has been demonstrated that highlycohesive teams tend to be more effectivewhen prevailing norms support high perfor-mance (Seashore, 1954). However, we alsoknow that it may be possible for teams to be

too cohesive. As was noted earlier, Janis(1982) believed that ‘‘groupthink’’ is likelyto occur when teams are so cohesive thatthey lose the capacity to realistically appraisetheir capabilities. In terms of team process,this is manifested in the suppression of dis-sent and inaccurate appraisals of the envi-ronment.

More recent research suggests that teamcohesiveness may be multidimensional, andthat this may have an impact on team effec-tiveness. According to Hackman (1992),it is possible to distinguish between interper-sonal-based cohesiveness and task-based co-hesiveness (see also Mullen & Cooper,1994; Siebold, 2006). When cohesivenessis interpersonal, members’ attraction to theteam is based largely on how much they likethe other team members and enjoy theircompany. Members of a college fraternityare likely to be cohesive for largely interper-sonal reasons. The same would apply tomore informal teams and friendship cliques.When cohesiveness is task based, members’attraction to the team is largely based on theirattraction to the task that the team is perfor-ming (see Table 12.1). The members of apresidential campaign staff may develop ahigh level of cohesiveness because of theirmutual support for the candidate they aresupporting. Members of professional sportsteams may also develop this form of cohe-siveness as they strive to win a champion-ship.

Distinguishing between interpersonal andtask-based cohesiveness may have importantconsequences for team effectiveness. Mullenand Cooper (1994) conducted a meta-analysis of the team cohesion literature andconcluded that task cohesion was a strongerpredictor of team performance than interper-sonal cohesion (see also Siebold, 2006). Thissuggests that just because people like eachother and enjoy being together, they will not


necessarily be an effective team. In fact, itwould not be hard to imagine a scenario inwhich a very high level of interpersonal cohe-siveness could even be counterproductive.Members of a team may enjoy each other somuch that they allocate very little time to theirtask.

According to Hackman (1992), this dis-tinction in forms of cohesiveness highlightsvery clearly the importance of a team’s taskdesign. If tasks are designed so that they areinteresting, challenging, and psychologicallyengaging, the chances are increased thatteam members will develop high task-basedcohesiveness that will ultimately enhance ateam’s performance. On the other hand, if ateam’s task is very mundane and uninterest-ing, there is less chance that high levels oftask-based cohesiveness will develop. Ulti-mately, team members may be very ‘‘luke-warm’’ about what they are doing, and thismay be reflected in the team’s performance.In the future, more research on the dimen-sionality of cohesiveness, and the implica-tions of this, is needed.

In addition to team cohesion, earlier inthe chapter we discussed the importance ofteam efficacy, sometimes also referred to ascollective efficacy. As indicated in Table 12.1,collective efficacy refers ‘‘a sense of collectivecompetence shared among members whenallocating, coordinating, and integratingtheir resources as a successful, concertedresponse to specific situational demands’’(Zaccaro, Blair, Peterson, & Zazanis, 1995,p. 309). In many ways collective efficacy canbe seen as the team members’ beliefs thatthey can accomplish the team’s objectivesunder different operational conditions. Notsurprisingly, collective efficacy has beenfound to predict better team performanceon a variety of tasks (Gully et al., 2002).Team members who have a collective beliefin their ability to perform their mission will

likely better coordinate their actions andcommunicate during episodes of team per-formance.

A variable related to collective efficacy isteam empowerment. Kirkman, Rosen, Tesluk,and Gibson (2004) defined team empower-ment as increased task motivation that is dueto team members’ collective, positive assess-ments of their organizational tasks. Whereascollective efficacy is focused on team mem-bers’ beliefs about their ability to performteam tasks, team empowerment refers tothe team members’ beliefs in the meaningful-ness of their tasks. According to Kirkmanet al., team empowerment is based on thestrength of the team (potency), the perceivedmeaningfulness of team tasks, a sense ofautonomy in being able to perform as a team,and a sense that the team is having a positiveimpact on broader organizational goals.These four variables are additive and subjectto change over time. Kirkman et al. foundthat team empowerment predicted theperformance of virtual teams as indexed byratings of customer satisfaction and improve-ment in the time it took to process tasks.Furthermore, the relationship between teamempowerment and process improvementwas especially strong when teams had fewerface-to-face meetings. This research suggeststhat team empowerment may be especiallyimportant when team members are less ableto interact and communicate.

The next cognitive/affective team-levelprocess presented in Table 12.1 is sharedmental models. Shared mental models allowteam members to organize information thatdeals with team goals, member tasks, andeven how to coordinate efforts. Salas et al.argue there are two forms of mental models:team related and task related. Team-relatedmental models deal with team functioning.Task-related models deal with materials thatare needed or information on how certain


equipment is used. It has been posited thatthese shared mental models can allow teamsto work more effectively and efficiently.Mathieu, Heffner, Goodwin, Salas, andCannon-Bowers (2000) found that team-related mental models were significantlyrelated to team performance, and task-relatedmodels were indirectly related to team perfor-mance through team processes.

The final important component of teamprocess that can influence team effectivenessis the manner in which teams handle con-flicts. Any time people come together toaccomplish a task collectively, some level ofconflict is inevitable. Members of a team maydisagree over a variety of issues, such as whoshould play a leadership role in the team orhow the task should be approached. Someindividuals may simply not like each otherbecause of fundamental personality andvalue differences. In addition to the sourcesof conflict, the level of conflict may vary fromvery minor friendly disagreement all the wayto physical aggression.

Although conflict is consistently nega-tively associated with team member atti-tudes, it is not always negatively associatedwith lower levels of team performance. Ac-cording to Jehn (1994), conflict within teamscan be characterized as being task relatedand emotion related. When conflict is taskrelated, team members have differences overtask-related issues. For example, they maydisagree on the way a team should prioritizeits work, which work methods are mostappropriate, or which team members shouldbe performing which tasks. Jehn states thattask-based conflict does not detract fromteam effectiveness and, in many cases, canactually enhance it. When team members’disagreements center on task-related issues,this may facilitate communication aboutthese issues and ultimately generate innova-tive ideas that enhance effectiveness.

In contrast, when conflict is emotionrelated, team members have differences overmore personal issues. Even reasonable peoplemay ‘‘see the world differently’’ because offundamental value differences, and thus notget along well. According to Jehn (1994),emotion-related conflict can and often doesdetract from team performance. In fact, if suchconflicts are allowed to escalate to extremelyhigh levels, they may ultimately lead to theabolishment of a team. Why is this form ofconflict so destructive? One reason is that it issimply unpleasant for all team members.Being in a heated conflict with another personis unpleasant, and it is also unpleasant towatch such conflicts unfold. Thus, in manycases, team members may simply withdrawfrom such unpleasant situations.

Perhaps a more important reason thatemotion-related conflicts reduce effective-ness is that they distract team members’attention away from important task-relatedmatters. When two people in a team becomebitter enemies, they may become so focusedon ‘‘one-upping’’ each other that they paylittle attention to the task that they are sup-posed to be performing.

Between-Team Processes

Although most of the research on processesin the team-performance literature has fo-cused on within-team dynamics, somerecent research has also emphasized theimportance of examining processes thatoccur between teams as predictors oforganization-level performance (Marks et al.,2005; Mathieu & Day, 1997). Mathieu andDay point out that organizational success isfrequently predicated on effective between-team communication, coordination, andcooperation, and that researchers examiningorganizational effectiveness need to map outcritical team-to-team interfaces and assess the


quality of the interactions between theseteams. These authors identified five importantprocesses that occurred between teams in anuclear power plant. Formalization wasdefined as the extent to which there are rules,procedures, and standard methods for teammembers to coordinate their activities. Coor-dination refers to the sequencing and spacingof activities between departments to accom-plish a common goal. Cooperation refers to theextent to which employees from differentdepartments maintain a positive interpersonalrelationship. Goal priority refers to the extentto which employees from different depart-ments agree about the priority of organi-zational goals. Finally, interdependence refersto the extent to which the actions of onedepartment affect the operations of anotherdepartment.

Mathieu and Day (1997) examined theinteractions between teams within a nuclearplant and were able to uncover that someteams shared a high level of interdepend-ence with many other teams (e.g., opera-tions, radiological control), whereas otherteams operated in relative isolation fromeach other. Importantly, there was variabilityin which teams within the plant scored highversus low on the team processes that wereassessed. Some teams were judged higher oncoordination and cooperation, whereas otherteams scored higher on goal priority. Mathieuand Day used this information to developtargeted recommendations for individualteams. Given the amount of cross-team inter-action necessary in modern organizations,research on between-team processes is likelyto increase in the future.

ENHANCING THEEFFECTIVENESS OF TEAMS

Assuming for the moment that an organiza-tion has identified tasks that are appropriate

for teams, what specific steps can organiza-tions take to increase the odds that teams willbe effective? In this section, we examinethree general approaches that organizationscan take to enhance team effectiveness. Asreaders will note, these steps correspondto activities that are typically conductedby Human Resources departments inorganizations. This was done intentionally,to emphasize that they are not just generalconcepts; they are concrete, ‘‘actionable’’steps.

Selection

Based on research by Barrick et al. (1998), itappears that one way of improving teameffectiveness is to simply use some of thesame factors that have been found to berobust predictors of individual performance(Barrick & Mount, 1991; Schmidt &Hunter, 1998). Specifically, they found thataverage levels of general mental abilitywithin teams were positively related to per-formance. In addition, it was found thataverage levels of conscientiousness were alsopositively related. Thus, to some degree,organizations can use selection to improveteam performance in much the same waythat they enhance individual performance.

Although much less research has accu-mulated on team performance as comparedto individual performance, individual char-acteristics such as general mental ability andpersonality appear to explain much less var-iance in team than in individual perfor-mance. For example, Schmidt and Hunter(1998) found that general mental abilityexplained approximately 31% of the vari-ance in individual performance. In compar-ison, Barrick et al. (1998) found that averagegeneral mental ability explained approxi-mately 5% of the variance in team perfor-mance. Therefore, there are probably other


team composition factors that need to beconsidered when selecting people for team-work.

Recall that Campion et al. (1993) pro-posed and found evidence for a positiverelation between preference for teamworkand several team-level performance criteria.Thus, organizations selecting people forteams might be wise to consider whetherthese individuals prefer working in teamsettings. In Campion et al.’s study, therewas a reasonable amount of variance in pref-erence for teamwork, suggesting that peopledo indeed vary in the degree to which theyenjoy working as part of a team. However, apotential problem with using such a meas-ure, particularly with job applicants, is thatthey may not be truthful in reporting theirpreferences. That is, if they know that a jobrequires teamwork, they may report thatthey enjoy working in teams (even if theydo not) to enhance their chances of beinghired. Although little research to date hasaddressed this issue, Nagel (1999) found ev-idence, in a laboratory simulation, for thistype of distortion. One way to deal with thisissue might be to use other selection methods,such as personal history, to measure appli-cants’ affinity for working in team situations.

Instead of considering individuals’ pref-erences, selection efforts might be aimed atassessing the knowledge, skills, and abilities(KSAs) needed in order to work in teams.Stevens and Campion (1999) used this ap-proach to develop a selection instrument thatthey labeled the Teamwork Test. The specificKSAs measured by this instrument are pres-ented in Table 12.2. As can be seen, theseare teamed into two general categories: In-terpersonal KSAs and Self-ManagementKSAs. In the Interpersonal category, theKSAs deemed most important have to dowith conflict resolution, collaborative prob-lem solving, and communication. The Self-

Management KSAs have to do with goal set-ting/time management and planning/taskcoordination.

Stevens and Campion (1999) alsoattempted to provide predictive validity evi-dence for this instrument in two organiza-tions. Overall, their findings providedsupport for the predictive validity of thisinstrument. More specifically, their instru-ment explained variance in team perfor-mance beyond that explained by thegeneral mental ability of team members. Thissuggests that this instrument may ultimatelybe a useful tool for organizations, althoughsome further development may be needed.These findings are also interesting becausethey suggest that working in teams mayrequire certain unique skills and abilities thatmay not be assessed by traditional selectioninstruments.

Hirschfeld and his colleagues (2006)recently examined knowledge of teamworkprinciples as a predictor of effective team-work. These authors assessed how well em-ployees performed on a test designed toassess their knowledge of multiple teamworkconcepts and found that the average teamscore on the objective test was relatedto expert ratings of how effective theemployees worked together as a team. Thisresearch suggests that knowledge of team-work principles may be another criterion forselection.

Organizational Reward Systems

Nearly all of the models of team effectivenesscovered earlier in this chapter mentionedrewards as being an important factor contri-buting to team effectiveness. Quite simply,teams tend to be more effective when orga-nizations reward their efforts. Unfortunately,in many organizations where teams areutilized, the rewards are focused almost


exclusively at the individual level. In fact,even in organizations that do have team-levelrewards, these must coexist with individual-level reward systems. How then do organiza-tions make sure that individual- and team-level reward systems complement each other?

According to Wageman and Baker(1997), two important considerations inthe design of team-level reward systems aretask interdependence and outcome interde-pendence. As defined earlier, task interde-pendence reflects the extent to which team

members must work collaboratively in orderto get their work done. Outcome interde-pendence reflects the extent to which themembers of a team share a ‘‘common fate’’;that is, do they all receive the same positiveor negative outcome if the team performswell or poorly? For a team-level reward sys-tem to work well, there must be alignmentbetween task and outcome interdependence.Team-level reward systems work best whenboth task and outcome interdependence arehigh. If a great deal of interdependence is

TABLE 12.2Dimensions Comprising the Stevens and Campion (1999) Teamwork Test

I. Interpersonal KSAs

A. Conflict Resolution KSAs

1. The KSA to recognize and encourage desirable, but discourage undesirable team conflict.

2. The KSA to recognize the type and source of conflict confronting the team and implement and

appropriate resolution strategy.

3. The KSA to implement an integrative (win-win) negotiation strategy, rather than the traditional

distributive (win-lose) strategy.

B. Collaborative Problem Solving KSAs

4. The KSA to identify situations requiring participative problem solving and to utilize the proper degree

and type of participation.

5. The KSA to recognize the obstacles to collaborative group problem solving and implement proper

corrective actions.

C. Communication KSAs

6. The KSA to understand communication networks, and to utilize decentralized networks to enhance

communication where possible.

7. The KSA to communicate openly and supportively; that is, to send messages that are (a) behavior- or

event-oriented, (b) congruent, (c) validating, (d) conjunctive, and (e) owned.

8. The KSA to listen nonevaluatively and to appropriately use active listening techniques.

9. The KSA to maximize the consonance between nonverbal and verbal messages and to recognize and

interpret the nonverbal messages of others.

10. The KSA to engage in small talk and ritual greetings as a recognition of their importance.

II. Self-Management KSAs

D. Goal Setting and Performance Management KSAs

11. The KSA to establish specific, challenging, and accepted team goals.

12. The KSA to monitor, evaluate, and provide feedback on both overall team performance and individual

team-member performance.

E. Planning and Task Coordination KSAs

13. The KSA to coordinate and synchronize activities, information, and tasks among team members.

14. The KSA to help establish task and role assignments for individual team members and ensure proper

balancing of workload.

Source: M. J. Stevens and M. A. Campion. (1999). Staffing work teams: Development and validation of a selection test for

teamwork settings. Journal of Management, 25, 207–228. Copyright # 1999. Reprinted with permission from Elsevier Science.


required to perform a task, it stands to reasonthat the outcomes team members receiveshould be similar.

As Wageman (1996) points out, how-ever, team reward systems are oftenineffective because these two forms of inter-dependence are misaligned. Consider, forexample, a situation in which the work per-formed by team members is highly interde-pendent, but the outcomes they receive arenot. This will likely create disincentives forthe members of a team to work cooperativelybecause those who receive the highestrewards will be seen as being over-rewarded.On the other hand, what if there is very littletask interdependence, but a great deal ofoutcome interdependence? In a case likethis, those who perform well may resentsharing a common fate with those whoseperformance is far below theirs. The majorpoint is that organizations often fail to con-sider these two forms of interdependencewhen rewarding teams, and thus may inad-vertently create a disincentive for people towork as a team.

In Chapter 9 we discussed specific waysin which organizations may reward the per-formance of teams, so they will be mentionedonly briefly here. The most common of theseinclude profit sharing and gain sharing, al-though Employee Stock Ownership Plansmay also be used for this purpose. The spe-cifics of all of these compensation plans dif-fer, but they all share one commoncharacteristic: They are appropriate primar-ily in situations where employees in teamswork on highly interdependent tasks. Thus,when they are designing team reward sys-tems, organizations must pay particularattention to the nature of the work that teamsare performing.

Organizations can also attempt toenhance the performance of teams throughnon-monetary rewards such as awards and

recognition. Again, as with monetary com-pensation, it is important that the awardsand recognition that are provided to indi-viduals do not negate the effects of those atthe team level. In professional team sports,individual-level honors and recognition areoften viewed as more important than collec-tive achievement. After all, individuals areinducted to sports halls of fame, not teams.Thus, organizations not only need to rewardteam performance, but they must also createa climate in which collective achievement isviewed as being at least as important as indi-vidual accomplishments.

Team-Development Interventions

A final way that organizations may enhancethe effectiveness of teams is through training orteam-development activities. Team develop-ment will be discussed in much more detailin Chapter 16, so only a brief overview will beprovided here. According to Dyer (1987),team development represents a variety ofteam-based training interventions that areaimed at one or more crucial aspects of teamfunctioning. For example, team-developmentactivities may be aimed at clarifying roleswithin a team, establishing goals and prior-ities, and tackling more sensitive interpersonalissues (Beer, 1976; French & Bell, 1995).

Specific forms of team development maybe carried out quite differently, but mostinvolve some common generic steps. Forexample, in most cases, teams participate inteam-development activities with the assis-tance of an outside consultant or facilitator;outside, in this case, means someone fromoutside the team (this can be either someonewho is external to the organization, or some-one who is an organizational member).Compared to a team member, an individualfrom outside the team can be much moreobjective about the team and its process.


Another characteristic common to mostteam-development activities is that they aredata based; that is, some data about thefunction of the team are collected priorto the actual intervention. Typically, pre-intervention data are collected throughsurveys and interviews of team members,although other methods may also be used(e.g., archival data, observations). After dataare collected and summarized, they are usu-ally used as the basis for choosing the spe-cific team-development intervention. Forexample, if data indicate a lack of role clarityamong team members, the focus of teamdevelopment would typically be on role clar-ification (for an example, see Schaubroecket al., 1993).

While not a great deal of research hasexamined the impact of team developmenton team performance, there is some indica-tion that it may improve team effectiveness.For example, in a review of the work-teamliterature, Sundstrom, DeMeuse, and Futrell(1990) found that team-development activi-ties, particularly those focused on task-related issues, have a positive effect onteam performance. In contrast, when team-development activities are focused on inter-personal issues, the results are much lesspositive. This may be due to the fact thatthe underlying causes of interpersonal prob-lems within teams (e.g., personality andvalue differences) are much less amenableto changes, as compared to the causes oftask-related problems.

THE FUTURE OF TEAMSIN ORGANIZATIONS

Given the challenges associated with design-ing and managing teams, it might be tempt-ing to conclude that teams are just anotherfad that will pass with time. In the authors’opinion, this is not the case; teams are now a

permanent part of organizational life and willcontinue to be in the foreseeable future. Onereason for this is that more and more of thework being performed in organizations isbecoming complex and mentally demanding(i.e., ‘‘knowledge work’’). Often, teams canbe much more effective in tackling suchcomplex tasks because they offer variety inthe skills and knowledge of team members.

Another reason that teams will continueto flourish in organizations of the future isthat they are highly compatible with thechanging nature of employee–employerrelationships. Organizations are becomingleaner; they are relying less and less on for-mal job descriptions and more and more ontemporary employees. In this type of envi-ronment, it is very convenient for organiza-tions to retain a small core staff of permanentemployees and bring in temporary employ-ees on an as-needed basis. Furthermore,when temporary employees are brought intoan organization, it is often advantageous tocreate project teams to accomplish tasks.

Given this trend, a future challenge fororganizations is their ability to assembleteams and have them working effectivelywithin a very short time frame. This mayinvolve many of the factors discussed in thischapter, but those likely to be most impor-tant in the future are the characteristics of theindividuals comprising teams. More specifi-cally, in the future, it may be especially cru-cial for individuals to have a set of genericknowledge, skills, and abilities (Driskellet al., 2006; Stevens & Campion, 1999) thatfacilitates team effectiveness. This obviouslyraises the issue of where individuals willacquire these generic KSAs. They might betrained by organizations, but it is unlikelythat this would be effective if temporaryemployees are being used. Perhaps, inschools of the future, teamwork will be ascommon as reading, writing, and arithmetic.


CHAPTER SUMMARY

In this chapter, we focused specifically onthe issue of team effectiveness. In doing so,the relative emphasis shifted from predom-inantly laboratory-based social psychologicalresearch to the organizational psychologyteam literature. Although many definitionsof team effectiveness have been proposed,Hackman’s (1987) is the most well-accepted.This defines team effectiveness in terms of taskperformance, team viability, and the satisfac-tion of team members. We also gave a moreelaborate definition of teams offered byKozlowski and Bell (2003).

There have also been a number of team-effectiveness models; several were reviewedin the chapter. We began by examiningMcGrath’s model (1964) and ended up withCampion’s attempt to synthesize many team-effectiveness models. Although the specificsof the models described are different, a com-mon characteristic of most is the idea thatteam process is a key mediating variable.Nevertheless, the importance of team pro-cess has not been supported well empirically,and it is possible that causal sequences otherthan the familiar input–process–outputmay explain team effectiveness.

From the team-effectiveness models de-scribed, several specific determinants of teameffectiveness were chosen for further review.These included team composition, task design,reward systems, and team process. As withindividual performance, teams need memberswith relevant skills/abilities in order to be suc-cessful. However, research has also shown thatit is important to have the right mixture ofabilities/skills, and that personality and atti-tudes are important as well.

Task design is an important determinantof team effectiveness, for several reasons.Teams are often ineffective because they areperforming tasks that are not well suited for

teams; organizations often overlook this factin their zeal to utilize teams. Task design isimportant in determining the most importantstrategies for teams to use, and it may alsohave motivational effects on team members.

As with individual performance, rewardsare a very important determinant of teamperformance. Simply put, if organizationswant teams to perform well, reward systemsmust be designed so that the teams’ accom-plishments are rewarded. In designing teamreward systems, however, organizationsmust consider the level of both task andoutcome interdependence. It is also crucialthat the individual and team reward systemsdo not work against each other.

Team process is ubiquitous in the team-effectiveness literature. Ironically, despite itsimportance, there is a lack of consensus onwhat actually constitutes the most importantdimensions of team process. No attempt ismade to resolve this issue here. Instead wediscussed a number of behavioral and cog-nitive/affective processes that have beenlinked to team effectiveness. These includedcommunication, coordination, collabora-tion, cohesion, team efficacy, empowerment,and conflict.

Based on the determinants of team effec-tiveness, three general approaches toenhancing team effectiveness were offered:(a) selection, (b) reward system administra-tion, and (c) team-development activities. Aswith individual performance, organizationscan enhance the performance of teams byselecting highly skilled individuals. Selectionfor teamwork, however, may require a con-sideration of individuals’ preferences as wellas specific team-related skills. The most fun-damental issue with regard to reward admin-istration is simply to make sure that teamsare rewarded for their performance. Organi-zations must also make sure that team-levelreward systems fit the tasks that teams

Chapter Summary 407


EDUARDO SALAS AND HIS TEAMMATES ALONG THE WAY

Studying work teams takes a team. I’ve been

fortunate to have had (and continue to do so)

great teammates along my career. These team-

mates supported and encouraged each other;promoted creativity, innovation and could

resolve conflicts; cooperated and communi-

cated well (most of the time); took leadership

roles when needed; ‘‘shared a mental model’’

of what, how, why, and when to do required

research tasks. I’ve had teammates that were

interested in applying the best industrial/orga-

nizational and human factors science possibleto solve team performance problems and who

had a passion for thinking and doing team

research, but we all ultimately were driven

to have impact—impact on how to compose,

manage, train, develop or assess aviation,

medical, financial, military, oil, engineering

and industry teams.

All began at Old Dominion University inthe early 80’s when two of my advisors (Ben B.

Morgan, Jr. and Albert S. Glickman) got inter-

ested in seeking funds to study teams over time.

They got me interested in work teams. And the

race began. As soon as I left graduate school, I

was hired by the U.S. Navy in Orlando (then

the Naval Training Systems Center; now NAV-

AIR Orlando). My first assignment was to

develop a research program aimed at under-

standing, measuring and improving team per-formance. We worked first, with gunfire and

submarine teams, then aviation.1 We began to

look at aircrews and their coordination

demands. We outlined and tested a number

of principles on how to design and deliver

Crew Resource Management (CRM) training

(now a popular strategy used in commercial

aviation and medical settings). We developed aset of CRM training guidelines for the Navy to

implement across all aviation platforms. These

early successes gave us momentum. We were

also interested in getting information out—to

other work team scientists, team training

developers and practitioners. So, we sought

to publish our work widely and in different

forums.In the late 80’s we had our biggest oppor-

tunity and challenge. The ‘‘TADMUS Program’’

came along.2 The Tactical Decision Making

under Stress research program funded by the

Office of Naval Research launched a new era in

team research. It was an era of theory building;

of developing expert teams; of studying teams

in complex and ‘‘natural’’ settings; of develop-ing process and outcome metrics; and of

designing team training strategies. So concepts

like shared mental models; mutual perfor-

mance monitoring, back-up behavior, inter-

positional knowledge, anticipation rater, team

leader training, team self-correction training,

cross-training and many others surfaced in the

literature done to this program. This programmade contributions to theory, measurement,

and training of work teams. This was quite a

ride—where we, I believe, changed how peo-

ple in the Navy (and maybe beyond) looked at

work teams. It was a ride with ups and downs;

with joy and moments of not much joy; with


perform, and that they are aligned withindividual-level reward systems.

In this chapter, the final method ofenhancing team effectiveness was team de-velopment. Team-development activitiesrepresent training interventions that aredesigned to improve various aspects of teamfunctioning (e.g., definition of roles, settinggoals). Most team-development interven-tions utilize the services of an outside con-sultant/facilitator, and most are data based.Research has shown that team-developmentactivities may enhance team effectiveness. Itappears, however, that interventions focusedon task-related issues are more successfulthan those focused on interpersonal issues.


Alper, S., Tjosvold, D., & Law, K. S. (2000).Conflict management, efficacy, and perfor-mance in organizational teams. PersonnelPsychology, 53, 625–642.

Ilgen, D. R., Hollenbeck, J. R., Johnson, M.,& Jundt, D. (2005). Teams in organizations:From input–process–output models to IMOImodels. Annual Review of Psychology, 56,517–543.

Kozlowski, S. W. J. & Bell, B. S., (2003).Work groups and teams in organization. InW. C. Borman, D. R. Ilgen, & R. J. Klimoski(Eds.), Handbook of psychology: Industrialand organizational psychology (pp. 333–375). Hoboken, NJ: Wiley.

Marks, M. A., DeChurch, L. A., Mathieu,J. E., Panzer, F. J., & Alonso, A. (2005).Teamwork in multi-team systems. Journal ofApplied Psychology, 90(5), 964–971.

Salas, E., Sims, D. E., & Burke, C. S. (2005).Is there a ‘‘big five’’ in teamwork? Small GroupResearch, 36(5), 555–599.

Simons, T. L., & Peterson, R. S. (2000). Taskconflict and relationship conflict in top man-agement teams: The pivotal role of intrateamtrust. Journal of Applied Psychology, 85, 102–111.

growth and intellectual stimulation, and great

personal satisfaction.

As I left the Navy in 19993, a new world

opened up to me. And now I spend my time

thinking and researching about shared cogni-

tion, multicultural teamwork, simulation-

based team training, and team performance

measurement and diagnosis. I still work withreal work teams like coalition forces, in medical

settings, at NASA, the oil and financial teams. It

is a new ride.

So to study teams it takes a team—a team of

good, passionate, motivated, fun, and engaging

colleagues. And by the way, I’ve learned that

effective teamwork is complex, dynamic, elu-

sive, and difficult to maintain; that maintainingshared mental models is not trivial, but possi-

ble; the importance of theory and of robust

metrics; the value of translating what we know

into practical terms and of working closer with

the customer; and that you need great set of

teammates to study working teams; and that

you need to have fun practicing our science.

AndI’vehadsomegreatonesalongtheway . . .

Eduardo Salas


University of Central Florida

1My teammates were Carolyn Prince, Randy Oser, Renee

Stout, David Baker, Clint Bowers, Florian Jentsch and

more . . .2My teammates were Jan Cannon-Bowers, Joan

Johnston, Kim Smith-Jentsch, Carol Paris, Scott

Tannenbaum, Marvin Cohen, Bob McIntyre, Daniel

Serfaty, Nancy Cooke, Steve Koslowski and more . . .3My teammates were Shawn Burke, Steve Fiore,

and my students Katherine Wilson, Heather Priest,

Kevin Stagl, Dana Sims, Cameron Klein, Mike

Rosen and more . . .


Chapter Thirteen

OrganizationalTheory and DesignA

major theme throughout thisbook has been that the primaryfocus of organizational psychol-ogy is individual behavior. Con-sistent with this theme, we have

explored a number of factors that impact thebehavior of individual employees in organi-zations. Some of these (e.g., compensationprograms) are aimed directly at individuals;others (e.g., groupthink) reside primarily atthe group level. In this chapter, we raise thelevel of analysis one step higher and explorethe impact of organizational design on em-ployee behavior and overall organizationaleffectiveness.

For the most part organizational psychol-ogists have tended to shy away from macro-level issues such as organizational theory anddesign. As a result, much of the work in thisarea has been done by sociologists, as wellas those trained directly in organizationaltheory and other areas of management (e.g.,strategic management). In recent years, therehas also been work done by those in oper-ations research who mathematically modelorganizational scenarios (Kujacic & Bujovic,2003).

The reluctance of organizational psychol-ogists to tackle macro-level issues such asorganizational design is unfortunate for tworeasons. First, anyone who has worked inan organization knows that organizationalstructure does impact individual employees,although such influences are admittedly of-ten indirect. For example, in a universitysetting, the manner in which academic de-partments are grouped often has implica-tions for resource allocations and the extentto which faculty in different departments

collaborate. Thus, to ignore organizationaldesign is to ignore a very important impacton employee behavior (Bourrier, 2005;Hornstein & de Guerre, 2006), and ulti-mately the effectiveness of an organization(Russo & Harrison, 2005).

A second reason organizational psychol-ogists should not avoid macro-level issues isthat those trained in fields such as sociology,organizational theory, and strategic manage-ment often provide an incomplete picture ofthe impact of organizational design. Just asorganizational psychologists tend to focus(or, some would say, overfocus) on individ-ual behavior, the analyses provided by thosetrained in other fields often exclude individ-ual behavior. To say that the design of an

411

organization contributes to its effectivenessbegs the question: How, specifically, doesthe design of an organization impact thebehavior of those in the organization, anddoes it make the organization more, or less,effective? Therefore, an interdisciplinary ap-proach to organizational design that includesthe contributions of sociology, organiza-tional theory, strategic management, andorganizational psychology will be taken inthe present chapter.

The chapter will begin by discussing thebroad field of organizational theory. Organi-zational theories are simply different ap-proaches to organizing human endeavorssuch as work. Organizational theories areimportant because they form the basis forconcrete organizational designs. The focus ofthe chapter then shifts to a discussion of themajor factors that impact the decisions orga-nizations make with respect to organizationaldesign. The chapter will then examine recentinnovations in the design of organizations;these designs depart widely from traditionalorganizational theories. The chapter con-cludes with a summary of empirical researchon the impact of organizational design and adescription of factors that will likely impactthe design of organizations of the future.

WHAT IS AN ORGANIZATIONALTHEORY?

In the organizational sciences (e.g., organi-zational behavior, organizational psychol-ogy), one of the more misunderstood termsis organizational theory. To some, organiza-tional theory is a field of study; to others, it isthe process of using metaphorical languageto describe organizational processes (e.g.,McKenna & Wright, 1992; Morgan, 1986),or it represents an attempt to determine thebest way to organize work organizations. The

term is used to indicate all of these things,but an organizational theory is really just away of organizing purposeful human action.Given the diversity of purposeful humanendeavors, there are numerous ways toorganize them, and, hence, a great manyorganizational theories.

The academic discipline that really hastaken the lead in theorizing about the orga-nization of purposeful human behavior issociology. Sociology is essentially the studyof macro-level forces (e.g., social stratifica-tion, social institutions) on human behavior(e.g., Merton, 1968; Parsons, 1951). A clas-sic illustration of this was sociologist EmilDurkheim’s classic study of the determinantsof suicide (Durkheim, 1951). Durkheimproposed and found support for the ideathat some forms of social structure result inalienation and a sense of hopelessness (calledanomie), which ultimately leads to higherrates of suicide. Given this historical back-drop, it is natural that sociologists would bemore interested in the impact of the structureand design of organizations than psycholo-gists would be. In fact, the Hawthorne stud-ies, which are considered one of the mostimportant historical events contributing tothe development of organizational psychol-ogy, were conducted under the direction ofsociologist Elton Mayo.

Given that organizational theory dealswith different ways of organizing humanactivity, how does one theorize about organi-zations? In most scientific disciplines, if onewants to theorize about something and ulti-mately study it, the most common approachis to bring it into a laboratory for closerinspection. Unfortunately, organizationaltheorists cannot do this because organiza-tions are largely abstractions, and thuscannot easily be subjected to laboratoryinvestigations. Although we can draw elegantorganization charts to represent reporting

412 Organizational Theory and Design

relationships, and so on, what keeps an orga-nization together is the fact that an organiza-tion’s employees understand its design andadapt their behavior accordingly (Katz &Kahn, 1978).

Given these constraints, organizationaltheorists have been forced to adopt more in-direct methods of investigation. Perhaps themost common of these methods has beenthe use of metaphors to describe and under-stand organizational structures (McKenna& Wright, 1992). A metaphor is simply atype of figurative language; that is, usingwords or phrases to communicate anythingother than their literal meaning. For exam-ple, if a person were to look out the windowand say that it is ‘‘raining cats and dogs,’’ thatperson would be using a metaphor.

The most common metaphor used in theorganizational theory literature has been toliken organizations to biological organismsand apply general systems theory to describethem (Katz & Kahn, 1978; von Bertalanffy,1956). If we liken organizations to biologicalorganisms, this leads to several impor-tant conclusions that may have importantimplications for organizational functioning.Perhaps the most important of these is thatorganizations are in constant interactionwith the environment around them, just asany biological system must interact with thelarger ecosystem in which it is embedded.Like biological systems that ignore the largerecosystem, organizations that ignore theirlarger environment (e.g., customers, publicopinion) may risk extinction. Because of this,organizations often spend considerable timeand financial resources attempting to under-stand (and sometimes control) their externalenvironments.

A second major implication of the bio-logical organism metaphor is that an organi-zation consists of a series of subsystems thatmust work together in order for the organi-

zation to function optimally. As an analogy,the human body, which is a highly complexbiological system, consists of a number oforgan systems that control physiological ac-tivities such as circulation, digestion, andrespiration. Most of the coordination amongthese systems is controlled by a complexarray of chemical messengers that arereleased by the brain. In organizations, coor-dination among subsystems is often achievedthrough the structural arrangements them-selves, but, in some cases, special mecha-nisms (e.g., coordinating committees) areestablished for it.

Another metaphor that is quite commonin organizational theory is that of the organi-zation as a machine. Like machines, organi-zations take environmental input, transformthe input in some fashion, and return thatinput, in an altered form, back into the en-vironment. Many of the implications of theorganization as machine metaphor are quitesimilar to the organization as biological organ-ism metaphor. That is, organizations mustpay attention to the external environmentand to the coordination of its internal com-ponents. An additional implication of themachine metaphor, however, is the impor-tance of making sure that the componentsof the machine (i.e., people and processes)are performing properly. This is typicallyaccomplished in a number of ways, includingselection, performance appraisal, perfor-mance coaching, and, in some cases, redesignof jobs in order to maximize the unique capa-bilities of individuals.

Although the biological organism andmachine metaphors have the longest histo-ries in organizational theory, McKenna andWright (1992) point out that other meta-phors may also be useful to organizationaltheorists. For example, they point out thatorganizations can be likened to the brain—with all its complex interconnections—and

What is an Organizational Theory? 413

to families—with their complex relation-ships—and to political arenas—with all oftheir influence and power dynamics. Whilethese metaphors have yet to be used ex-tensively by organizational theorists, theyprobably reflect the realities of many organi-zations today. As we will see later in thechapter, the design of many organizationstoday is highly complex so these new meta-phors are needed.

MAJOR ORGANIZATIONALTHEORIES

Having provided a brief overview of the fieldof organizational theory, we now move on toa consideration of the major organizationaltheories themselves. Organizational theoriessimply represent ideas or models of the formin which human activity can be organized.Obviously, there could be an almost infinitenumber of ideas about organizing humanactivity; however, over the years, three gen-eral types of organizational theories havebeen developed. These three types are de-scribed next.

Classical Organizational Theories

Historically, the term classical organizationaltheory has been used to denote models oforganizing that were developed fromapproximately the early twentieth centuryuntil the mid-1940s. The best known ofthese classical theories were scientific man-agement (Taylor, 1911), ideal bureaucracy(Weber, 1947), and administrative manage-ment (Fayol, 1984). Each is discussed brieflyin the following paragraphs.

The term scientific management was firstintroduced in describing the history of orga-nizational psychology (see Chapter 1) andwas examined further in our discussion ofjob design (see Chapter 9). We resurrect this

term yet again because it also has implica-tions for the ways organizations are designed.As might be remembered from our earlierdiscussion, a fundamental principle of sci-entific management was that those who de-sign the work should be separate from thosewho actually perform the work. The impli-cations of this approach for organizing workare quite important. For example, it impliesthat there should be distinct status or hier-archical differences among employees. InTaylor’s writings, it is fairly obvious thatthose who design the work occupy a higherstatus than those who perform it. An orga-nization that is designed from a scientificmanagement perspective thus has many lev-els and many ways to distinguish amongthose levels.

Recall that another fundamental princi-ple of scientific management is that workshould be broken down into the smallestand simplest components possible. For ex-ample, the various steps involved in the pro-duction of an automobile would be brokendown into the simplest possible instructions.The implication of this emphasis on simpli-fication is that an organization should bestructured such that employees performingvery similar tasks should be grouped to-gether. Furthermore, in most cases, the bestway to manage these groups of employeesis to create departmental structures that arebased on these highly specialized activi-ties. Thus, an organization that is designedaccording to scientific management will con-sist of a large number of departments, witheach performing a highly specialized func-tion.

Beyond these more concrete organiza-tional design implications of scientific man-agement, there are also a number of lesstangible implications. Perhaps one of themost important of these is that a scientificallymanaged organization will have a great


number of rules and procedures for employ-ees to follow. Frederick Taylor becamefamous for using empirical research to deter-mine the most efficient way to carry out worktasks. The underlying assumption behindthis quest for efficiency was that there is‘‘one best way’’ to do any job or accomplishany task. The trick is to find it. This tendencyis likely to be generalized to other organiza-tional tasks; thus, such organizations arelikely to have thick policies and proceduresmanuals that have scripted procedures foremployees to follow in the event of any con-tingency.

The term ideal bureaucracy may strikesome as an oxymoron because the termbureaucracy is often used as a codeword fororganizational inefficiency and administra-tive inflexibility. In reality ideal bureaucracysimply represents one idea or theory of howhuman activity should be organized. Mostattribute the development of ideal bureauc-racy to Max Weber, who is generally creditedwith being one of the pioneers of the macroside of organizational psychology. Weber, asyou may recall, wore many intellectual hatsand made contributions to history, econom-ics, political science, and sociology duringhis lifetime. In Weber’s time, there were feworganizations in the form that we see themtoday. Instead, a great many ‘‘organizations’’of his era were loosely run family businesses,or they consisted of an individual craftsmanwho worked independently. Given theseorganizational forms, there was not a greatneed for organizing, per se.

With the advent of the Industrial Revo-lution, however, there arose a great need toorganize large numbers of people who hadleft rural areas to work in factories located inlarge cities. Furthermore, the organizationalmodels that were predominant in the latenineteenth and early twentieth centurieswere not adequate for handling these large

organizations. For example, at that time, agreat deal of one’s success was based onsocial connections or the nature of one’sfamily bloodline—not surprising, given thata good number of organizations of that erawere family run. The problem with thismodel being implemented in industrial set-tings was that those who possessed the bestsocial or family connections were not alwayscapable of performing the tasks required bythe work.

Weber proposed ideal bureaucracy as analternative that would result in more efficientoperations and more effective use of talent inorganizations. One of the primary assump-tions behind ideal bureaucracy (and one thatmany people forget) is a noble one: Rewardsshould be based on one’s contributions tothe organizations, as opposed to social orfamilial connections. Unfortunately, manyof the more negative assumptions underlyingbureaucracy have tended to overshadow thisone.

As in scientific management, there is, inideal bureaucracy, a strong reliance on pre-viously developed rules and procedures toguide behavior. In a truly efficient idealbureaucracy, there should be a rule or pro-cedure covering almost any situation thatemployees may encounter. This may oftenexplain why bureaucracies have so muchdifficulty when there is a highly novel situa-tion, and, in many cases, why people have anegative view of this organizational form.

Another hallmark of ideal bureaucracy isvery close supervision of employees. Bureau-cratic organizations typically are character-ized by very narrow spans of control; that is,each supervisor does not supervise a largenumber of employees. One thing is accom-plished by a narrow span of control: It iseasier for a supervisor to meet the needs ofhis or her subordinates. Answering ques-tions and helping to train 4 employees is


obviously much easier than doing the sametasks with 40. A narrow span of control alsomakes it far easier for supervisors to monitorthe behaviors of their subordinates. This hasled many to conclude that bureaucracy isbased on the assumption that employees willnot work unless their behaviors are tightlymonitored—not a particularly positive viewof human nature.

Another important principle of ideal bu-reaucracy is unity of command. In a bureau-cratic structure, each employee shouldideally have one (and only one) direct super-visor. For anyone prone to cynicism, a logicalconclusion would be that this principle, likethe narrow span of control, would make iteasier for an organization to control its em-ployees. Unity of command, however, mayhave some tangible positive benefits for em-ployees as well. If an employee reports toonly one supervisor, this reduces the oddsthat he or she would have to meet the com-bined and potentially conflicting expecta-tions of more than one individual. As willbe discussed later in this chapter, one of theproblems with some modern organizationaldesigns that violate this principle is that theyincrease role conflict among employees (e.g.,Joyce, 1986).

A third important principle of ideal bu-reaucracy is unity of direction. This meansthat information within a bureaucratic orga-nization flows in one direction—typically,from the top of the organization down tolower levels. The primary benefit of informa-tion flowing in one direction is that it in-creases predictability and stability. Thismakes everyone’s lives easier when vastamounts of information must be processedand also makes it much easier for the topmanagement of an organization to controlemployees’ access to information. Top man-agers can make sure that employees ‘‘knowonly what they need to know,’’ and nothing

more. This principle, like the others previ-ously discussed, makes it far easier for em-ployees to be controlled than would be thecase if information within an organizationflowed freely.

Ultimately, like any other organizationaltheory, ideal bureaucracy is neither goodnor bad. The extent to which it enhancesor detracts from organizational effectivenessdepends to a large extent on how it is imple-mented, and whether it is appropriate forthe organization’s environment. As will beshown later in the section on contingencytheories, there are some situations in whichbureaucracy is very appropriate. In othersituations, however, organizations that useit are highly prone to failure.

The term administrative management wasfirst coined by Henri Fayol, who was an en-gineer by training and eventually became thechief executive of a French mining company.Fayol (1984) sought to develop a relativelyuniversal set of organizing principles formanagers to apply in organizations. To givethese principles meaning, however, Fayolpresented them in the context of the activ-ities of managers, or behaviors he calledmanagement functions. According to Fayol,the major functions of managers includedplanning, organizing, commanding, coordi-nation, and controlling. The principles thatFayol proposed were designed to assist man-agers in carrying out these essential func-tions.

Table 13.1 presents Fayol’s 14 organizingprinciples. As can be seen, many items onthis list are similar to the principles of idealbureaucracy that were described earlier. Forexample, Fayol advocated, among other bu-reaucratic principles, the division of work,having a well-defined authority structure,unity of command, unity of direction, order,and equity in the way individuals are re-warded. Among the unique principles that


Fayol added were stability of personnel,encouraging people to take initiative, andhaving a high level of cohesion and camara-derie among employees. These three addi-tions are interesting, simply because they donot seem to be in line with those of idealbureaucracy. In fact, they seem more com-patible with organizational theories that weredeveloped due to perceived limitations ofbureaucracy.

Although many of Fayol’s principles areuseful, and in fact put into practice in manyorganizations even today, they have also beenwidely criticized on a number of counts.Perhaps the most vigorous criticism of theseprinciples, and classical theories in general, isthat they ignore the ‘‘human element’’ inorganizations (e.g., McGregor, 1960). Thatis, these principles paint a picture of employ-ees in organizations as being interchangeable

cogs in a grand machine, rather than humanswith emotions, desires, and creative talents.The other major criticism of Fayol’s princi-ples has been that they are simply too gen-eral. To take one example, Fayol advocatesfairness in the remuneration of employees,which most managers would likely agree is adesirable principle. However, even thoughhe does offer some suggestions on how toaccomplish this goal (Fayol, 1984), thesesuggestions are rather vague. The same prob-lem plagues many of the other principles aswell.

Humanistic Organizational Theories

Recall from Chapter 1 that the Human Rela-tions movement began in the 1940s, largelyas a response to organizations being designedaccording to the classical theories described

TABLE 13.1Fayol’s Classic Principles of Organizing

Principle Fayol’s Comments

1. Division of work Individuals and managers work on the same part or task.

2. Authority and responsibility Authority: right to give orders; power to exact obedience; goes with

responsibility for reward and punishment.

3. Discipline Obedience, application, energy, behavior. Agreement between firm and

individual.

4. Unity of command Orders received from one supervisor.

5. Unity of direction One head and one plan for activities with the same objective.

6. Subordination of individual

interest to general interest

Organizational objectives come before the objectives of the individual.

7. Remuneration of personnel Fairness of pay to the organization and the individual; discussed various

forms.

8. Centralization Amount of discretion held by the manager compared to that allowed to

subordinates.

9. Scalar chain Line of authority from lowest to top.

10. Order A place for everyone, and everyone in his or her place.

11. Equity Emphasis on kindness and justice.

12. Stability of tenure of personnel Ability in managerial ranges; time to adapt to work.

13. Initiative Power of thinking out and executing a plan.

14. Esprit de corps Harmony and union among personnel is strength.

Adapted from: From H. Fayol. (1984). General and industrial management. Belmont, CA: Lake Publishing.


in the previous section. According to those atthe forefront of the Human Relations move-ment, most notably Douglas McGregor andRensis Likert, classically designed organiza-tions dehumanized the workplace and vastlyunderutilized the creativity and initiativeof employees. The primary reason for thisnegative impact of classical theory can betraced largely to its underlying assumptionsabout human nature. With their strongemphasis on order and control, classicalorganizational theories are based on theassumption that employees will not workunless they are prodded to do so, and theylack the creativity and initiative to define theirroles on their own. The best known of thesehumanistic organizational theories wereMcGregor’s Theory X/Y Leadership Distinc-tion and Likert’s concept of The HumanOrganization.

In his 1960 book, The Human Side ofEnterprise, McGregor made a distinction be-tween two types of managers: Theory X andTheory Y. Managers characterized as TheoryX operate under the assumption that mostpeople have an inherent dislike for workand, as a result, need to be coerced andsupervised very closely if they are to worktoward the goals of the organization. Anotherfundamental assumption of the Theory Xmanager is that people have little ambition,are not self-directed, and value security overall else. This again suggests the need for highlevels of managerial control and close super-vision of employees.

In direct contrast to the Theory X man-ager, the Theory Y manager operates underthe assumption that work is a natural part ofpeoples’ lives, and rather than avoid it, mostpeople seek greater meaning in it. As a result,individuals are capable of some degree ofself-control and will work toward the goalsof the organization to the extent thatthey find doing so to be personally reward-

ing. Another fundamental Theory Y assump-tion is that, under the right conditions,many people will seek out responsibilityand will creatively solve organizationalproblems if they are allowed to do so. A finalkey assumption of Theory Y is that mostorganizations are designed in a way thatresults in the underutilization of employees’skills and talents. Given that classical orga-nizational designs were still dominant at thetime McGregor wrote this book, this was anobvious criticism of classical organizationaltheory.

Although the Theory X/Y distinction ismade at the individual manager level andthus is not technically a theory of organizing,it is not difficult to extend these ideas to theorganizational level. An organization that ispopulated with Theory X managers is likelyto have very narrow spans of control, strictlines of authority, and a vast number of rulesand procedures—in short, the design wouldbe based largely on classical organizationaldesign principles. The Theory Y organiza-tion, in contrast, would include wider spansof control, more flexible lines of authority,and rules only for very important issues.These are logical design features if oneassumes that employees are capable of doingtheir work without close supervision, andthat they are bright enough to engage increative problem solving when novel prob-lems occur.

The primary benefit of the Theory Yorganization is that it is inherently more hu-mane and potentially more psychologicallyfulfilling for employees than Theory X orga-nizations. Within such organizations, workcan be a source of personal growth ratherthan a ‘‘necessary evil’’ that one must endurein order to make a living. There are, how-ever, potential downsides to the Theory Yorganization. Specifically, many have arguedthat it is rather naıve to assume that all


employees want to achieve personal growththrough their work experiences (e.g., Hack-man & Oldham, 1980), and that everyonecan work productively with only minimalsupervision. Another potential drawbackassociated with the Theory Y organizationis that because lines of authority are not asclear as in classically designed organizations,there is greater potential for confusion andconflict regarding employees’ roles and re-sponsibilities.

The other organizational theory that hascome to be seen as highly representative ofthe humanistic perspective is Likert’s idea ofthe human organization. In his 1961 book,New Patterns of Management, Likert proposedthat organizations could be classified intofour different types, which corresponded toSystem 1 through System 4. A System 1organization was described by Likert as anexploitive authoritarian type. This type oforganization is very similar to Theory Xbecause it is characterized as having verylittle trust in employees, little communica-tion between employees and management,very centralized decision making, and con-trol achieved in a very ‘‘top-down’’ manner.This type of organization, according to Lik-ert, would result in largely dissatisfied em-ployees and, ultimately, a low level oforganizational performance.

The System 2 organization, which waslabeled benevolent authoritative, was pro-posed to be much like the System 1 organi-zation, although some notable differenceswere present. For example, in this type oforganization, there is some level of trust inemployees and, at times, management usestheir ideas. There is also more communica-tion in this type of organization and, onoccasion, employees have an opportunityto communicate their ideas to upper man-agement. To sum up this type, employees arestill treated in a largely authoritative manner,

but the organization is a bit nicer to them.According to Likert, employees in this typeof organization may derive some moderatelevel of satisfaction toward their work andorganizational performance may be ‘‘fair togood.’’

The System 3 organization was labeledconsultative by Likert. This type of organiza-tion is very different from System 1 in thatthere is substantially greater trust in employ-ees and their ideas are used a great deal more.In addition, there is more overall communi-cation in such an organization and, com-pared to the other organizational forms,much more of it flows from the bottom up.Decision making is still primarily in thehands of those at higher organizational lev-els, but the manner in which this authority isexerted is different than in the System 1 andSystem 2 organizations. In System 3 organi-zations those at the top of the organizationset broad policies, and more specific opera-tional decisions are made by those at lowerlevels. Employees in the consultative organi-zation work on goals that they adopt aftersome discussion, and, at times, they mayresist organizational goals. Some level of con-trol resides at lower organizational levels andthe information that flows within the orga-nization is accurate. According to Likert, theconsultative organization is capable of‘‘good’’ performance, although it may neverreach a level of extremely high excellence.

System 4, the final type of organization inLikert’s typology, is labeled participativegroup. This type of organization is essentiallythe complete opposite of the System 1 orexploitive authoritarian organization. For ex-ample, in this type of organization, managershave complete trust in subordinates and, as aresult, are always seeking their input beforemaking decisions. Communication in thistype of organization is free-flowing in alldirections, and there is often a great reliance


on teamwork. As might be expected, deci-sion making occurs at all organizational lev-els and with a high level of involvement.Employees in the participative group organi-zation work on goals that they have a sayin developing and, as a result, there is agreat deal of acceptance of them. Controlmechanisms are applied at all levels of theorganization, and the information availableto employees is complete and accurate. Ac-cording to Likert, the participative grouporganization is the only one of the four thatis truly capable of ‘‘excellent’’ performance.Therefore, it is probably not a coincidencethat highly successful organizations (e.g.,Peters & Waterman, 1982) often havemuch in common with Likert’s System 4organization.

Likert (1961) originally proposed fourorganizational types. More recently, J. Likertand Araki (1986) proposed that there is also aSystem 5. The System 5 organization is essen-tially identical to System 4, but differs in oneimportant respect. Specifically, in a System 5organization, leadership is truly a sharedenterprise. Essentially, the organization hasno ‘‘bosses.’’ This model is common in uni-versities where academic departments areessentially self-managed. Outside of academiathis model is rare, although it may eventuallybecome more prevalent in the future.

Although many would agree that the Sys-tem 4 or 5 organizations are most likely to besuccessful, it is questionable whether thiswould always be the case. For example, hav-ing completely free-flowing information maybe an advantage in some ways, but it also maylead to problems such as information over-flow or distortion. Similarly, having completetrust in employees may be a good policy,provided people are in fact trustworthy. Fur-thermore, in the more recently developedSystem 5 organization, although the absenceof bosses may contribute to an egalitarian

atmosphere, it may also be problematic attimes when someone needs to take charge.

The overall point is that there is muchabout McGregor’s Theory Y or Likert’s Sys-tem 4 or 5 organizations that can lead toorganizational effectiveness. However, themajor weakness in these organizational the-ories is the notion that there is one mosteffective way to run an organization. Thatwas essentially the problem with classicalorganizational theories. This dissatisfactionwith the ‘‘one best way’’ approach to orga-nizations led to the development of contin-gency theories of organization.

Contingency Organizational Theories

Recall from Chapter 10 that contingencytheories of leadership developed largelybecause leadership theorists recognized thatthere is no singular set of personal traits orone set of behaviors that will always distin-guish good leaders from poor ones. Simi-larly, organizational theorists graduallycame to the realization that neither a classicnor a humanistic organizational form wasappropriate all of the time. Thus, the basicpremise of contingency organizational theo-ries is that design of an organization must beconsistent with the situation (Lawrence &Lorsch, 1967).

More recently organizational scholars la-beled this the congruence perspective (Nadler& Tuchman 1992; 1997). The idea, muchlike the early contingency approach, is thatorganizations should be designed in waysthat are congruent not only with an organi-zation’s strategy, but with other factors aswell, including the work that an organiza-tion does, characteristics of the individualswithin the organization, and the informalprocesses within the organization. Given thisperspective, the obvious question thenbecomes: What specific internal and external


factors do organizations need to pay mostattention to when making organizational de-sign decisions? In the following section, wediscuss five factors that have most often beencited as driving organizational design deci-sions.

DETERMINANTS OFORGANIZATIONAL DESIGN

Organizational design decisions should bemade on some rational basis; that is, the de-sign of an organization should serve somepurpose. Keeping this in mind, there aremany purposes that organizational designdecisions can serve. Most commonly, designdecisions are made to support an organiza-tion’s strategy, to help it cope with environ-mental uncertainty, to reflect the beliefs andassumptions of those in power, and to sup-port its core technology—in some cases,however, design is essentially preordainedbecause of the size of an organization. Eachof these factors is discussed next.

Strategy

An organization’s strategy consists of itslong-range goals and the tactics it uses toreach those goals. Strategic differences inorganizations have a number of implicationsfor organizational issues such as staffing,compensation and reward systems, and per-formance appraisals. According to Galbraith(1995), the aspects of organizational struc-ture that tend to have the greatest impact onstrategy implementation include the level ofspecialization, the shape of the organization,the distribution of power within the organi-zation, and the departmental structure. Spe-cialization refers simply to the types andnumbers of specialties that an organizationuses in performing its work. When organi-zations have a high degree of specialization,

this tends to improve the performance ofvarious subtasks—excellence in engineeringis more likely when there is a high concen-tration of engineers. Unfortunately, highspecialization also makes it more difficultto integrate all of the different specialties. Ifan organization’s strategy is to produce ahighly specialized product—and produce itvery consistently over time—then high spe-cialization is preferred. On the other hand, ifan organization pursues a strategy thatinvolves changing products quickly in re-sponse to consumer demand, then a highlevel of specialization is a liability becauseit does not allow an organization to reactquickly to the market. For many organiza-tions the trend in recent years has been toveer away from specialization, largelybecause more and more organizations arepursuing strategies that require quick re-sponses to changes in market conditions,and because many highly specialized taskscan be automated.

A second organizational design factor im-pacting strategy implementation is the shapeof the organization. According to Galbraith(1995), the shape of an organization is re-flected by the number of people who formdepartments at each hierarchical level. A nar-row organization is one that has a relativelylarge number of levels but relatively fewindividuals at each level. In contrast, a veryflat organization is one that has a relativelyfew number of levels, but each level is com-prised of a large number of employees.Figure 13.1 highlights this distinction.

From a strategic point of view, the shapeof an organization has some important con-sequences. When organizations are narrow,a great deal of time and energy is typicallyinvested in communication, supervision,and, especially, decision making. Stated dif-ferently, a narrow organization expends agreat deal of resources on ‘‘running itself.’’


This time spent on internal issues obviouslytakes time away from externally focused ac-tivities such as interfacing with customers ordetermining trends in the competitive mar-ket. As a result, a narrow organization ismost appropriate for an organization pursu-ing a strategy that involves producing highlyspecialized products or services in marketswhere the demand is fairly consistent.

A very flat organization, in contrast,spends far less time on internal processessuch as supervision and decision making.With such large spans of control, these typesof organizations often rely on teams or co-ordinating committees to provide typicalsupervisory functions. Also, with a flatter or-ganizational structure, decisions can bemade without having to go through manyorganizational layers. Flatter organizationalstructures thus allow much quicker re-sponses to consumer demand or the pene-tration of highly volatile market segments.

This is likely the reason why organizations innew or emerging sectors of the economy (e.g.,Internet-based commerce) often have flatorganizational structures (Sine, Mitsuhashui,& Kirsch, 2006).

A third organizational design principlethat impacts strategy implementation is dis-tribution of power within an organization.As Galbraith (1995) notes, the distribution ofpower in organizations is typically reflectedin the manner in which decisions are made.This can be seen both vertically and horizon-tally. Vertical distribution of decision-makingpower reflects the extent to which the deci-sion making in an organization is centralizedversus decentralized. Centralization wouldbe present if the headquarters of an organi-zation made all important decisions; de-centralization would be present if decisionmaking was pushed down to the level ofthose who either produce the product ordirectly interact with the customers.

FIGURE 13.1Contrast between a Flat and Narrow Organizational Structure

Vice-PresidentAdministration

Vice-PresidentMarketing

StaffAccountant

SalesTeams

President

FlatOrganizational

Structure

Vice-PresidentAdministration


MarketResearchManager

SalesManager

CustomerRelationsManager

MarketResearchAnalysts

SalesTeams

CustomerService

Representatives

President

NarrowOrganizational

Structure


Horizontal distribution of power is re-flected in whether managers shift decision-making power to the department or unit thathas the best information or is in the bestposition to make a decision. An example ofhorizontal decision-making power that isbecoming increasingly common is organiza-tions’ shift of decision making to unitsthat have the most customer contact, partic-ular in highly volatile customer-drivenindustries.

An organization that concentrates powerin the hands of a few top managers is likely tohave difficulty implementing a strategy thatinvolves quick responses to market condi-tions. When decisions need to be made veryquickly—often, by those who are close toconsumers (e.g., salespeople, customer serv-ice employees)—this type of organizationalstructure is simply inefficient. Similarly, anorganization that is very hesitant to sharepower horizontally is often unable to takeadvantage of cost-saving opportunities asso-ciated with contracting out aspects of theirbusiness. They simply lack the knowledge totake advantage of such opportunities.

The final aspect of structure that willimpact strategy implementation is an orga-nization’s departmental structure. Accordingto Galbraith (1995), departments may beformed according to a number of criteria,including function, product line, customersegments, geographical areas, or work-flowprocesses. Traditionally, departmentalizationby function (e.g., marketing, production,human resources) has been the most com-mon form in organizations (see Fig. 13.2).The primary advantage of this form of de-partmentalization is that it promotes a highlevel of specialization and, as a result, a highlevel of excellence in each particular func-tion. This type of structure, for example,makes it is easier for groups of product engi-neers to exchange ideas that will potentially

lead to new and creative products (Daman-pour, 1991).

Despite this benefit, a departmentalstructure based on functional specializationhas two primary weaknesses. First, this typeof organizational structure is very good if anorganization produces a single product or avery limited number of products or services.However, as the number of products or serv-ices increases, managers who lead functionalteams can easily become overwhelmedbecause of the complexity that is introduced.A second disadvantage of this type of struc-ture is that it does not promote a great deal ofcross-fertilization of ideas among functionalgroups. In this type of structure employeesmay become very compartmentalized andthus not take advantage of the ideas of otherspecialties in the development of new prod-ucts or services. This is often the reason thatuniversities, most of which have departmen-tal structures based on academic disciplines,tend not to have a great deal of collaborationacross departments.

FIGURE 13.2Example of a Functional Department Structure

Finance

Researchand

Development

ProductMarketing

HumanResources

Operations

GeneralManager

Source: J. R. Galbraith. (1995). Designing organizations: An

executive briefing on strategy, structure, and process. San

Francisco: Jossey-Bass. Copyright # 1995 Jossey-Bass.

Reprinted by permission of John Wiley & Sons, Inc.


In terms of strategy, a functional struc-ture is likely to work best in a relatively stableenvironment—one in which technologicalchange is not rapid. Organizations withfunctional structures have a difficult time inindustries where competitive advantage isdetermined largely by the speed with whichorganizations are able to respond to marketdemands.

A second basis on which many currentorganizations create departmental structuresis the products or services offered.Figure 13.3 provides an illustration of howsuch an organization might look. Notice thatimmediately below the CEO are the financeand human resources departments. Thedepartments at the next level, however, cor-respond to the various products or servicesin which this organization specializes (e.g.,electronic instruments, medical instruments,computers). In this type of structure, thefunctional specialties that were describedearlier are included under each of the prod-uct groupings. That is, there would be a

marketing group for electronic instruments,another for medical instruments, and so on.

A product-based departmental structureis an advantage if an organization’s strategyis to penetrate many markets or to have ahighly diverse set of products. As stated ear-lier, a functional departmental structurewould be quickly overwhelmed if an organi-zation pursued this type of strategy. Accord-ing to Galbraith (1995), however, there aretwo principal disadvantages associated witha product-based departmental structure.First, general product managers often tendto think of themselves as entrepreneurs andthus want considerable autonomy and inde-pendence. Although this may be a way tofoster innovation (e.g., Puranam, Singh, &Zollo, 2006), it also can result in an organi-zation’s consistently ‘‘reinventing the wheel’’because these individuals act independentlyof each other.

A second and related problem is that it isdifficult to achieve functional ‘‘economies ofscale’’ that benefit the organization as awhole in product-based structures. Note inFigure 13.3 that there is a marketing depart-ment for each of this organization’s threeproduct lines. While beneficial in someways, this can also be wasteful because itprecludes the organization’s taking advan-tage of complimentary approaches to mar-keting or advertising. In effect, eachmarketing department focuses on its ownproduct and may not have the larger interestsof the organization as its first priority.

A third departmental structure proposedby Galbraith (1995) is based on the variousmarkets that an organization serves. Thismay lead to a departmental structure basedon the customers the organization serves orthe various industries in which the organi-zation operates. Market-driven departmentalstructures are becoming increasingly popu-lar, given the decline in large-scale heavy

FIGURE 13.3Example of a Product-Based Structure

Finance

ElectronicInstruments Computers

HumanResources

MedicalInstruments

CEO






manufacturing and the increasing numberof firms in the service sector. This type ofstructure is particularly advantageous inthe service sector because it enables theorganization to react very quickly when, forsome reason, customers’ preferences change.The primary concern, as with the product-based structure, is that organizations maywaste a considerable amount of time dupli-cating functional activities within each mar-ket.

A fourth type of departmental structure isbased on geographic location. For example,an organization may create a ‘‘Northeast’’division that is responsible for all operationsin the New England states and a ‘‘Midwest’’division that is concerned with all businessin the Great Lakes region. Most typically,geographical structures are found in indus-tries where service is provided on-site andregional differences may be important tothe business. Fast-food companies such asMcDonald’s or Burger King are organized inthis fashion. The primary service is providedon-site (you can’t e-mail someone a BigMac!), and there may be important regionaldifferences in food preferences.

The primary advantage of a geographicstructure is that organizations can moreeasily provide personal services to customersand make adjustments to those servicesbased on regional preferences. A potentialdisadvantage of this structure again, how-ever, is duplication of functional activities.Also, geographical dispersion makes itmore difficult to maintain consistency andcompany-wide quality standards. Recentadvances in communication technology havemade consistent quality somewhat less of aconcern, but it still remains an issue formany products and services.

A final way that organizations may struc-ture their departments is according to majorwork-flow processes. According to Galbraith

(1995), this is the newest type of departmen-tal structure. Figure 13.4 shows an exampleof a departmental structure that is processbased. Notice that the three departments inthis organizational structure correspond tothe processes of new product development,order fulfillment, and customer acquisition/maintenance. Functional groups wouldreside within each of these process-relateddepartments.

When compared to the functional orga-nizational structure, the process-based struc-ture has one advantage: There is potentiallya lower amount of duplication of effortbecause there is less need to duplicate func-tions across departments responsible fordifferent processes. For example, customerservice personnel would reside exclusively inthe customer acquisition and maintenancedepartment, so these efforts would not beduplicated across departments. This struc-ture may also force an organization to take ahard look at its major work-flow process; infact, this type of structure cannot be createdunless an organization does so. By looking at

FIGURE 13.4Example of a Work-Flow Process Structure

New ProductTeams

ProductTeams

CustomerTeams

CustomerAcquisition and

Maintenance

OrderFulfillment

Process

New ProductDevelopment

Process

GeneralManager






its major processes, an organization may gainvaluable insights into how to become moreeffective and efficient. For example, bydeveloping a greater understanding of itsdistribution systems, a consumer productcompany may be able to adopt a strategy ofreducing costs by maintaining smaller inven-tories.

Level of Environmental Uncertainty

Organizations differ greatly in the degree ofenvironmental uncertainty they must face.One obvious factor that impacts environ-mental uncertainty is the number of yearsan organization has been in existence. Newor start-up organizations typically face ahighly uncertain environment as they try tosurvive (Aldrich, 1999). Conversely, organi-zations that have been in existence for manyyears have an established track record andconsequently a higher level of certainty.

Another important factor impactinguncertainty is the level of competition anorganization faces. An organization operat-ing in an industry with few competitors willface a far more certain environment com-pared to one operating in an industry withmany competitors. A related factor impact-ing uncertainty is the degree of stability in anorganization’s competitive environment.Some organizations produce goods and serv-ices that have rather stable demand; forothers, the demand may fluctuate verywidely. A third factor impacting an organi-zation is the extent to which its markets areaffected by external factors such as govern-ment regulations. Examples of very stableenvironments would include governmentand regulated utilities. Examples of veryuncertain market sectors would include hightechnology, consumer products, and trans-portation.

Uncertainty is related to organizationaldesign because the degree of uncertaintyoften impacts the speed at which organiza-tions must adapt to external conditions. Acomputer software company, for example,must be constantly prepared to bring newproducts to market and to satisfy the rapidchanges in technology. A regulated utility,on the other hand, typically does not have toadapt nearly as quickly because it has muchless volatility in its environment. Given thisconsideration, as a general rule organizationsthat operate in highly uncertain environ-ments need organizational structures thatallow them to act quickly when changeoccurs (Burns & Stalker, 1961).

Based on the organizational theories thathave been discussed up to this point, itseems fairly clear that highly bureaucraticorganizational structures are not very adapt-able in highly uncertain environments(O’Toole & Meier, 2003). Having a largenumber of rules and procedures, clear linesof authority, and small spans of control maykeep things humming along very smoothlyin stable environments. When thingschange, however, it is very difficult for highlybureaucratic organizations to change coursequickly. More humanistic organizations faremuch better in uncertain environments. Inthese types of organizations, power is shiftedto lower levels, and employees at these levelsare empowered to make decisions withoutfirst having lengthy deliberations with supe-riors.

Recent research, however, has shownthat the relationship between organizationalstructure and environmental uncertaintymay be more complicated than Burns andStalker (1961) originally proposed. Since,Mitsuhashui, and Kirsch (2006) found thathaving a more humanistic (or organic) orga-nizational structure was not beneficial for allorganizations operating in highly turbulent


emerging segments of the economy. Specif-ically, they found that new organizationsworking in highly turbulent environmentsbenefited from many attributes of bureau-cratic organizing (e.g., rules, administrativecontrol). This suggests that start-up organi-zations, which are typically formed by entre-preneurs not professional managers, stillneed to pay close attention to organizationaland administrative issues.

Beliefs and Assumptions of Thosein Power

Organizations are created by people; there-fore, at some level organizations typicallyreflect the beliefs and assumptions of peoplewho hold positions of power. In some casesthis would be the organizational founder,but it could also be those in high-level posi-tions. What specific beliefs of such individ-uals would impact organizational design?One of the fundamental judgments thatmanagers in organizations must make iswhether their employees can be trusted(Dansereau, Graen, & Haga, 1975). Organi-zations that are run by managers who placelittle trust in their employees will likely favororganizational designs that allow for highlevels of managerial control. These designsoffer very small spans of control, limited andhighly centralized communication, andother mechanisms that discourage indepen-dent action on the part of employees.

A related issue is the extent to whichmanagers respect their employees’ job-related skills or believe in their competence(Liden & Maslyn, 1998). This is obviouslyrelated to trust but is not the same thing. Forexample, a manager may believe that his orher employees are trustworthy people, yet heor she may have little confidence in theirskills. Managers may also see some employ-ees as highly competent, yet have little trust

in them. Organizations run by managerswho have little confidence in their employeeswill probably be designed to resemble orga-nizations where there is low trust. The sim-ilar designs achieve the same goal: a highlevel of control over employees. Conversely,when managers in organizations have a greatdeal of confidence in employees, it is morelikely that the organizational design willallow employees to use their unique skillsand abilities. Organizational arrangementsmay allow employees to exercise indepen-dent judgment and decision making, and tocommunicate freely with each other. Fur-thermore, in some cases, organizations cre-ate designs that specifically facilitatecreativity (Galbraith, 1995). An organizationwith a long history of doing this is 3M (seeComment 13.1).

A final belief or assumption of those inpower that may impact organizational designis the level of organizational performancethat is expected or desired. As we saw earlier,Likert (1961) proposed that there is a strongconnection between the desire to maintainmanagerial control and organizational per-formance. In the System 1 organization, man-agers maintain very high levels of control, yetthe price they pay for this control is high—such organizations will likely only be medio-cre. In contrast, in the System 4 and System 5organizations, managers give up a good dealof control, which is risky. However, the re-wards that go along with doing so are poten-tially very high because theses organizationsare capable of truly excellent performance.

Although it seems illogical, there may bereasons why managers in an organizationwould be comfortable with minimal perfor-mance. For example, managerial compensa-tion packages often provide disincentives fortaking steps to create long-term organiza-tional excellence. Specifically, stock optionplans often reward managers for taking steps


to boost short-term stock prices (e.g., layoffs,cuts in research and development expendi-tures) at the expense of long-term excellence(Gomez-Mejia, 1994). Thus, managers maysacrifice long-term success for short-termgains.

Another potential reason has to do withthe culture of the organization. According toPeters and O’Connor (1988), over time, many

organizations develop a ‘‘culture of justifica-tion’’: Employees’ primary concern becomesjustifying minimal levels of performance. Insuch an organizational culture, high levels ofperformance are not rewarded; in fact, theymay actually be punished. As a result, overtime, those who strive for excellence eitherleave the organization or simply retreat to aminimal level of performance themselves. A

COMMENT 13.1

THE 3M PATH TO INNOVATION

THE MINNESOTA MINING and Manufacturing

Company, more commonly known as 3M,

was founded in 1902 by five businessmen

whose original intent was to mine the mineralcorundum and sell it to grinding-wheel man-

ufacturers. The manufacturers needed this

mineral to make abrasive materials such as

sandpaper. Unfortunately, the mineral found

in the mine they purchased in Crystal Bay,

Minnesota, contained another material that

was not suitable for abrasives; thus, they were

not able to carry out their original plan. Thecompany ultimately decided to manufacture

abrasives themselves and, after several years of

struggle, expanded into the production of

adhesives. Ultimately, as most readers are well

aware, 3M became the leader in both indus-

tries. Today, 3M is one of the most successful

corporations in existence. Annual sales are

approximately $15 billion, half of whichcomes from outside the United States.

By any measure, 3M’s credentials as an

innovative company are impressive. For exam-

ple, 30% of the company’s sales come from

products that have been introduced within the

past 4 years. In 1998 alone, 3M gained 611

patents, and it consistently ranks in the top 10

U.S. companies in patents granted. How dothey do it? One obvious factor is their recruit-

ment and hiring of top scientific talent. An-

other factor that has been widely cited is 3M’s

long-standing ‘‘15% rule’’ among its research

personnel. More specifically, researchers areencouraged to spend up to 15% of their time on

projects that are based purely on their own

interests. The most visible outcome of the 15%

rule is the ubiquitous ‘‘post-it’’1 note. Many

other product innovations have resulted from

this policy as well.

Another major factor in 3M’s success as an

innovator is its organizational structure. Re-search personnel are organized into Corporate

Laboratories, Divisional Laboratories, and

Technology Centers. Researchers in each of

these areas are focused on somewhat different

activities (e.g., ‘‘basic’’ research, modification

of existing products and technologies), but

they are encouraged to communicate freely

with each other and with customers. This typeof structure facilitates the sense of scientific

community that is needed for innovation and

the probability that innovations have market

potential.

Source: E. Gundling. (2000). The 3M way to innovation:

Balancing people and profit. Tokyo: Kodansha Interna-

tional.


logical corollary to this sequence is that man-agers who demand excellence typically do notfare well in such an environment.

Organizational Size

As a general rule, organizational design in-creases in importance as organizations in-crease in size. An organization consisting offive family members has no need for formaldepartmental or divisional structures. Insuch situations a highly informal organiza-tional structure will probably suffice. Asorganizations grow in size, however, coordi-nating the efforts of individuals with a highlyinformal organizational structure becomesincreasingly difficult. Thus, as organizationsgrow in size, the level of formalization inorganizational structure needs to increase.The ability and/or willingness to adopt amore formal organizational structure oftendistinguish new businesses that succeedfrom those that fail (Since, Haverman, &Tolbert, 2005).

Empirical research has supported therelationship between organizational sizeand structure. Perhaps the best-known re-search was conducted by the Aston researchgroup in Great Britain (Hickson, Pugh, &Pheysey, 1969). These researchers foundthat organizational size, along with majororganizational technology (which will be de-scribed next), were both associated withorganizational structure. Organizationsthat were large (as measured by number ofemployees) tended to exhibit the character-istics of bureaucracy, or what has also beenlabeled mechanistic organizational structure(Burns & Stalker, 1961). In contrast, smallerorganizations tended to resemble character-istics of humanistic organizations, or whathas been labeled organic organizationalstructure.

Larger organizations gravitate towardmore bureaucratic designs because this helpsthem cope with the complexities that are in-evitable when large numbers of people areinvolved. For example, having free-flowingcommunication in a large organization couldpotentially lead to information overload and,ultimately, chaos. Unfortunately, many orga-nizations face a competitive environmentthat is not well served by having a highlybureaucratic organizational structure. Thus,many organizations deal with this issue bycreating what may be described as a hybridtype of organizational structure, in which theorganization as a whole might be describedas a bureaucracy. However, within smallerorganizational units, the structure and cul-ture are more like those of a humanisticorganization.

Small organizations obviously have lessneed for coordination and thus are able toadopt a purer version of humanistic orga-nizational structure. For example, in theseorganizations, communication is more free-flowing, roles are less well defined, andemployees are consulted on a regular basis.Highly organic structures often work verywell, but there may be a point at whichmore formalization is necessary. As an ex-ample, many successful high-technologycompanies are started by technical expertswho have little, if any, formal managerialexperience. As a result, most of these com-panies tend to adopt highly informal orga-nizational structures and cultures. As theirsales revenues increase, these types of orga-nizations are often acquired by larger orga-nizations wishing to capitalize on theirtechnical innovations. When these acquisitionsoccur, a major dilemma for the parent com-pany is how much structure to impose onthese small innovative organizations (seeComment 13.2).


Major Technologies

The word technology typically conjures upimages of complex machinery and manufac-turing processes. Although this applies tosome, technology need not be highly com-plex. Technology is defined simply as theprimary means by which inputs from theenvironment are converted into somethingtangible that can be returned to the envi-ronment (Katz & Kahn, 1978; Scott, 1990).For an organization that makes paper prod-ucts, for example, technology representsthe processes used to convert wood to

products such as paper napkins and tissuepaper. The ‘‘technology’’ of a drug treatmentcenter represents the therapeutic methodsand interventions that are used to ‘‘convert’’people from being addicted to alcohol anddrugs to being free from these addictions.

An organization’s primary technology isrelated to organizational design in severalways. At the most general level, the two mustbe compatible. Since an organization’s tech-nology is typically established prior to itsstructure, the structure is typically createdto support technology rather than the re-verse. One of the most widely cited studies

COMMENT 13.2

THE COORDINATION-AUTONOMY DILEMMA

FOR MOST ORGANIZATIONS one of the biggest

challenges is to come up with highly innova-

tive products or services for their customers.

Many organizations respond to this challengeby spending large amounts of money on re-

search and development. Other organizations,

however, take a different approach—they

acquire highly innovative small organizations

and use this innovation to their advantage.

According to Puranam, Singh, and Zollo

(2006) the major dilemma large organizations

face when they acquire a small, innovativecompany is striking the optimal balance be-

tween coordination and autonomy. In other

words, do they completely absorb the new

company into their structure, or do they allow

their new acquisition to operate autono-

mously? Too much coordination and control

could potentially decrease the very quality that

they were seeking in the acquired company—namely, innovation and creativity. On the

other hand, allowing the new company to

operate with complete independence may pre-

vent it from using the resources the larger

company has to offer. So what is the optimal

balance?

Based on a study of large manufactur-

ing organizations that had acquired small,technology-based firms, Puranam et al. (2006)

concluded that a high level of coordination and

control is probably not advisable if the acquired

organization does not have a great deal of expe-

rience in developing new products; in these

cases a high level of autonomy is better. On

theotherhand,acquiredorganizationsthathave

alongertrackrecordofproductdevelopmentarenot adversely affected by coordination.

The overall conclusion one might draw

from this research is that organizations acquir-

ing very young companies with creative

‘‘potential’’ must be prepared to give these

organizations the autonomy and freedom to

reach that potential.

Source. Puranam, P., Singh, H., & Zollo, M (2006).

Organizing for innovation: Managing the coordination-

autonomy dilemma in technology acquisitions. Academy

of Management Journal, 49, 268–280.


that relates structure and technology wasconducted by Woodward (1965). In thisstudy, a number of organizations were classi-fied as having one of three primary technol-ogies. Large-batch technology organizationswere those that used traditional assembly-line production processes; that is, productswere mass produced, and this was done in asequential fashion. The second type of orga-nization was described as having small-batchtechnology. This type of organization pro-duced products that were essentially custommade, and thus it was not possible to obtainthe economies of scale that are possiblewith mass production. The third type oftechnology in this study was described ascontinuous-process technology. Organiza-tions using this type of technology are notmass-producing a product; they are convert-ing material from one state to another.

Based on this classification of technologytypes, Woodward (1965) found a distinctrelationship between technology and organi-zational design. The largest difference inorganizational design was found betweenorganizations that used large-batch andthose that used small-batch technology.Organizations that used large-batch technol-ogy tended to have bureaucratic or mecha-nistic organizational structures. Presumably,the need to maintain control and certaintyduring the production process would beresponsible for this finding. Organizationsusing small-batch technology tended to havemore humanistic or organic structures. Onewould assume that this would be due to thefact that custom-made products require ahigh level of adaptability and flexibility inan organization—both of which are hard toachieve in a typical bureaucratic organiza-tion. The third type of organization, theone using continuous-process technology,was found to be a hybrid between bureau-cratic and humanistic structures.

Although Woodward’s (1965) findingshave been widely cited over the years, theyhave also been criticized. The major criticismof her study was that organizational technol-ogy and organizational size were confounded.Because technology tends to covary quitestrongly with organizational size, it has beenasserted that her findings are due primarilyto organizational size. Researchers in thepreviously mentioned Aston project empiri-cally tested this idea and found that, althoughsize was a predictor of organizational design,it was not completely confounded with tech-nology (Hickson et al., 1969)—that is,technology and organizational design hadindependent effects on the design of organi-zations.

Another potential limitation of Wood-ward’s (1965) research is that many organi-zations do not rely on only one technology;rather, they utilize multiple technologies.Universities, for example, often use quitedifferent instructional methodologies forundergraduate education than they do forgraduate programs. This leads to a question:What type of organizational structure wouldbe present in an organization that utilizesvery different technologies at the same time?At present, there are no definite answers tothis question. However, it is probably logicalto assume that organizations often deal withthis issue by having several structures.

A final limitation is that the classificationdeveloped by Woodward over 40 years agomay not be as applicable today. For example,because of changes in manufacturing tech-nology, organizations are now capable ofproducing customized versions of productsthat were mass-produced at the time ofWoodward’s study (e.g., Zammuto & O’Con-nor, 1992). In addition, many tasks that wereonce performed by assembly-line workershave now been automated. Thus, employeesinvolved in mass production today are much


more likely to be involved in monitoring tasksas opposed to repetitive tasks. As a result, theorganizational structure needed for mass pro-duction may be quite different today than itwas 40 years ago; for example, it may be moresimilar to structures that support continuous-process technology.

RECENT INNOVATIONSIN ORGANIZATIONAL DESIGN

While the distinction between classical andhumanistic organizations is useful, manyorganizations today have adopted organiza-tional designs that resemble neither of these.These different organizational designs haveevolved for a number of reasons. One is thatmany organizations have found that purelybureaucratic or purely humanistic organiza-tional designs have not allowed them to meetthe challenges of their competitive environ-ment. Also, different organizational designshave developed as a way to better leverageboth employee skills and organizational re-sources. Finally, evolution of these differentorganizational forms has corresponded torecent changes in managerial philosophy.In this section, we examine three of the mostcommon recent innovations in organiza-tional design: (1) the team-based organiza-tion, (2) the matrix organization, and (3) thevirtual organization.

The Team-Based Organization

In the past 20 years, research on group orteam effectiveness within organizational psy-chology has grown rapidly (Guzzo & Shea,1992). Along with this increasing interest,organizations have dramatically increasedtheir use of teams. Most organizations utilizeteams under the assumption that teams rep-resent a much more effective way to utilize

and combine employees’ skills rather thansimply having employees work as indi-viduals. Teams can be very effective tools inorganizations, yet they are certainly not theanswer to all problems. In fact, teams maycreate problems of their own.

A team-based organization is one thatgoes beyond the occasional ad hoc use ofteams and uses them as the basis for theirorganizational structure. Figure 13.5 illus-trates an organizational design that is basedentirely on teams. As can be seen, this smallorganization consists of four cross-functionalteams that correspond to the organization’sfour primary product lines. According toMohrman and Quam (2000), in a team-based organization, teams are used to ‘‘carryout their core work, to develop and deliverthe products and services that provide valueto customers’’ (p. 20). In this type of organi-zation, teams have responsibility for activitiessuch as planning work, staffing, and com-pensating team members.

According to Galbraith (1995), a numberof important issues must be addressed foreven a simple team-based organizationalstructure to work effectively. One of the most

FIGURE 13.5Example of a Team-Based Organization

ProductTeams

SalesTeams

ProductDevelopment

Teams

Vice-PresidentResearch andDevelopment


Vice-PresidentProduction

CEO


important of these is coordination of theactivities of the different teams comprisingthe structure. This is important because theactivities of one team often impact the activ-ities of others. As an example, let’s say themarketing efforts of each of the teams inFigure 13.5 are not coordinated well. Thiscould, for example, result in sales represen-tatives from each team calling on the samecustomers at the same time; this decreasesthe chances of any sale being made and isembarrassing to the organization. One of themost common ways to achieve coordinationis to create what is essentially an executivecommittee consisting of the leaders of each ofthe teams. That is, the leader of each teambelongs to a higher-level team that meetsperiodically to consider issues of coordina-tion between teams. Given the size and com-plexity of an organization, there may beseveral such executive or coordinating com-mittees at various levels of the organization.A variant of this method would be to createvarious coordinating committees from differ-ent organizational levels.

Many academic departments in univer-sities utilize this mechanism by having anexecutive committee comprised of the chair-person, the director of the undergraduateprogram, and the directors of the graduateprograms. These individuals are oftenresponsible for setting overall departmentalpolicy in a way that balances the needs of thevarious programs offered by the department.Such a committee might also be charged withcreating long-term strategic plans. A second-ary benefit of this type of committee is that itcan be an excellent mechanism for trainingmembers of an academic department toassume leadership positions in the future.

Another crucial issue in implementing ateam-based organizational structure is creat-ing a compatible reward system. As wasstated in Chapter 12, a common problem

with the use of teams in general is that orga-nizational reward systems are often aimedprimarily at individual-level performance.As a result, individuals often have little in-centive to put the interests of the team aheadof their own. This issue becomes even moreimportant with a team-based structure be-cause overall organizational performance isintimately tied to the performance of indi-vidual teams.

One way to align reward systems with ateam-based structure is to base rewards pri-marily on the team performance or even theorganizational performance. Compensationmethods such as profit sharing and gainsharing are common ways of accomplishingthis objective. This is not to say that organi-zations should completely ignore individualperformance. However, if organizational per-formance depends primarily on team perfor-mance, this should be the focal point of thereward system.

Organizations implementing team-basedorganizational structures must also considerthe nature of the work being performed bythe teams. One obvious question must beasked before implementing a team-basedorganizational structure: Does the work lenditself to a team-based structure? This mayseem like a rather obvious point, yet manyorganizations overlook it. In their zeal toreap the benefits of teamwork, many organi-zations overlook the fact that the work beingperformed does not lend itself to a team-based structure. If the work does not lenditself to a team-based structure, an organiza-tion can either abandon this structure orredesign the work.

Finally,organizationscommitted to imple-mentinga team-basedorganizational structuremust take a hard look at their selection proce-dures. As with any organization, employeesmust be selected based on whether they areable to perform important job-related tasks.


However, as discussed in Chapter 12, team-basedorganizationsmustalsoconfront the factthat not all individuals want to work within ateam-based structure. Some individuals aremore comfortable than others working in ateam-based environment (Campion et al.,1993). There is also evidence that some indi-viduals possess better team-based skills thanothers do (e.g., Stevens & Campion, 1999).

The Matrix Organization

In a matrix organization there are essentiallytwo separate organizational structures at thesame time (Davis & Lawrence, 1977). One ofthese structures is often represented by tradi-tional functional departments such as mar-keting, engineering, accounting, and so on.At the same time, a second structure is super-imposed on this traditional departmentalstructure. Most typically, this second struc-ture is based on organizational projects, al-though there are many other bases forcreating this second structure. Consumerproduct companies often have secondarystructures based on brands or, in some cases,different markets that they serve.

In a typical matrix structure managers incharge of different projects draw employeesfrom each of the functional departmentsuntil a project is completed. Figure 13.6provides a simple illustration of how thismight work in practice. Notice that projectA draws employees from all three functionaldepartments, project B draws employeesfrom both marketing and engineering, andproject C draws employees from both mar-keting and accounting. Projects in matrixorganizations may be of limited duration,but in some cases they are relatively perma-nent. For example, in consumer productcompanies that have matrix structuresbased on brands, these structural arrange-

ments are more or less permanent. It ishighly likely, however, that the specificneeds of each of the brands may changeover time, and thus the specific functionalresources devoted to each brand may alsofluctuate over time.

The primary advantage of a matrix struc-ture is that it allows an organization toquickly shift the focus of its design to themost important parts of its business. Forexample, the lifeblood of an organizationthat is a major defense contractor is obvi-ously defense contracts. Therefore, if such anorganization receives a large contract todevelop an advanced weapons system, a ma-trix structure allows it to quickly shift a greatdeal of its internal resources to that project.When the project is completed, resourcescan then be quickly reallocated to other pro-jects as needed.

Matrix structures are used frequently byconsumer product companies because theyallow an organization to use a unique ap-proach in the way it handles different prod-ucts or product lines. For example, a well-established product line may require a vastlydifferent advertising approach than a productthat has just emerged on the scene. In theconsumer product environment, matrix

FIGURE 13.6Example of a Matrix Organizational Structure

Project A

ProjectsEngineering

Functioning Departments

Accounting

Project B

Project C

Marketing

X X X

X X

X X


structures also help to guard against compla-cency. Having people assigned to products(e.g., brand managers) makes it more likelythat the organization will continue trying toimprove them through research and develop-ment efforts and through the use of consumerdata.

Beyond these more obvious benefits,there are also intangible benefits. For exam-ple, being a project or brand manager can bea highly developmental experience for man-agers. In fact, Procter and Gamble—one ofthe largest consumer product companies inthe world—is known to use brand manage-ment positions explicitly for developmentalpurposes. Brand managers not only learn agreat deal about a particular product orproduct line, they also learn a great dealabout the organization as a whole as theydraw resources for their brand. Project orbrand managers in any organization alsolearn a great deal about negotiation and com-promise because, in many cases, they are notsimply handed resources; instead, they mustnegotiate with functional managers in orderto get them.

Matrix structures also pose a number ofchallenges. One of the biggest challenges isthat matrix structures often end up pittingproject or brand managers against functionalmanagers. This can result in a great deal ofdysfunctional conflict and political games-manship (e.g., de Laat, 1994). Some organi-zations address this issue by establishingclear priorities; for example, the needs ofprojects take priority over the needs of func-tional departments. While this might behelpful, it introduces some dysfunctionaleffects of its own. Functional managers maycome to see themselves as ‘‘second-class citi-zens’’ in comparison to project or brandmanagers. As a result, functional manage-ment is often seen as an undesirable position,

particularly if one wants to be promoted toupper management.

Matrix structures can also have a nega-tive impact on employees working in func-tional departments. In a large organizationwith a matrix structure, employees workingin a functional department often have tojuggle the demands of several projects atonce. This may result in work overload andcan often impose conflicting demands onemployees (Joyce, 1986). Beyond theseworkload issues, a by-product of matrixstructure is that employees may have to workfor several ‘‘bosses’’ at the same time. Fur-thermore, these different bosses may havevery different performance standards andinterpersonal styles. Thus, from an employ-ee’s point of view, working under a matrixstructure could be highly stressful. Func-tional managers must communicate fre-quently with project managers to ensurethat they are not overworking employees.

The Virtual Organization

According to Galbraith (1995), an exampleof a virtual organization is one that decides toproduce a product or service but contractswith other firms to provide a key part ofthat product or service. The major automo-tive manufacturers have, to a large extent,become virtual organizations. Althoughthey continue to assemble automobiles,most of the parts that were once producedinternally are now produced by externalsuppliers.

The primary motivation for forming vir-tual organizations is cost reduction. In thecase of the auto companies, the capital andother internal resources needed to produceautomotive parts have simply increased to alevel that is prohibitive. It is therefore muchcheaper to simply buy the parts from external


suppliers. A second motivation is that con-tracting out more peripheral functionsallows an organization to concentrate all ofits energy on its core business. In collegeathletics, for example, many universities hireoutside companies to make travel arrange-ments when they play road games. Since thisused to be a task performed by coaches,contracting this out allows coaches to focuson other more important aspects of their jobs(e.g., recruiting, developing game plans).

Despite the potential advantages of virtualorganizations, this type of organizational de-sign also has its drawbacks. When an organi-zation enters into a partnership with asupplier, there is some degree of risk. Forexample, when auto companies do not pro-duce their own parts, ensuring that those partsmeet quality standards is obviously more dif-ficult. Organizations have some leverage inthis regard because suppliers obviously wantto maintain the relationship. According toGalbraith (1995), organizations typically dealwith this uncertainty by thoroughly investi-gating potential business partners beforeentering into relationships. Another way isto build ‘‘escape clauses’’ into business part-nerships, as consultants often do.

Employing a virtual organizational struc-ture makes it difficult for an organization tomaintain a coherent culture. In the extreme,an organization that contracts out nearlyeverything ceases to become a distinct or-ganizational entity; instead, it becomes arather large conglomeration of differentorganizational cultures. This status maymake it very difficult to instill a sense ofcommitment and loyalty in employees. Thismay also explain why some organizationshave stayed away from the virtual organiza-tion concept. As the virtual organizationbecomes more prevalent, though, organiza-tions will probably find effective ways to dealwith this issue.

RESEARCH ONORGANIZATIONAL DESIGN

Perhaps one of the clearest themes in theorganizational design literature over the yearsis that organizational designs can be viewedas adaptive responses to the environment. Forexample, the designs of organizations differ asa function of environmental uncertainty(Burns & Stalker, 1961), technology (Scott,1990; Woodward, 1965), strategy (Galbraith,1995), and sheer organizational size (Hicksonet al., 1969). Thus, one thing that empiricalresearch has shown is that organizationaldesigns do not just appear; rather, they typ-ically come about as a response to an organi-zation’s environment. Interestingly, though, ithas also been proposed that organizationaldesign may at times reflect the personalityquirks of individual managers (Kets de Vries& Miller, 1986). Therefore, organizationaldesign decisions may not always have arational or functional basis.

Another clear theme in organizationaldesign research is that there is no one bestway to design an organization. Research hasshown that organizational designs comeabout as a result of environmental factors.However, there is not as much evidence thatorganizations with designs matching theirenvironments are always successful. If any-thing, organizational design research hasshown, with respect to design, that thereare many paths to organizational success. Itappears, though, that a key factor in whethera given organizational design leads to successis whether internal policies and proceduresare congruent with the design. Perhaps thebest example of this is the literature on team-based organizations, which has shown thatorganizations using this type of design aremuch more successful if their compensationpractices are congruent with it (Mohrman &Quam, 2000).


A third major theme is clearly evident inorganizational design research: The design oforganizations has tangible effects on employ-ee behavior. It has been shown, for example,that the design of an organization mayimpact employees’ willingness to contributeideas (Boode, 2005), managers’ attention toenvironmental issues (Russo & Harrison,2006), and employees’ use of safe work prac-tices (Bourrier, 2005). It has also been foundthat organizational structure has an impacton individual employees’ perceptions of fair-ness (Schminke, Ambrose, & Cropanzano,2000), which in turn may have an impact onmany subsequent behaviors.

Findings of this sort are important becausethey suggest that the design of an organizationmakes a difference. They also suggest possiblemediating mechanisms (e.g., fairness percep-tions) by which organizational designs—orany other organizational intervention, for thatmatter—impact organizational effectiveness.Perhaps most importantly, they force us toshift our thinking from viewing organiza-tional designs as ‘‘boxes and arrows’’ to seeingthem for what they really are—purposefulinterventions designed to shape and influencehuman behavior in a particular direction.For those in the field of psychology, this isimportant because it drives home the pointthat macro-level factors such as organiza-tional design are important. Conversely, forthose trained in organizational theory or strat-egy, it suggests that something very importantis in the ‘‘black box’’ between macro-levelvariables such as design and organizationaleffectiveness.

THE FUTURE OFORGANIZATIONAL DESIGN

How will organizations be designed in thefuture? This question obviously cannot beanswered with complete certainty. We do

know, however, that a number of trends willclearly shape organizations in the future. Onetrend that has impacted many aspects oforganizational functioning is the ever-increasing sophistication of information tech-nology. Business transactions that used totake weeks to accomplish can now be com-pleted in a matter of seconds. This increasedsophistication of information technology hasbeen a double-edged sword for organizations.On the positive side, it has resulted in a levelof efficiency and speed that is unprecedented.At the same time, however, it has contributedto a highly volatile environment in manybusiness sectors that were previously muchmore stable.

With respect to organizational design,information technology has led organiza-tions to place a premium on speed and flex-ibility. Thus, based on the organizationaldesigns that were discussed in this chapter,there will likely continue to be an increasingtrend toward designs that resemble Likert’sSystem 4 and System 5, as well as the team-based structures. The reason for this is thatspeed often demands that lower-level em-ployees must be empowered to make manydecisions that were once reserved for man-agement. Organizations simply will not havetime to go up through a large chain of com-mand to make every decision.

The other implication of informationtechnology is that it will likely increase theuse of more flexible organizational designs. Itwill become increasingly more common fororganizations to consist of individuals whospend some or all of their time ‘‘off site’’through telecommuting. This will also makevirtual organizations much more appealing,and therefore more likely to be used. Forexample, through videoconferencing, regu-lar meetings can take place between those atorganizational headquarters and suppliers indifferent parts of the country, or even the

The Future of Organizational Design 437

world. Thus, in the future, we may need toseriously rethink what an organizationactually is.

A second trend, which is actually some-what related to the first, is the globalization ofthe economy. With the fall of communism inthe early 1990s, coupled with free trade agree-ments and a trend toward standardization ofcurrency, the world has increasingly em-braced free-market capitalism. Furthermore,countries that were at one time essentiallyclosed to Western society (e.g., China) arenow becoming active trading partners withthe West. Obviously, the extent to which thisglobalization of the economy will lead toglobal prosperity depends on a number offactors (e.g., maintaining world peace, stabi-lizing new free-market economies). However,it will have a major impact on organizations.

With respect to organizational design,globalization will clearly increase the incen-tive of organizations to expand their marketsbeyond national borders. Thus, an increas-ing number of organizations will add foreignsubsidiaries to their existing structures. Howthese foreign subsidiaries are managed andintegrated with the existing structure will bethe major issue these organizations will face.Organizational design issues will focus pri-marily on determining the appropriate com-bination of direction and autonomy to givethese foreign subsidiaries (Boode, 2005).

A third trend that will likely impact orga-nizational design is the growing contingentworkforce (Beard & Edwards, 1995). Recallfrom Chapter 1 that this is an importanttrend that was predicted to have an impacton many areas of organizational psychology.In the realm of organizational design, theincreasing availability of a contingent work-force, combined with the previously dis-cussed trends, will make the virtualorganization much more prevalent in thefuture. Having contingent employees readily

available makes it much easier for organiza-tions to quickly reconfigure themselves inorder to take advantage of market opportu-nities. In the past, more permanent bureau-cratic structures made this type of flexibilitynearly impossible. Of course, along withthese new opportunities come a number ofchallenges, such as maintaining consistentperformance standards and instilling acoherent sense of organizational culture inthese transient employees.

CHAPTER SUMMARY

In this chapter, we shifted the focus fromindividual- and group-level phenomena to abroader level. This shift is important becausea complete understanding of behavior inorganizations requires that we look at it fromall three vantage points. The chapter beganwith a discussion of the field of organiza-tional theory, primarily focusing on its intel-lectual roots and its linkage to the closelyrelated field of organizational design. Thefocus then shifted to a description of thethree most general types of organizationaldesigns. Classical organizational designs arerepresented by Scientific Management, IdealBureaucracy, and Administrative Manage-ment. Humanistic organizational designsare best represented by McGregor’s TheoryX/Y and Likert’s System 4 organization. Clas-sical and humanistic organizational designshave both advantages and disadvantages.Furthermore, the appropriateness of eachof these organizational types depends largelyon situational factors, and this leads to thethird type of organizational design: contin-gency organizational design.

Based on the premise that contingencytheory is the dominant paradigm in organi-zational design today, the chapter thenshifted to the major factors that are taken


into consideration when organizations makedesign decisions. These included strategy,level of environmental uncertainty, beliefsand assumptions of those in power, organi-zational size, and the dominant technology.

Given this number of factors, there will obvi-ously be a variety of organizational designs.Furthermore, it is very likely that differentorganizational designs may coexist withinthe same organization.


DAVID NADLER AND MULTIPLE APPROACHES TO UNDERSTANDING HUMAN

ORGANIZATIONS

When I was in college, I had summer jobs, and

in two successive summers I worked in one of

the best jobs I ever had, and subsequently oneof the worst. I was puzzled by how two orga-

nizations could be so radically different—one

extremely effective with motivated staff, and

one very bureaucratic with disaffected staff.

That began my interest in organizations,

and as I read, I got very interested in the whole

area of change, leadership, and interpersonal

relations.When I went to business school at Har-

vard, I was exposed, however, to a whole new

set of ideas. I was lucky enough to have classes

with Jay Lorsch and Paul Lawrence, who were

working on ideas of systems, environment, and

structure. At the same time, I was taking a

seminar in psychoanalytic theory and organi-

zations, which stretched me in other direc-tions. Much of this came together in my

doctoral work at Michigan. People there were

thinking about systems theory and organiza-

tions. In a course with Ed Lawler, he asked us,as a mid-term assignment, to ‘‘develop a frame-

work to explain the behavior of organizations.’’

My mid-term paper was the beginning of work

on developing an integrated theory of organi-

zations, which ultimately became the Congru-

ence Model that Michael Tushman and I

developed and have used for several decades.

I ended up with a strong view of organi-zations as both social and technical systems.

You need to think about both the ‘‘software’’

(people, behavior, culture, etc.) and the ‘‘hard-

ware’’ (structure, systems, processes) in the

context of strategy, to understand what’s going

on.

After seven years in academia, I left to create

a consulting firm. I used the same approach inworking with clients, but have tried to always be

open to new ideas, new models, and the value of

scientific inquiry in solving real problems. On

reflection, no one theory, approach, or frame-

work explains everything. Each concept or

model gives us a different set of lenses to under-

stand this very complex creation we call the

human organization. Academics and leadersalike need to be ready to use different tools

and approaches.

David A. Nadler

Senior Partner, Delta

Organization & Leadership

Vice Chairman,

Marsh & McLennan Companies, Inc.

Chapter Summary 439

The chapter then shifted to focus on threerelatively recent trends in the design of orga-nizations: (1) team-based organizationalstructure, (2) matrix organizational structure,and (3) the virtual organization. Althougheach of these organizational designs is differ-ent, they all allow organizations to respondmore quickly to market opportunities and tomake better uses of their internal resources.Furthermore, all three of these organizationaldesigns require that other organizational sub-systems must be properly aligned in order forthem to work well.

There has been considerable research onthe impact of organizational design, and thiswas summarized according to the dominantthemes. One of these themes is that designsappear to be at least partially traceable toadaptive responses to the environments inwhich organizations operate. Research hasalso shown that there is no ‘‘right way’’ todesign an organization; however, organiza-tions that tend to align their various subsys-tems with their structure tend to be the mosteffective. A third theme is that organizationaldesigns do influence the behavior of employ-ees. This serves as an important linkage be-tween micro- and macro-level organizationalbehavior.

The chapter concluded with a brief dis-cussion of factors that are likely to influenceorganizational designs of the future. Theseinclude information technology, globaliza-

tion of the economy, and the increasingnumber of contingent employees. Thesetrends may have many influences on organi-zations, but their most likely impact on de-sign will be to increase the use of virtualorganizational designs. Organizations willthen be better able to expand and contractquickly, and to move much more quicklyinto previously untapped global markets.


Boode, G. (2005). Boundaries on themove: The impact of cultural values andlanguage on organizational design andcommunication within an organization inThailand. Asian Pacific Business Review, 11,519--533.

Bourrier, M. (2005). The contribution oforganizational design to safety. EuropeanManagement Journal, 23, 98--104.

Russo, M. V., & Harrison, N. S. (2005).Organizational design and environmentalperformance: Clues from the electronicsindustry. Academy of Management Journal,48, 582--583.

Since, W. D., Mitsuhashi, H., & Kirsch, D.(2006). Revisiting Burns and Salker: Formalstructure and new venture performance inemerging economic sectors. Academy ofManagement Journal, 49, 121--132.


Chapter Fourteen

OrganizationalCultureP

erhaps the best way to appreciateorganizational culture is to imag-ine entering an unfamiliar organi-zation for the very first time—either as a new employee or a

customer. In some ways, this experience issimilar to entering a foreign country. Forexample, members of the organization mayuse words and phrases that are not fullyunderstood; they may engage in behaviorsthat they take quite seriously but have littlemeaning to outsiders; and they may tell jokesand stories that only they can fully under-stand. If we were to stay in the organizationlong enough to make the transition to full-fledged organizational members, or interactwith members of the organization frequentlyenough, many of the unfamiliar things thatwe initially observed would become muchmore meaningful.

While the study of culture has a longtradition in anthropology and sociology,the study of organizational culture is actuallyvery new. In fact, most researchers have tracedits beginning to the late 1970s (Pettigrew,1979). However, the fact that organizationalpsychologists have studied organizationalculture for only a short period of time doesnot decrease its importance. To the contrary,culture is an extremely important key tounderstanding many behavior patterns inorganizations. In fact, all behavior in organi-zations occurs in a cultural context. This mayexplain why some things (e.g., incentive pay)work well in some organizations yet fail mis-erably in others. Culture may also help us tounderstand why some organizations are suc-cessful and why others are not (e.g., Mason,2004).

This chapter provides an overview oforganizational culture and many of its impli-cations. We begin by defining what is meantby organizational culture—no small feat,considering that this concept comes not onlyfrom psychology, but also from culturalanthropology and sociology. We then de-scribe two recent attempts to describe thedimensions underlying organizational cul-ture. The chapter then shifts to an explana-tion of the various ways in which culture isreflected in organizations. As we’ll see, someof these are rather obvious, but culture isoften reflected in very subtle ways. We thenexplore the factors that shape the culture ofan organization. The chapter will then focuson the various methods that can be used to

441

study organizational culture. The chapterthen shifts to a discussion of organizationalculture change, and concludes with anexamination of the impact of organizationalculture, both on the success of the organiza-tion as a whole and on individual organiza-tional members.

DEFINING ORGANIZATIONALCULTURE

There are many definitions of organizationalculture in the organizational psychology lit-erature (e.g., Louis, 1983; Martin, 2002).Fortunately, however, most of these have agreat deal in common. In an effort to inte-grate these various definitions, Ravisi andSchultz (2006) state that organizational cul-ture is ‘‘a set of shared mental assumptionsthat guide interpretation and action in orga-nizations by defining appropriate behaviorfor various situations’’ (p. 437). There aretwo important aspects of this definition.First, this definition implies that culture isthe ‘‘view of the world’’ that the members ofan organization share. By ‘‘view of theworld,’’ we mean that culture essentially rep-resents the ‘‘lens’’ through which employeesof an organization learn to interpret the en-vironment. Secondly, and perhaps mostimportant, this definition also implies thatculture guides the behavior of employees inan organization. For example, cultureimpacts how employees treat each other,the quality of decisions that are made, andultimately whether or not an organization issuccessful.

According to Schein (1985, 1992), thereare three levels of an organization’s culture,and each succeeding level is more difficultfor outsiders to decipher. The most visiblelevel of organizational culture is reflected inartifacts, technology, and behavior patterns.Artifacts, which will be discussed in more

depth later, are aspects of the physical envi-ronment that communicate cultural mean-ing. Technology represents the means bywhich organizations transform input fromthe outside environment. Behavior patterns,of course, simply represent what employeesin the organization do.

The next level of culture, according toSchein (1992), is represented by the sharedvalues within the organization. Values sim-ply represent individuals’ broad tendenciesto prefer certain things, or states of affairs,over others (Hofstede, 1980). The valuesthat might be salient within an organi-zation could be a number of things: loyalty,customer service, collegiality, and self-preservation, to name a few. According toSchein, values are less accessible to an out-sider than are things such as behavior pat-terns, and, typically, they must be inferredby the outsider through symbolic means. Forexample, if an organization rewards and pro-motes employees largely on the basis of se-niority, one might infer that the organizationtends to place a high value on loyalty andretention.

In considering values in organizations,it is important to distinguish between thevalues that are espoused by the organization,and those that are actually in operation(e.g., the ‘‘true’’ values). In many organiza-tions, there is a strong relationship betweenthe espoused and the true values. For exam-ple, innovation has always been an espousedvalue at 3M, and studies of this organization(e.g., Gundling, 2000) have shown that gen-erally company practices are consistent withit. In some organizations, however, there is adisconnect between what an organizationclaims to value, and the values that appearto be guiding overt behavior. As an example,many organizations claim to place a highvalue on diversity, yet have few minorityemployees in management positions; many

442 Organizational Culture

organizations claim to place a high value onperformance, yet tolerate consistently poorperformance from employees; and manyorganizations claim that customer service isone of their core values, yet customers aretreated rudely. The important implication ofthis is that in many cases researchers must gobeyond espoused values in order to trulyunderstand organizational culture.

The third layer of culture, according toSchein (1992), is represented by the basicbeliefs and assumptions held by the mem-bers of an organization. These beliefs andassumptions are so deeply ingrained thatpeople take them for granted; they areimportant, however, because they impactthe more visible aspects of culture (Denison& Mishra, 1995). To better understand basicbeliefs and assumptions, let’s consider abasic assumption that we operate under inour daily lives, at least in Western society.For example, when people greet each other,it is common for one or both persons in-volved in such an encounter to ask the other,‘‘How are you doing?’’ or ‘‘How’s it going?’’Most people understand that these ques-tions, particularly when people do not knoweach other very well, are merely forms ofgreeting, and appropriate responses mightbe ‘‘Fine, thanks’’ or ‘‘Not bad; how areyou?’’ On the other hand, most people wouldbe uncomfortable if a person responded tothis question with a detailed 30-minuteexplanation of all of the challenges they havefaced in the past month.

What are the basic beliefs and assump-tions that people hold in organizationalsettings? This is a difficult question to answerbecause organizations, and the people inthem, differ so widely. However, if one thinksabout it, there are probably some beliefsand assumptions that may be salient, regard-less of the situation. For example, employeesin organizations have basic beliefs and

assumptions about things such as whetherthe organization can be trusted, whether theorganization supports them, whether thepsychological environment is threateningor supportive, or whether hard work anddedication pay off. There are obviously otherbasic assumptions that are quite specific to agiven organizational setting. For example,members of an accounting firm may havebasic assumptions about the ethics surround-ing the tax deductions they seek for theirclients, or the teachers in an elementaryschool may hold common basic assumptionsregarding the benefits of parental involve-ment in children’s education.

Compared to the other two levels of cul-ture discussed, basic beliefs and assumptionsare difficult to study because they are soingrained; in fact, Schein (1992) argued thatthey are not at a conscious level. Because ofthis, it is extremely difficult for a naıve orga-nizational outsider to determine what thesebeliefs and assumptions are. It is also diffi-cult for employees, particularly those whohave been around for a long time, to articu-late the basic beliefs and assumptions of theirorganization because they are so ingrained.Most typically, basic beliefs and assumptionsare determined only through painstaking re-search processes such as field observation,use of informants, and careful study of orga-nizational archives. More will be said aboutstudying organizational culture later in thechapter.

Having defined organizational cultureand described its various levels, an importantissue to consider is that, even though mostorganizations have what could be describedas an ‘‘organization-wide’’ culture, they alsocontain a number of identifiable subcultures.Janson (1994) proposed that, in most orga-nizations, there are six subcultures; these arepresented in Table 14.1. As can be seen, thefirst subculture is labeled ‘‘Elite culture/


corporate culture’’ and is essentially repre-sented by those at the highest levels of theorganization. The subculture in which thechief executive and the top executive groupof an organization live is much different thanmost other employees. These individualstypically have more pleasant surroundingsthan other employees, and they have a greatdeal of control over information dissemina-tion in the wider organization.

The next form of subculture described byJanson is labeled ‘‘Departmental.’’ Indi-viduals within the same department workvery closely together, face many of the samechallenges, and collectively experience suc-cess and failure. Because of this, individualswithin departments may develop many ofthe same views and thus have many of thesame basic assumptions about the organiza-tion. In universities, this is very evidentwhen one looks at the different cultures thatdevelop in academic departments (see Com-ment 14.1).

The next level of subculture developmentis at the ‘‘division’’ level. In a business orga-nization, for example, the marketing divisionmay consist of the sales, market research,and advertising departments. In a university,the equivalent of a division is a college that iscomposed of several academic departments.

Divisional subcultures develop for essen-tially the same reasons as departmentalsubcultures. Employees in the same divisionmay work under many of the same policiesand may experience many of the same chal-lenges. As a result, individuals within thesame division may begin to develop manyof the same basic assumptions, and hence asubculture develops.

The next level of subculture that maydevelop is labeled ‘‘Local culture’’ and isbased on geographic regions. Local subcul-tures may be identical to divisional subcul-tures when an organization’s structure isbased on geographic region. However, thisis not always the case. Local subculturesdevelop largely based on local customs andnorms of the region in which a unit works.For example, as a graduate student, the firstauthor worked as a contractor in the Floridadivision of a large telecommunications com-pany. Based on conversations with others inthat organization, it became evident that therewere distinct regional differences betweenthis division and other regional divisions ofthe company. Due to the warm weather inFlorida, the dress code was a bit morerelaxed, and the manner in which peopledealt with each other was a bit more informalthan in other parts of the organization.

TABLE 14.1Possible Subcultures within an Organization Proposed by Janson (1994)

Elite culture/corporate culture—‘‘For your eyes only’’ or ‘‘For public consumption’’

Departmental culture—Horizontal slice; for example, sales department

Divisional culture—Vertical slice; for example, a division

Local culture—Within a geographical location/unit

Issue-related culture—Metaphorical, related to an important issue throughout the

organization; for example, safety culture or quality culture

Professional culture—On the basis of professional background and training

Source: J. V. Mbijen. (1998). Organizational culture. In P. J. Drenth and H. Thierry (Eds.),

Handbook of work and organizational psychology (2nd ed., Vol. 4, pp. 113–131). Hove, England:

Psychology Press. Reprinted with permission of publisher.


Subcultures may also develop due tosome important issues faced throughoutthe organization. Recently, much work hasbeen done on the construct of ‘‘safety cul-ture’’ in organizations (Hofmann & Stetzer,1998). In reality, in most organizations,there are probably many safety ‘‘subcul-tures’’; that is, safety is likely to be viewedand practiced quite differently in many dif-ferent parts of an organization. Other impor-tant issues that may be the basis forsubculture development may include affirm-ative action, whether pay should be based onperformance, and views of the trustworthi-ness of management, to name a few.

The final basis for subculture develop-ment proposed by Janson is the professionaltraining of employees. In some organiza-tions, this could be the basis for the wider

organizational culture (e.g., accountingfirms, law firms, consulting firms); but, inmany other cases, organizations employgroups of individuals who have obtainedvery different forms of professional train-ing. For example, an organization thathires groups of chemical engineers mayfind that these individuals constitute adistinct subculture within the organiza-tion. In fact, in some cases, employeesmay have a much greater identificationwith their professional subculture thanwith the organization or division in whichthey work. Physicians, for example, oftenidentify more strongly with the medicalprofession than they do with the hospitalsor clinics in which they are employed.

Given that subcultures coexist with theoverall organizational culture, a logical

COMMENT 14.1

ACADEMIC DEPARTMENT CULTURES AND SUBCULTURES

AS MOST READERS have probably noted in taking

courses in a variety of departments, there

tends to be a great deal of variation in the

cultures of various academic departments.For example, the cultures of academic depart-

ments in the arts and humanities tend to be

more ‘‘laid back’’ and relaxed in comparison to

academic departments in engineering and the

hard sciences.

Academic departments in business schools

also tend to be distinct in their own way. For

example, many business school faculty dressformally (many business schools require that

faculty wear business attire when teaching),

and many of their offices resembled those in

corporations. Considering that most faculty

members in business schools are trained in

business schools, and that their primary mis-

sion is to train students for business careers, it

is understandable that a businesslike culture

would develop.

Psychology departments tend to have very

interesting cultures because of the dominanceof subcultures. Particularly in large psychology

departments with several doctoral programs,

the subcultures that develop in each of the areas

may be quite different. For example, the culture

of a clinical psychology faculty might be very

different from the culture of an industrial/orga-

nizational psychology faculty. The cultures of

both groups, in turn, may be very different thanthe culture of a social psychology faculty.

The next time you’re in an academic

department, look around and see if you can

find any clues about the culture of that depart-

ment. Better yet, do this with two or more

departments and see what the differences

are. You might be surprised and fascinated

by what you find!


question is which of these has the strongestimpact on employees? Adkins and Caldwell(2004) examined this issue with employeesin a large consulting firm that was comprisedof four distinct subcultures based on typeof service provided (Strategic Consulting,Technology Consulting, Process Reengineer-ing, and Change Management Consulting).What these researchers found was that jobsatisfaction was positively associated with thedegree to which employees fit into both theoverall culture and the subculture in whichthey worked. This suggests that while sub-cultures are important to employees, theyalso do not completely negate the importanceof the overall culture of the organization.

MODELS OF ORGANIZATIONALCULTURE

Having defined what organizational cultureis, we will now discuss common dimensionsalong which organizational cultures can bedescribed. Anyone who has worked in sev-eral different organizations knows that, to alarge extent, no two organizational culturesare completely alike. Therefore, it is probablyfutile to develop a finite typology of all orga-nizational culture types or dimensions. Overthe years, however, researchers have discov-ered what they have considered to be clustersof cultural attributes that are common tomost organizations. In this section, we exam-ine two of these common models of organi-zational culture.

The O’Reilly, Chatman, and CaldwellModel

O’Reilly, Chatman, and Caldwell (1991)developed a self-report measure of organiza-tional culture, the Organizational CultureProfile (OCP). While the specifics of theOCP will be discussed later in the chapter,

the substantive model underlying the OCPwill be described briefly here. The substan-tive model behind the OCP is simply thatcultures can be distinguished based on thepredominant values that are reinforcedwithin a particular organization. Accordingto O’Reilly et al., most organizational cul-tures can be distinguished according to theseven values that are presented and definedin Table 14.2. According to this model, whatmakes each culture unique is its ‘‘profile’’ onthese seven cultural dimensions. For exam-ple, the culture of one organization mightplace a high value on innovation, respect forpeople, and it may have a strong team ori-entation. In contrast, the culture of anotherorganization might place a high value onstability, attention to detail, and it may placea low value on innovation.

Unlike other models of organizationalculture that have sought to link culturaldimensions to organizational performance,much of the research on the O’Reilly et al.(1991) model has been focused on the fitbetween organizational culture and person-alities of individual employees (e.g., Judge &Cable, 1997). One might imagine, though,that in certain circumstances some of thesecultural dimensions would likely be associ-ated with organizational performance. It ishard to imagine, for example, a successfulhigh-techology company with a culture thatplaces a low value on O’Reilly, Chatman, andCaldwell model – ‘‘ . . . innovation and avery high value on stability. It is also likelythat an organization providing a service toconsumers would need to place a high valueon respect for people and attention to detailin order to be successful.

The Denison Model

A second model of organizational culture,which also served as the basis for a culture


measure, was proposed by Daniel Denisonand colleagues (Denison, 1990; Denison &Mishra, 1995). The model, which is pre-sented in Figure 14.1, is much more com-plex than that proposed by O’Reilly et al.(1991). The basic idea is that organizationalculture can be described according to thefour general dimensions of Adaptability,Mission, Involvement, and Consistency.Each of these general dimensions, in turn,can be described in terms of three subdimen-sions. For example, the general dimensionof Mission is subdivided into Strategic Dir-ection and Intent, Goals and Objectives,and Vision. Adaptability is subdivided intoCreating Change, Customer Focus, andOrganizational Learning. Involvement issubdivided into Empowerment, TeamOrien-tation, and Capability Development.Finally, Consistency is subdivided into CoreValues, Agreement, and Coordination/Inte-gration. Readers will also note that thismodel allows the cultures of organizationsto be described along two broad dimensions(External versus Internal Focus; Flexible ver-

sus Stable) based on where they score on thevarious subdimensions.

While the Denison model is relativelynew, it has been subject to a great deal ofempirical scrutiny and been used in manyorganizations to facilitate diagnosis of cul-tural problems (Denison, Haaland, &Goelzer, 2004). In the future this model willlikely be the focus of continued research.

MANIFESTATIONS OFORGANIZATIONAL CULTURE

Given that organizational culture can be de-scribed along a common set of dimensions,what makes these cultural dimensions visi-ble to both employees and organizationaloutsiders? In other words, what exactlycontributes to differences in the ‘‘feel’’ ofdifferent organizations? This is an importantquestion because culture cannot be studied,diagnosed, or in some cases changed if we donot understand the various ways that cultureis communicated. Fortunately, organiza-tional culture researchers have come up with

TABLE 14.2O’Reilly et al. (1991) Seven Organizational Values

Value Definition

Innovation The extent to which an organization encourages employees to be innovative,

seek out new opportunities, and take risks

Stability The extent to which an organization emphasizes rules and values predictability

Respect for people The extent to which an organization emphasizes mutual respect, fairness, and

tolerance of differences among employees

Outcome orientation The extent to which an organization encourages employees to take action and to

strive for excellence in their work

Attention to detail The extent to which an organization encourages employees to be precise and

detail-oriented in doing their work

Team orientation The extent to which an organization emphasizes collaboration and teamwork

among employees

Aggressiveness The extent to which an organization encourages competition and aggressiveness

among employees

Source: Q’Reilly, C.A, Chatman, J., & Caldwell, D.F. (1991). People and organizational culture: A profile comparison approach to

assessing person-environment fit. Academy of Management Journal, 34, 487–516.


a number of ways, and they are described inthis section.

Symbols and Artifacts

According to Cohen (1974), symbols are‘‘objects, acts, relationships, or linguistic for-mations that stand ambiguously for a multi-plicity of meanings, evoke emotions, andimpel men to action’’ (p. 23). In most orga-nizations, symbols provide us with informa-tion on the nature of the culture. Perhapsone of the most revealing symbols in anorganization is the physical layout in whichemployees work. In some organizations, em-

ployees’ offices are located in large openareas; in others, however, employees are giv-en a great deal more privacy by having theiroffices placed in more remote locations. Inthe former setting, the office layout may besymbolic of a culture that places a high valueon sociability and openness of communica-tion and, in fact, employees may engage inthese types of behaviors. In the latter, thelayout may be symbolic of a culture charac-terized by a high degree of secrecy, or per-haps just a great deal of respect for privacy;employees in such a culture may tend toprovide information to each other only ona ‘‘need to know’’ basis.

FIGURE 14.1The denison organizational culture model


According to Schein (1983) an artifact isvery similar to a symbol; the only differenceis that artifacts represent a more directattempt to convey cultural meaning, where-as symbols are more indirect. As with sym-bols, artifacts are most easily found in thephysical environment of organizations. Oneof the most typical cultural artifacts in orga-nizations is the physical manifestation ofthe major technology that is used. In educa-tional settings, for example, classrooms areartifacts in that they convey the fact thatstudents are to be reasonably obedient recip-ients of the knowledge that is passed down tothem. In the Army, the uniform is a powerfulartifact to remind everyone that they are allsoldiers regardless of the setting in whichthey work and the job they perform.

Rites and Rituals

Rites represent ‘‘relatively elaborate, dramatic,planned sets of activities that consolidate var-ious forms of cultural expressions into oneevent, which is carried out through socialinteractions, usually for the benefit of an audi-ence’’ (Trice & Beyer, 1984, p. 655). The mostcommon rites carried out in organizations aresummarized in Table 14.3. As can be seen,rites of passage are often used to symbolize the

socialization from organizational outsiders tofull-fledged organizational members. Themilitary’s use of basic training is probablythe most dramatic organizational rite of pas-sage, but other organizations have these aswell. For example, a familiar rite of passagein academic settings is the oral defense ofone’s master’s thesis or doctoral dissertation.

In some cases, rites are designed to sanc-tion or, in a more general sense, to conveynegative information to employees. Rites ofdegradation often occur when there is aproblem in the organization or when theremust be a change in personnel. A recent andhighly publicized example of this was theexecution of former Iraqi dictator SaddamHussein. When someone is denied tenurein a university there is no public event, butthe year following the denial of tenure is atype of degradation ceremony. During thisyear, a faculty member must face his or herpeers each day, knowing that he or she hasfailed to meet tenure standards and thus willnot be employed there the following year.

In direct contrast, rites of enhancementare designed to convey positive information.This can be positive information about theorganization or public recognition of indivi-duals for exceptional levels of performance.To illustrate this type of rite, Trice and Beyer

TABLE 14.3A Summary of Organizational Rites

Type of Rites Example

Rites of passage Induction and basic training in the U.S. Army

Rites of degradation Firing and replacing top executives

Rites of enhancement Mary Kay seminars

Rites of renewal Organizational development activities

Rites of conflict reduction Collective bargaining

Rites of integration Office Christmas party

Source: H. M. Trice and J. M. Beyer. (1984). Studying organizational culture through rites and

ceremonials. Academy of Management Review, 9, 653–669. Reprinted with permission of the Copyright

Clearance Center.


(1984) provide the example of the employeeseminars conducted by the Mary Kay cosmet-ics company. During these seminars, thecompany legacy is celebrated, and individualemployees are recognized for outstandingsales performance—all of which is done witha great deal of fanfare and glamour. Many ofthe activities at the annual meetings of pro-fessional organizations often serve this pur-pose as well (see Comment 14.2).

In most organizations, there are timeswhen problems need to be addressed andemployees need to renew their sense of pur-pose within the organization. Rites of renewalserve this purpose. Trice and Beyer (1984)cite the use of organizational-developmentinterventions as a prime example of rites ofrenewal in organizations. For example, inter-ventions such as team building, survey

feedback, and Management by Objectives(MBO), which are often part of organiza-tional development programs, can be seenas ritualistic activities that ultimately serveto renew employees’ sense of purpose. Whilesuch activities provide employees with reas-surance that something is being done aboutthe problems in the organization, they mayalso mask the real causes of problems. Bydoing this, they may reinforce the existingpower structure and social arrangementswithin the organization.

This view of organizational develop-ment proposed by Trice and Beyer (1984)is certainly provocative, although manyorganizational-development professionalswould probably disagree with it. In fact,there is some empirical evidence that orga-nizational-development interventions can

COMMENT 14.2

PROFESSIONAL MEETINGS: GOING TO SIOP?

FOR NEARLY 20 years, a yearly spring ritual for

many industrial/organizational psychologists

has been attendance at the annual conference

of the Society for Industrial and Organiza-tional Psychology (SIOP). The most recent

(2007) SIOP conference was held in New York

City; this conference has also taken place in

Boston, Miami, St. Louis, Montreal, San Fran-

cisco, San Diego, Nashville, New Orleans,

Dallas, Atlanta, and Orlando.

Why do so many industrial/organizational

psychologists make the SIOP conference aregular event? One reason is that going to

the conference allows them to keep up on

the latest developments in both the science

and practice of industrial/organizational psy-

chology. Each year, the conference program

includes symposia and poster sessions that

allow researchers and practitioners to discuss

their findings and exchange ideas. This is

particularly important for researchers, because

much of what appears in academic journals is

often 1 or 2 years old!

Another important (and perhaps lessunderstood) function of the SIOP conference

is that it serves a socialization function. Each

year, many graduate students attend this con-

ference for the first time and receive their first

taste of what it is like to be in this profession.

They learn who the important people in the

profession are and how to conduct themselves

as professionals. They are also educated aboutthe major issues facing the profession. These

things are obviously important in transmitting

a professional culture that will live on far longer

than any individual. Furthermore, graduate

students attending this conference for the first

time leave feeling very enthused about the

profession they have chosen and eager to

attend the next year.


facilitate positive change in organizations(e.g., French & Bell, 1995), and thus aremore than expensive ‘‘feel good’’ rituals.On the other hand, it is possible for organi-zational development to be applied in asomewhat ritualistic fashion—an employeeopinion survey is conducted, a report iswritten and filed away. While everyonemay feel good about the process, the realityis that after this process is completed none ofthe substantive problems within the organi-zation are even addressed, much less solved.

Rites of conflict reduction are often con-ducted in organizations when potentiallydebilitating conflict needs to be addressed.Perhaps the best example of this type of ritein unionized organizations is the collectivebargaining process. According to Trice andBeyer (1984), this activity is a rite because, inmost cases, each side knows that an agree-ment is ultimately going to be reached. How-ever, on the way to getting there, each sidemust ‘‘play a game’’ that is consistent with itsrole. For example, representatives of thecompany must initially present an unaccept-able contractual offer in order to show thatthey are good stewards of organizational re-sources. The union representatives, in turn,must reject that offer and make contractualdemands that they know the organizationcannot agree to, just to show that they areprotecting the interests of the union mem-bership. Ultimately, this give-and-take proc-ess will produce a contract that is acceptable,if not ideal, to both sides.

The final type of rites described by Triceand Beyer (1984) are rites of integration.The major purpose behind rites of integrationis to encourage and revive common feelingsthat serve to bind members of the organizationtogether. In most organizations, the commonexample of this form of rite is the annualholiday office party. At most holiday gather-ings employees typically suspend normal rules

of protocol and simply have fun together.This experience of having fun together pre-sumably serves to make the social ties that bindthese people together that much stronger, evenif it is only for an afternoon or evening.

Rituals are closely related to rites becausethey are also enacted through behavior pat-terns. Trice and Beyer (1984) define a ritualas ‘‘a standard, detailed set of techniques andbehaviors that manage anxieties, but seldomproduce intended, technical consequencesof practical importance’’ (p. 655). Perhapsthe most visible examples of ritualistic be-havior come from the world of sports—inparticular, from baseball. Many baseballplayers, for example, believe that it is badluck to step on the chalk lines when runningonto the field, and often make a visible effortto avoid doing so (just watch closely some-time!). Perhaps the most elaborate ritualsever seen in baseball came from former majorleague baseball player and hall of famerWade Boggs. Boggs would eat only chickenon the day of a game, and he had to fieldthe same number of ground balls prior toeach game. In Boggs’s case, however, thesemight not be considered rituals because theyevidently did him some good—he won sev-eral American League batting titles andended his career with over 3,000 hits!

Employees in most organizations do notengage in ritualistic behaviors similar tothose of professional athletes. Organizationalrituals, however, do exist and they do conveyinformation about the organizational culture.For example, employees in many organiza-tions develop nearly ritualistic behavior thatcenters on daily breaks and lunch time. Eachday, employees may congregate in the samelocation or eat at the same restaurant atprecisely the same time. In contrast, in someorganizations, individuals may spend thesetimes eating at their desks or perhaps readinga book. In the former case, such rituals


convey strong social bonds within the orga-nization; in the latter case, they may suggest aculture that values individuals’ privacy andsolitude.

Other common ritualistic behaviors inorganizations can be seen at the beginningand end of the workday. Employees, forexample, may congregate around the coffeemachine and exchange pleasantries, or per-haps talk about sports or current events. Inother organizations, each employee may be-gin the day by quickly going to his or herdesk and immediately beginning to work. Inthe former case, one might again presumethat the social ties are a bit stronger; infact, they reach to the point where employeesmay feel that such daily activities are highlyvital to their work, even though the informa-tion exchanged may actually be quite trivial.In the latter case, this behavior, at least on thesurface, may convey a high level of diligenceand a desire to accomplish tasks. It may alsobe indicative of a high level of conflict andsuspicion among the employees of an orga-nization.

Another ritual that can be very revealingabout the culture of organizations—or, inmany cases, subcultures—is the type of social-izing after work hours. In some academicdepartments a common ritual on Fridayafternoons is socializing over drinks at a localbar. In contrast, in some academic depart-ments, faculty rarely, if ever, socialize outsideof work hours. In the former case, this weeklyritual conveys that members of a departmentsee themselves as more than just coworkers,and they wish to extend the social bondsbeyond the confines of the work environment.A lack of socializing outside of work couldmean that coworkers do not find each other’scompany appealing; it may also signify anorganizational culture in which employeesget along quite well, but place a very highvalue on spending time with their families.

Stories, Legends, and Dramas

It is certainly well documented, from fieldssuch as cultural anthropology (e.g., Geertz,1973) and communication theory (Pacanow-sky & O’Donnell-Trujillo, 1983), that storytelling and passing on legends are veryimportant ways of passing on informationabout a culture. In organizational settings,stories are defined as ‘‘narratives based ontrue events—often a combination of truthand fiction’’ (Trice & Beyer, 1984, p. 655).Employees in organizations tell many stories,some of which may be completely irrelevantto cultural transmission. What makes a storya vehicle for cultural transmission is that itis intentionally meant to convey somethingimportant about the culture of the organiza-tion—in many cases, to organizationalnewcomers. A good example of this can beseen in a brief story contained in Derek Jet-er’s 2001 book The Life You Imagine: LifeLessons for Achieving Your Dreams. In thisbook Jeter describes an instance during hisrookie year when he and Don Mattingly(former Yankee great and current benchcoach) are leaving the field at the end of aworkout during spring training. Eventhough the two players were alone and couldleisurely walk off the field, Jeter states thatMattingly suggests they sprint off the field(which they do) because ‘‘you never knowwhose watching.’’ This story obviously sayssomething about the integrity of Mattingly,but more importantly, about the culture ofthe team. More specifically, hard work andeffort are not just for show, but are importantall the time.

A legend is a ‘‘handed-down narrative ofsome wonderful event that is based on his-tory, but is embellished with fictionaldetails’’ (Trice & Beyer, 1984, p. 655). Inschools throughout the United States, chil-dren learn about how the founding fathers,


such as Thomas Jefferson, Benjamin Frank-lin, and Alexander Hamilton, cooperatedto produce the Constitution, and what thissignifies about our national culture. As his-torical analysis has shown, however, theprocesses surrounding the development ofthe U.S. Constitution were anything butcooperative (Tolson, 2001, February 26).Many of the framers of the Constitution werehighly political and very self-interested. Infact, the partisan politics seen today is rela-tively tame in comparison. Notice, however,that teaching this slightly inaccurate versionof history does serve to transmit culturalvalues that are important in a democracysuch as the United States.

Legends are also used in organizations toconvey important cultural details. The specificlegends passed on typically focus on impor-tant milestones such as the founding of theorganization, a critical organizational crisis, oran important innovation that has had a greatimpact on the organization. Within 3M, thedetails surrounding many product innova-tions, such as a Post-it1 note, take on alegendary status, and the individuals respon-sible for these innovations are seen as almostlarger than life (Gundling, 2000). Passing onthese legends to new employees within 3Mserves the purpose of communicating the factthat innovation and creativity are importantparts of the culture.

A final mechanism for the transmissionof culture is through what Pettigrew (1979)termed organizational dramas. An organiza-tional drama is simply a significant or defin-ing event in the history of an organization.The occurrences of organizational dramasprovide researchers with a window into theculture of an organization; relating thesedramas to new organizational members alsoprovides a way for organizations to transmitorganizational culture to newcomers. Ac-cording to Pettigrew, the most typical orga-

nizational dramas are the entry and exit oforganizational leaders, changes in the struc-ture of an organization, and noteworthy suc-cesses or failures.

Language and Communication

Language is one of the key things that dis-tinguish humans from other species. It standsto reason, then, that the culture of an orga-nization would be reflected in the languageof organizational employees; in fact, eachorganization typically has its own uniquevernacular. Similarly, the manner in whichemployees in an organization communicatewith each other may also reveal importantinformation about an organization’s culture.Each of these is discussed in the followingparagraphs.

With respect to spoken language, organi-zational terminology can be quite revealing.Several years ago one of the authors conducteda brief training seminar in an organizationin which virtually all employees referred totheir various departments as worlds ratherthan by more standard terms such as depart-ments or units. Although use of this terminol-ogy may have been completely coincidental, italso could have been indicative of a great dealof ‘‘turf battles’’ and compartmentalizationwithin the organization. Another example ofthis, which may be familiar to many readers, isDisney’s long-standing practice of referring topark visitors as guests rather than customers(Van Maanen, 1991). This signifies that peo-ple who pay to visit the Disney theme parksshould be treated by employees as thoughthey were visitors in their homes. Disney hasalso traditionally used theatrical terminology(e.g., employees are cast members) to reinforcethe point that they are in the business ofproviding entertainment.

The mode of communication used byemployees in organizations can also provide


insights into organizational culture. Employ-ees in some organizations favor highlyimpersonal one-way modes of communica-tion such as written messages, voice mail,and e-mail. What does reliance on theseforms of communication suggest about orga-nizational culture? It may simply mean thatpeople want to save time. However, it couldalso mean that people really do not want tocommunicate with each other. While imper-sonal modes of communication such ase-mail are very efficient, they also get em-ployees into the habit of ‘‘issuing directives’’and making ‘‘declarations’’ to their fellowemployees rather than engaging in two-waycommunication and meaningful dialogue. Asa result, this may be indicative of aculture characterized by high suspicion andconflict.

In contrast, the preferred mode of com-munication in some organizations is muchdifferent. Employees may favor highly per-sonal, face-to-face communication ratherthan more impersonal modes such as e-mailor written memos. In terms of organizationalculture, this may indicate that there is a greatdeal of emphasis on interpersonal harmonyand on making sure that others’ feelings areconsidered when making decisions. Thismay also indicate a highly participative cul-ture in which a great deal of consultationmust take place prior to decisions beingmade.

THE DEVELOPMENT OFORGANIZATIONAL CULTURE

According to Schein (1992), the two majorfunctions of organizational culture are exter-nal adaptation and internal integration. Thenotion of external adaptation reflects ananthropological, or even an evolutionary,approach to organizational culture. To cul-tural anthropologists, the totality of a culture

reflects behaviors and beliefs that have sur-vived over time because they have helped agroup of people adapt more successfully totheir environment. This obviously has evolu-tionary overtones because adaptation is acentral part of the evolutionary process.

When we apply the concept of externaladaptation, we come up with the propositionthat organizational cultures develop and per-sist because they help an organization tosurvive and flourish. This concept is quiteeasy to illustrate if one looks at organizationsthat possess cultural attributes that mostobservers would consider very positive. Forexample, developing a culture that empha-sizes innovation kept 3M from going out ofbusiness, and this continues to help it remainone of the most successful corporations inthe world. Similarly, developing a culturethat puts customer service and comfortabove all else helped Disney make the tran-sition from a small film-animation companyto a large entertainment conglomerate.

External adaptation can also explain whysome organizations ultimately develop cul-tures that possess what some would considernegative attributes. According to Mason 2004,the organizational culture of NASA changedfrom one that strongly emphasized safetyand technical excellence, to one primarilyconcerned with cost efficiency and meetingschedules. When one considers the pres-sures faced by this organization over theyears (e.g., decreased congressional funding,pressure to meet launch deadlines), thischange in culture is understandable. How-ever, as Mason points out, the price of thisemphasis on efficiency has been less vigi-lance about safety issues; in fact, this mayhave contributed to some of its failures suchas the Challanger and Columbia space shuttleaccidents.

In addition to facilitating externaladaptation, Schein (1992) proposed that


organizational culture facilitates internalintegration. Consider for a moment how anorganization could function if it had no iden-tifiable culture. In such a scenario, howwould new members be integrated into theorganization and taught how to assume theirnew roles? Thus, culture can be thought of asa sort of ‘‘glue’’ that bonds the social struc-ture of a larger organization together. This iscritical because, when all is said and done,organizations are ultimately social construc-tions and, without social integration, theywould cease to exist (Katz & Kahn, 1978).

This integrative function can be seen atvarious levels of an organization, and thusserves as an explanation for the developmentof organizational subcultures. Furthermore,some of these subcultures may result fromindividuals in a particular department orfunction sharing common experiences orsimilar academic training. Because of this,the various areas that are represented bygraduate programs (e.g., Clinical, I/O, Social,Experimental) within large psychology de-partments often develop very distinct sub-cultures based on commonality of academictraining and experiences. The developmentof these subcultures, provided they aren’t atodds with each other, increases social cohe-sion within these areas and enhances theprofessional socialization of graduate stu-dents. The existence of subcultures doesnot mean, however, that there is no overalldepartmental culture. Regardless of the area,all faculty typically have had at least someoverlap in their training as doctoral-levelpsychologists.

A final factor that often shapes an orga-nization’s culture is its founder or its chiefexecutive. How do influential founders andhigh-level executives put their own ‘‘per-sonal stamp’’ on the culture of an organiza-tion? This question has not achieved a greatdeal of empirical attention, but several mech-

anisms are possible. One is that these indi-viduals have a great deal of control over whois hired, particularly at the highest levels.Because people generally like to be in thecompany of others whom they perceive to besimilar to them (Byrne, 1971), it is highlylikely that employees hand-picked by afounder or chief executive have similar val-ues. Furthermore, because those who reallydidn’t share his or her values either declinedto join the company or ultimately left(Schneider, 1987), those remaining prob-ably shaped a culture that was very similarto his or her personality.

Founders and influential executives alsohave a great deal of influence over the strategyan organization decides to pursue (Finkel-stein, 1992). Choice of strategy, in turn, mayultimately impact the culture that develops inthe organization. For example, an organiza-tion that chooses to pursue a strategy of offer-ing a very limited number of highlyspecialized products will likely develop a verydifferent culture, compared to an organizationwhere the primary source of competitive ad-vantage is high-quality customer service. Inthe former case, the culture that develops mayplace a premium on technical expertise. Incontrast, in the latter case, a culture maydevelop that places a much higher value onsocial skills and the reduction of conflict.

A final issue to consider, particularlywith respect to founders, is whether theycontinue to impact the culture of an organi-zation when they are no longer involved withit on a day-to-day basis (e.g., after retirementor death). Again, little empirical research hasexamined this issue. However, based on whatwe do know about culture, the legacy of anorganizational founder may be reflected inthe culture for quite some time. That is,through processes of cultural transmission(e.g., rites, stories), cultures will typically per-petuate themselves, and thus outlive the

The Development of Organizational Culture 455

founding member of the organization. This isparticularly true if the original culture of theorganization has led to success and is thusseen as having some adaptive value. Disney isa good example of an organization that hasworked hard to preserve the legacy of thefounder, Walt Disney, and it has been verysuccessful in doing so.

MEASURING ORGANIZATIONALCULTURE

To scientifically determine the effect of orga-nizational culture, we need to be able tomeasure it and to do so with a great deal ofprecision. Like many variables in organiza-tional psychology, organizational culture isvery complex and thus very difficult to meas-ure. In this section, we examine common ap-proaches to measuring organizational culture.

Self-Report Assessments of Culture

The most direct way to measure the cultureof an organization is to create some type ofself-report measure, administer this measureto a sample of organizational employees, andthen create a numerical index to describe theculture. The most popular self-report meas-ure of organizational culture over the yearshas been the Organizational Culture Profile(OCP), which is based on the previouslydescribed model of organizational cultureproposed by O’Reilly, Chatman, and Cald-well (1991). The OCP measures employeeperceptions of the predominate valueswithin an organization, which are summa-rized in Table 14.2. Because the OCP pro-vides measures of organizations and notindividuals, the scores for each of these val-ues are typically formed by averaging indi-vidual employees’ ratings.

Another relatively common self-reportmeasure of organizational culture is Hofstede’s

(1980) measure of organizational values. Thisself-report instrument, which is based onHofstede’s work on differences in nationalcultures, assesses the following potentialorganizational values: process-orientedversus results-oriented; employee-orientedversus job-oriented; parochial versus profes-sional; open system versus closed system;loose control versus tight control; normativecontrol versus tight control. As with theOCP, individual employees’ scores are aggre-gated to come up with the scores for theorganization. The organization’s unique cul-ture is then determined by examining thepattern of the scores on this instrument.

The most recent self-report measure ofculture is the Denison Organizational CultureSurvey (Denison, Cho, & Young, 2000).This measure consists of 60 items that aredesigned to measure the 12 dimensions ofthe previously described Denison model oforganizational culture (see Fig. 14.1). Likethe other measures described, responsesof employees in an organization are aver-aged. While this measure is still relativelynew, compared to the OCI and Hofstede’smeasure, Denison and colleagues haveshown it to be a useful diagnostic tool inmany types of organizations.

Self-report measures of organizationalculture are relatively easy to administer,and they provide quantitative indexes thatresearchers can use to describe and compareorganizational cultures. Unfortunately thereare also some serious limitations associatedwith self-report assessments of organiza-tional culture. Recall that the essence oforganizational culture is the basic assump-tions shared by the employees in an organi-zation. Because these basic assumptions arerarely questioned by employees, they are, toa large extent, unconscious. Thus, employ-ees who are immersed in the culture of anorganization are probably going to be able to


report only the surface aspects of that cul-ture, such as values, which are exactlyassessed by self-report measures.

Self-report measures of culture may alsobe problematic because they impose a some-what arbitrary structure on the respondent.Although researchers have found that certainvalues or dimensions of organizational cul-ture are important to assess, there may beothers that are more specific to a given orga-nization and contribute greatly to its culture.For example, compared to a business orga-nization the culture of an educational insti-tution may be heavily influenced by highlyunique external factors such as the level ofeducational funding that is provided by thestate government. These highly specific fac-tors are typically not measured in standardself-report culture measures.

A more serious problem with self-report isthat there is no way of assessing whetherrespondents are describing the actual cultureof the organization or the idealized cultureof the organization. In many organizations,there is a great deal of difference betweenwhat employees would like the culture to beand what it actually is. Employees completingself-report measures may very well report,essentially, an idealized version, and not thereality, of the culture. This occurs simplybecause of the many weaknesses inherent inself-report measurement (e.g., Spector, 1994)and the fact that employees, particularly thoseat higher organizational levels, may have‘‘blind spots’’ regarding the culture.

Ethnographic Methods of CultureAssessment

Ethnography is the use of qualitative, obser-vational methods of assessing behavior.Researchers conducting ethnographic assess-ment of organizational culture (herein refer-red to as ethnographers) typically observe

and record behavior in an organization foran extended period of time. In some cases,ethnographers present themselves as outsideresearchers. In other cases, however, ethnog-raphers may actually become members ofthe organizations they are trying to analyze.The most notable example of this type ofresearch in the organizational literature wasVan Maanen’s analysis of police culture (VanMaanen, 1975). In order to study this organi-zation, Van Maanen actually went through apolice academy as a recruit and recorded hisobservations.

Other than direct observation, anothertool often used by ethnographers is inter-viewing informants (Johnson, 1990). Aninformant is a member of the organizationto whom ethnographers can go for informa-tion. In many cases, informants help ethnog-raphers make sense out of what they haveobserved in the organization. According toJohnson, there is no ideal informant in anyethnographic study; however, it is obviouslyimportant that any informant should possessa detailed knowledge of the organizationbeing studied.

When choosing an organizational inform-ant, ethnographers often seek out long-tenured employees. Indeed, these individualsmay be very helpful because they are able toprovide a historical context for understandingmuch of what goes on in an organization. Apotential drawback of long-tenured employ-ees, though, is that they may be so immersedin the culture that they are unable to describeit accurately. The ‘‘first impressions’’ of a rel-atively new employee may ultimately provideas much (or more) insight into an organiza-tion’s true culture. The best course of actionfor ethnographers, if possible, is to seek orga-nizational informants who represent a varietyof tenure levels.

The obvious benefit of ethnographicassessment of organizational culture is that

Measuring Organizational Culture 457

it does not require the researcher to directlyask employees about the culture of the orga-nization. If we accept the notion that orga-nizational culture represents the shared‘‘basic assumptions’’ of employees, then qua-litative methods are more likely than self-report measures to capture these assump-tions, simply because basic assumptions areat a level of consciousness that is very diffi-cult for employees to access. Thus, moreinformation about culture can probably begleaned from observing employees’ behav-iors, rather than directly asking them ques-tions. Unfortunately, ethnography is a labor-intensive and, at times, painstaking process.Many researchers do not have the time toobserve an organization for long periods orthe capability of coding all of those observa-tions. There is also a potential for observerbias in ethnographic research. There areways that ethnographers can address thisissue (e.g., via informants or multiple obser-vers), but observation ultimately involves agood deal of subjectivity.

Other Methods of CulturalAssessment

By far, the most common methods of cul-ture assessment are self-report surveys andethnography. Given the vast methodologicaltool-kit available to organizational psychol-ogists, however, there are certainly otherways culture could be assessed. One meth-od, which is not used very often in orga-nizational culture research, is the use ofarchival information from the organization.Most organizations produce a good deal ofarchival information, and some of this mayprovide clues about culture. For example,an organization’s annual report could beanalyzed through content analysis to pro-vide information about culture. If a goodportion of the text of the annual report

deals with customer service, this is a signthat customer service is a major part of anorganization’s culture. Similarly, if all of anorganization’s top executives are long-tenured employees, this may be a sign that,in the organization, a strong value is placedon experience.

Another method, which has not beenused frequently, is to assess culture throughmeasuring employees’ cognitive maps of theorganizations (see Silvester, Anderson, &Patterson, 1999). Cognitive mapping is sim-ply a way of determining the underlyingheuristics that employees use to process in-formation about the organization. To con-struct cognitive maps, employees areinterviewed, and the information from theseinterviews is subjected to a standardizedcoding process. Although this is a relativelynew process, it certainly holds great promisefor future researchers as a method of cultureassessment.

CHANGING ORGANIZATIONALCULTURE

So far, in defining organizational culture, wehave emphasized the values and basicassumptions that have been passed downthrough many generations and, as a result,are shared among employees. Another keypoint in the previous discussion is that orga-nizational cultures do not develop in a ran-dom fashion; rather, they develop and aresustained over time because they help anorganization adapt to its competitive environ-ment. However, what happens when thatcompetitive environment changes? The attri-butes of an organization’s culture that helpedit compete in the previous competitive envi-ronment may be irrelevant, or perhaps evencounterproductive, in the new environment.Thus, at times organizations have to changetheir cultures in order to survive.


Even if their competitive environment isrelatively stable, organizations may attemptto change their cultures for other reasons.Often, the desire for culture change accom-panies significant changes in the top manage-ment team of an organization. In such caseschange is driven more by the desires of thosein top management than by necessity. Cul-tures also change because the people in anorganization change over time. Although it isoften argued that this process favors culturalstability (e.g., Kristof (1996); Schneider,1987), it may not always be the case. Whendifferent people come into an organization,they may gradually change its interpersonaldynamics and, ultimately, its culture.

Many of the organizational developmentinterventions that will be described in Chap-ter 15 are ultimately aimed at changing theculture of an organization, so the topic willnot be covered in great depth here. In thissection, however, two important questionsabout culture change are addressed:

1. Why is changing the culture of an org-anization so difficult?

2. What are some of the common mecha-nisms by which organizational cultureschange?

Why Is Culture Change Difficult?

In almost any comprehensive treatment oforganizational culture it is concluded thatorganizational culture is hard to change onceit has been established (Denison, 1990;Hatch, 1993; Schein, 1985, 1992). That’snot to say that organizational culture iscompletely intractable; in fact, over a longperiod of time most organizational culturesevolve and change, due to a number of fac-tors that will be covered in the next section.What is difficult, however, is for organiza-tions to change their cultures very quickly. A

manager cannot simply write a memo toemployees on a Friday informing them that,as of Monday, the culture will be different(though some organizations may naıvelythink this is possible).

One reason that it is difficult to changethe culture of an organization has to do withthe definition of culture that was presentedat the beginning of the chapter. As waspointed out, the essence of organizationalculture resides in the basic assumptionsshared by employees. Recall that basicassumptions can be about anything, butthose relevant to organizational culture typ-ically have something to do with the organi-zation and its major activities. What makesthese assumptions ‘‘basic’’ is that they areshared among employees and, as such, arerarely if ever questioned or put under objec-tive scrutiny.

Because of this, ‘‘basic assumptions’’ arehighly resistant to change. Furthermore, inthose cases when basic assumptions are chal-lenged in organizations, the challenge mayactually serve to strengthen employees’beliefs in those basic assumptions. For ex-ample if a new employee refuses to accept thebasic assumptions that are inherent in anorganization’s culture, this will typicallyforce other employees to ‘‘bring that personinto line.’’ In the process of doing so, the corevalues and assumptions may be strength-ened, regardless of whether that personeventually accepts them, actively resiststhem, or ultimately leaves the organization.Although basic assumptions are viewed bymost organizational culture experts as beinghighly resistant to change, it should be notedthat this view is not shared by all (see Com-ment 14.3).

Another reason organizational culture isdifficult to change is because there are typ-ically those who stand to benefit if theculture remains static. A logical corollary


to this is: There are often some who stand tolose (or perceive this to be the case) if theculture is changed. Perhaps the best way toillustrate this is through one of the mostfundamental assumptions shaping organi-zational culture: how employees should berewarded. Let’s say that one of the mostbasic assumptions of an organization’s cul-ture is that rewards should be based pri-marily on seniority. Further assume that anew organizational president is hired and isdetermined to change the culture to one inwhich rewards are instead based primarilyon performance.

Given this scenario, consider first theissue of who benefits from the present cul-ture of this organization. It is very clear thatthe primary beneficiaries of the present cul-ture are those who have been employedin the organization for a long period of

time, assuming of course that the organiza-tion’s reward policies actually reflect theculture. Now consider who stands to loseif the culture of the organization changesand performance is then valued above allelse. Not all long-tenured employees will behurt by this change, because some of theseindividuals may be among the organiza-tion’s best performers. Rather, employeeswho perform their jobs poorly, regardlessof tenure, stand to lose the most from such aculture change.

Among those actively resisting sucha change, however, long-tenured poor-performing employees would probably bethe most vigorous. These individuals mayresist such a culture change by actively argu-ing against it, or even failing to follow poli-cies that are based on it. These individualsnot only must give something up (rewards

COMMENT 14.3

IS ORGANIZATIONAL CULTURE REALLY SO DIFFICULT TO CHANGE?

IN NEARLY ALL reviews of organizational culture,

one of the common assertions is that once the

culture of an organization is established it is

extremely difficult to change. This is because,by definition, culture represents beliefs and

assumptions that are so rarely questioned that

they are not even conscious. Thus, it takes a

lot of effort to get people to question basic

assumptions and, in the process, change cul-

ture.

Although this is by far the dominant view-

point, there are some organizational cultureresearchers that disagree. Wilkins and Ouchi

(1983), for example, point out that the idea of

organizational culture being difficult to change

comes from cultural anthropology. Cultural

anthropologists, as many readers know, are

primarily interested in societal cultures. Soci-

etal cultures are obviously very difficult to

change because most people become totally

immersed in their societal culture.

In the case of organizational culture, how-

ever, there is a great deal of variation in thedegree of ‘‘enculturation.’’ Some employees do

become ‘‘true believers’’ and faithfully espouse

the values and assumptions of their organiza-

tion. On the other hand, employees may be

very much opposed to the values and assump-

tions of their organization. Most employees are

probably somewhere between these two

extremes. Given this variation in employeeenculturation, it is probably easier to change

the culture of an organization than an entire

society.

Source: A. L. Wilkins and W. G. Ouchi. (1983). Efficient

cultures: Exploring the relationship between culture and

organizational performance. Administrative Science Quar-

terly, 28, 468–481.


based on seniority), but they may also behurt again (based on their performance, theywill not be highly rewarded) by the culturechange. Regardless of how bad or dysfunc-tional an organizational culture may appear,there are typically those who benefit fromhaving it remain that way, and those whostand to lose by changing it. Many attemptsto change organizational culture end up infailure because those initiating the changehaven’t recognized this.

A final reason that organizational culturecannot be easily changed goes back to thefactors that shape culture in the first place.Recall that the most important of these isadaptation. Cultures develop and flourishover time because they serve some purposeor help some group adapt more effectively toits environment. That’s not to say that mal-adaptive cultures never develop. For themost part, though, cultures remain stablebecause they serve some adaptive function(Mason, 2004). It follows, then, that super-ficial or misguided attempts to change orga-nizational culture would probably encounterresistance.

The Nature of OrganizationalCulture Change

According to Schein (1992), organizationsare like individuals in that they pass throughdistinct ‘‘life’’ stages. These stages are impor-tant because they help us to understand howorganizations change and evolve over time.During the Birth and Early Growth phase, theorganization is founded and is beginning todevelop a distinct culture. As one mightimagine, during this phase, organizationalculture is strongly impacted by the organi-zational founder(s) or the family of thefounder. Such individuals can often literallyhire or fire at will and are in a position to

demand a great deal of loyalty. Furthermore,just to survive, organizations at this stagemay demand a great deal of commitmenton the part of employees. Also, at this stage,when the organization is most vulnerable,external events can potentially have greateffects on the organization and, in fact,become part of organizational folklore.

In the second stage, Organizational Mid-life, an organization typically becomes ‘‘big-ger’’ structurally. This may also be a time ofgrowth and expansion, as organizationsdecide to explore new markets or productlines. With respect to organizational culture,the great structural complexity that oftenaccompanies this stage may result in a num-ber of organizational ‘‘subcultures.’’ Thesesubcultures may be based on a number ofthings, such as geographic location, productlines or divisions, or even functional special-ties. The obvious danger at this stage is thatthe subcultures may become so distinct thatthe organization begins to lose its more gen-eral, overarching culture.

The third and final stage in this model isOrganizational Maturity. This is essentiallythe ‘‘crossroads’’ in the life of an organiza-tion. At this point, an organization is oftenfaced with the choice of renewal (e.g., con-tinuing on indefinitely) or stagnation and,ultimately, death. In this sense, organiza-tions have an advantage over people—theycan live on indefinitely whereas people can-not. Organizational culture is a key factor indetermining this choice between renewaland stagnation. Organizations that fail tochange any aspect of their cultures stand agood chance of failure. On the other hand,organizations that live on indefinitely mustdecide which aspects of their cultures needto be changed and which ones need to bepreserved over time.

Schein (1985, 1992) proposed a numberof mechanisms by which organizational


cultures change over time within the contextof these organizational life stages. These aresummarized in Table 14.4. During the Birthand Early Growth phase, organizational cul-ture may change through any of four primarymechanisms. The mechanism of natural evo-lution represents the processes that shape anorganizational culture when it attempts toadapt to its environment. When viewed ina general sense, this simply represents thoseaspects of organizational culture that con-tribute to its survival. For example, duringthe early life of an organization, changingfrom a highly autocratic to a highly collabo-rative culture is adaptive, and this maybecome part of the organization’s permanentculture.

A second mechanism that frequentlyleads to culture change in the early stagesof an organization is referred to by Schein as‘‘self-guided evolution through organiza-tional therapy’’ (2003). In other words, thereis a deliberate attempt or intervention con-ducted to change the organizational culture.The term organizational therapy is used torefer to a variety of interventions (some of

which will be described in Chapter 15) thatare designed to facilitate culture change. Asan example, the top management of a rela-tively new organization may decide that theculture of the organization should be veryteam oriented. In this case, the therapy usedto achieve this culture change might be in theform of training on topics such as team de-cision making or resolving interpersonalconflict in teams. In other cases, organiza-tions use reorganization or restructuring inorder to change a culture (Hannan, Polos, &Carroll, 2003).

Another mechanism used for culturechange is referred to by Schein as ‘‘managedevolution through hybrids.’’ In this case, cul-ture change is also initiated intentionally,although the mechanism is much differentfrom the one in the previous example. In thiscase, the mechanism is through the appoint-ment of ‘‘hybrids’’ in key positions within theorganization. A hybrid is an individual whohas grown up in the present organizationalculture but, at the same time, may not acceptall of the underlying assumptions on which itis based. By putting these types of individuals

TABLE 14.4Culture Change Mechanisms at Different Stages of the Organizational Lifecycle

Stage Change Mechanisms

1. Birth and Early growth 1. Natural evolution

2. Self-guided evolution through organizational therapy

3. Managed evolution through hybrids

4. Managed ‘‘revolution’’ through outsiders

2. Organizational midlife 1. Planned change and organizational development

2. Technological seduction

3. Change through scandal and explosion of myths

4. Incrementalism

3. Organizational maturity 1. Coercive persuasion

2. Turnaround

3. Reorganization, destruction, and rebirth

Source: E. H. Schein. (1985). Organizational culture and leadership: A dynamic view. San Francisco: Jossey-Bass.

Copyright 1985, Jossey-Bass. Reprinted by permission of John Wiley & Sons.


in key positions, the culture may not changeradically, but it may shift in a way that isultimately more adaptive for the organization.

The final mechanism used for culturechange during the early life of an organiza-tion is referred to as ‘‘managed revolutionthrough outsiders.’’ This mechanism is likethe one just described, except that the‘‘agents of change’’ are individuals from out-side of the organization who are much lessfamiliar with the organizational culture thanorganizational insiders. Bringing these typesof individuals into an organization canpotentially initiate a great deal of culturechange because they are likely to questionmany of the basic assumptions on which thecurrent culture is based. While this may be adifficult process, for both current employeesand the outsiders that are brought in, it mayhelp an organization rethink outdated ordysfunctional aspects of its culture and ulti-mately lead to culture change.

By the time an organization reaches themidlife stage, the culture is relatively wellestablished and different mechanisms maybe needed to facilitate change. The first ofthese, in Table 14.4, is referred to as plannedchange and organizational development. Thisrepresents a deliberate attempt to guideand facilitate the change process. This is asign of organizational maturity because itshows recognition that adaptation is neces-sary for success. While all organizational-development programs have somewhatdifferent goals, the fundamental purpose ofmost is to change the culture of the organi-zation, or at least provide an organizationwith the capability to do so. Organizationsusing this type of strategy typically bringin outside consultants, although, in somelarge organizations, an organizational-development function may be established.

The second change mechanism in orga-nizational midlife is referred to as technolog-

ical seduction. This refers to the use oftechnology as a lever for organizational cul-ture change, and it may occur in two differ-ent ways. Technology may driveorganizational change because of the tech-nologies that emerge within the organiza-tion. For example, a ‘‘high-tech’’ culturemay develop in a computer company, dueto the types of employees needed to fill manyof the positions in such an organization. Inaddition, organizations can sometimesinduce culture change by introducing newand unfamiliar technology. The idea thattechnology can shape the social environmentis well known and can be traced back to thesociotechnical systems perspective and theTavistock studies of coal mining (Trist &Bamforth, 1951).

A third mechanism of culture changeduring organizational midlife is throughscandal and the explosion of myths. Forexample, a scandal involving an organizationmay force organizational members to rethinksome of their basic assumptions, which mayultimately lead to culture change. This mayoccur, for example, when a charismatic lead-er in an organization is caught engaging inillegal behavior. On a societal level, one couldcertainly argue that the Watergate scandal inthe early 1970s led many to rethink theirassumptions about government officials. Ulti-mately, this has led to a great deal of mistrustand skepticism toward these people.

The explosion of myths occurs when oneof the generally accepted organizationalmyths is publicly proven to be false. As anexample, a myth commonly held in manyorganizations is that employees’ jobs aresecure. If layoffs do occur, this results in anexplosion of this myth and the culture of anorganization may change as a result. At thesocial level, a myth that persisted amongAmericans was that there was little possibil-ity of terrorism being carried out on our own


soil. Most thought that terrorism was some-thing that occurred only in the Middle East.The events on September 11, 2001, obvi-ously shattered this myth, and it is probablyone of the reasons that people were soshocked by these acts.

The final mechanism for change duringorganizational midlife is referred to as incre-mentalism. This means that change doesoccur, but it occurs very slowly. For exam-ple, most organizations in the midlife stagehave employees who represent a variety oftenure levels. Some have been with the orga-nization for a long time, others have beenaround for a few years, and others are new.Over time, as new employees come into anorganization and others either retire or leave,the organization will undoubtedly change,although in subtle ways. As an example,academic departments in many universitiesare changing because a large number of fac-ulty hired during the early 1970s are nowretiring. These changes are incremental,however, because all of these individualsare unlikely to retire at once. New facultyare brought in gradually, and the change isoften very subtle and hard to detect.

During the final stage, organizationalmaturity, an organization is really faced withthe choice of stagnation/decline or changingin ways that will facilitate its renewal. Thus, achange in organizational culture may be avery critical issue. One way that change maybe achieved at this point is through whatSchein (1985) described as coercive persua-sion. In this case, organizations use a varietyof coercive tactics to facilitate changes inindividuals, which will ultimately lead tochanges in the culture. A common way thatorganizations use this mechanism is by pro-viding long-tenured employees with theoption of early retirement. Another way thatorganizations may facilitate change in thismanner is through the threat of undesirable

work assignments or by altering workingconditions in ways that are undesirable toany employees who will not change.

The second change mechanism duringOrganizational Maturing is described bySchein (1985) as turnaround. To a large extent,turnaround embodies many of the changemechanisms that were previously described.During turnaround, the organization recog-nizes the need for a cultural change and takesthe steps necessary for the change to occur. Inmany cases, this may be through the applica-tion of organizational-development methods,but it could also be through a change in per-sonnel. As Schein (1985, 1992) points out, forturnaround to be successful, it must be acomprehensive effort and involve all membersof the organization.

The final change mechanism in organi-zational maturing is referred to as reorganiza-tion, destruction, and rebirth. This is probablythe most extreme form of culture changebecause it essentially involves destroyingthe present culture and instituting a newone. Given the extremity of this method, itis typically reserved for times of crisis ortimes when the only alternative to culturechange is failure. An example of this changemechanism can often be seen in the federalgovernment when the U.S. president isreelected to a second term in office. Specifi-cally, many of the cabinet members and keymembers of the administration from the firstterm resign or are replaced with new ap-pointees. The impact of such changes, onewould assume, is to change the culture sur-rounding the administration, and henceenhance its effectiveness.

THE IMPACT OFORGANIZATIONAL CULTURE

So far, we have examined a variety of issuespertaining to organizational culture—how it


is defined, how it is measured, how itchanges—and two models that provide acomparison of different organizational cul-tures. However, some important questionshave yet to be examined: Does organizationalculture make a difference in important orga-nizational outcomes? Do organizations withcertain cultural attributes tend to be moresuccessful than organizations without suchattributes? Do organizations with culturalattributes tend to attract, hire, and retainbetter employees than organizations withoutsuch attributes? Do employees in organiza-tions with certain cultural attributes tend tobe more satisfied and to have a better qualityof work life than employees in organizationswithout such attributes? In this section, abrief summary of research evidence bearingon each of these questions will be provided.

Culture and OrganizationalPerformance

The issue of linking organizational culturewith organizational performance has cer-tainly received some attention (e.g., Denison,1984; Denison & Mishra, 1995; Peters &Waterman, 1995; Wilkins & Ouchi, 1983).Early attempts to do this, as in Peters andWaterman’s 1982 best-selling book In Searchof Excellence, were focused primarily on qual-itative comparisons of successful and unsuc-cessful companies. While such an approachcan provide useful insights, it is certainlylimited from a methodological perspective.Thus, more recent research has attempted tocorrelate dimensions of organizational cul-ture with organizational performance.

Kotter and Heskett (1992) conducted whatis perhaps the most comprehensive empiricalstudy.They investigated207U.S.organizationsspread among 25 different industries. Theymeasured the strength of organizational cultureand examined how this related to a number of

performance indexes, such as revenues, stockprice, expansion of the work force, and netincome. The fact that culture and performanceweremeasuredusingdifferentsourcesisimpor-tant because it decreases the possibility thatculture and performance were related simplybecause of a common-method bias.

The results of this study suggest thatorganizational culture does make a differ-ence in bottom-line organizational perfor-mance. For example, organizations withcultures that these authors labeled adaptiveperformed much better than organizationswith cultures labeled unadaptive. The majordifferences between adaptive and unadap-tive cultures are highlighted in Table 14.5.When one looks at these differences, itbecomes fairly clear that an organizationwith an adaptive culture would be a muchmore enjoyable place to work, compared toan organization that is unadaptive. Further-more, from the results of this study, itappears an adaptive culture translates intoorganizational success.

Denison and colleagues have conductedresearch that is very similar to Kotter andHeskett (1992), where they correlate dimen-sions of culture with organizational perfor-mance. Table 14.6 contains correlationsbetween the 12 culture dimensions fromthe Denison model and overall effectivenessfrom Denison, Haaland, and Goelzer (2004).These correlations are based on responsesfrom 36,820 individuals from 210 organiza-tions in three regions of the world (NorthAmerica, Asia, and Europe). As can be seen,all culture dimensions are positively corre-lated with organizational effectiveness, butnot to the same degree. For example, in NorthAmerican organizations the three strongestpredictors of effectiveness were CapabilityDevelopment, Coordination, and Empower-ment. The weakest were Customer Focusand Creating Change. In Asian organizations


none of the dimensions were significantlycorrelated with overall effectiveness becauseso few (n ¼ 7) were included in the study.However, based on the magnitude of thecorrelations it appears that the best predic-tors were Creating Change, OrganizationalLearning, and Team Orientation. The weak-est were Customer Focus and CapabilityDevelopment. Finally, in European organi-zations the three strongest predictors oforganizational performance were StrategicDirection and Intent, Coordination and Inte-gration, and Agreement. The weakest wereCapability Development and OrganizationalLearning.

Like Kotter and Heskett’s (1992) study,Denison’s results suggest that culture doescontribute to the success of an organization,though not all dimensions contribute thesame. Denison’s data also suggest that theimpact of organizational culture is impactedby national culture. This is often overlookedin organizational culture research, but it isvery important given that many organiza-tions have global operations.

Probably the most important issue infuture research in this area will be an attempt

to explain the mediating linkages betweenculture and organizational performance. Ithas been proposed, for example, that orga-nizational culture may impact the level ofemployee creativity (Tesluk, Farr, & Klein,

TABLE 14.5Key Differences between Adaptive and Unadaptive Corporate Cultures

Adaptive Corporate Cultures Unadaptive Corporate Cultures

Core values Managers care deeply about customers,

stockholders, and employees. They

place a high value on people and proc-

esses that create useful change.

Managers care about themselves, their

immediate work group, or some product

or technology. They value the orderly

and risk-reducing management proc-

esses.

Common behavior Managers pay close attention to all their

constituencies, especially customers; ini-

tiate change when needed; take risks.

Managers behave politically and

bureaucratically. They do not change

their strategies quickly to adjust to or

take advantage of changes in their

business environments.

Source: J. P. Kotter and J. L. Heskett. (1992). Corporate culture and performance. New York: Free Press. Reprinted with

the permission of The Free Press, a division of Simon & Schuster, Inc. Copyright # 1992 by Kotter Associates Inc.

and James L. Heskett.

TABLE 14.6Correlation between dimensions from Denison’sOrganizational Culture Model and OverallEffectiveness by Region

DimensionNorth

America Asia Europe

Empowerment .65� .57 .60�

Team orientation .61� .71 .53�

CapabiIity development .70� .48 .50�

Core values .58� .62 .73�

Coordination and integration .69� .62 .74�

Creating change .48� .87 .68�

Customer focus .36� .19 .62�

Organizational learning .50� .82 .52�

Strategic direction/intent .55� .66 .79�

Goals and objectives .60� .54 .62�

Vision .53� .71 .67�

Number of organizations 169 7 34

Note: �p < .05

Source: Denison, D.R, Haaland, S., & Goelzer, P. (2004).

Corporate culture and effectiveness: Is Asia different than

the rest of the world? Organizational Dynamics, 33, 98–109.


1997), the strength of employee motivation(Weiner & Vardi, 1990), and the reportingof unethical behavior (Ellis & Arieli, 1999).Much more work is needed, however, toexplain why culture makes a difference inorganizational performance.

The Impact of Culture onRecruitment and Retention

Compared to the literature on organizationalculture and performance, much more empir-ical research has investigated the impact thatorganizational culture has on attracting,recruiting, and retaining employees. (Thisissue was examined in some detail in Chap-ter 3 and will not be covered extensivelyhere.) The basic finding in both of these areasis that individuals tend to be attracted toorganizations that possess cultures that theyperceive to be compatible. Furthermore,once people are in organizations, they willtend to remain in organizations that theyperceive to be compatible.

In terms of retention, fit is probably alsoimportant but far less theoretical and empir-ical work has been done compared to thework on attraction. One of the reasons forthis lack of information is that theoreticalmodels of turnover (e.g., Mobley, 1977) havetypically focused on characteristics of the jobrather than organizational-level variablessuch as culture. Furthermore, we know thatturnover is a complex process and is affectedby variables (e.g., economic conditions, fam-ily considerations) that have little to do withthe job or the organization (e.g., Carsten &Spector, 1987; Lee & Mitchell, 1994). Itseems plausible, though, that if an employeeperceives that the culture is incompatiblewith his or her values or personality, thiswould certainly prompt a search for a newjob. High turnover may be one mediatingfactor in the relationship between culture

and organizational performance (Kotter &Heskett, 1992). Further research needs tobe done to address these issues.

Culture and Employee Satisfaction/Well-Being

Given the pervasiveness of organizationalculture, to say that it impacts employee sat-isfaction and well-being seems to be statingthe obvious. Surprisingly, there is not a greatdeal of empirical evidence bearing on thisissue—perhaps because of the difficulty ofconducting the multiorganizational studiesnecessary to test such hypotheses. What littleevidence exists, however, suggests that cul-ture makes a difference in the quality ofemployees’ work lives. For example, Hattonet al. (1999) found that a mismatch betweenthe actual culture of the organization andwhat employees felt that culture should bewas associated with a number of negativeoutcomes. For example, perceived mismatchwas associated with lower job satisfaction,higher job strain, general stress, and turn-over intent. These findings suggest that thereis no universally appropriate culture. Rather,the key again appears to be whether the cul-ture meets employees’ expectations. It hasalso been shown that safety climate, whichis certainly an aspect of an organization’sculture, is related to an organization’s safetyrecord (Clarke, 2006), which ultimatelyimpacts health.

To provide more explanation of theimpact of organizational culture on employ-ees’ quality of life, Peterson and Wilson(1998) proposed the model presented inFigure 14.2. Note that the key mediatingfactor in the relation between culture andemployee health is business and manage-ment systems. Culture directly impacts thebusiness and management systems that aredeployed by the organization. These, in turn,


may then impact employee health. As anexample, an organization with a very con-trolling culture may have a human resourcessystem that requires employees to accountvery carefully for their time. There is, in fact,evidence that culture does impact organiza-tional choices of human resources systems(Aycan, Kanungo, & Sinha, 1999). This highlevel of control, in turn, may detract fromquality of life and ultimately detract fromemployees’ health.

Although it clearly needs empiricalassessment, the connection between organi-zational culture and employee well-beinghas certainly been recognized (e.g., Monroy,Jonas, Mathey, & Murphy, 1998; Murphy,1996). More specifically, there has been

increasing emphasis on examining the char-acteristics of ‘‘healthy organizations’’—thosethat are economically successful and possesshealthy employees. Extensive models oforganizational health await development,and a key factor in that development is likelyto be organizational culture. In the future,linking macro-level variables such as organi-zational culture to employee health and well-being will likely become a major focus of theemployee health literature (e.g., Bliese & Jex,1999).

CHAPTER SUMMARY

This chapter examined the important topicof organizational culture. Although culture

FIGURE 14.2Model of the Relation between Culture,Work, and Health

Profit andProductivity

EmployeeHealth

Health CareUse and Illness

Business andManagement

Systems

Quality ofWork Life

CorporateHealth

CorporateCulture

Source: M. Peterson and J. Wilson. (1998). A culture-work-health model: A theoretical conceptualization.

American Journal of Public Health, 22, 378–390. Reprinted by permission of PNG Publications.


has been defined in a variety of ways, theessence of culture lies in the basic assump-tions and values held by the members ofan organization. This definition is widelyaccepted in organizational psychology, andit reflects the impact that cultural anthropol-ogy and sociology have had on the study oforganizational culture.

All organizational cultures are unique tosome degree, but there have been efforts todevelop ‘‘models’’ of organizational culture.O’Reilly et al. (1991) developed a model ofeight dimensions of organizational culturethat are based on employees’ perceptions ofunderlying values. Denison and colleaguescontend that organizational culture can bedescribed according to four broad dimen-sions and that each of these four can be bro-ken down into more specific subdimensions.Both of these models have been very useful inboth research and organizational diagnosis.

The culture of an organization is reflectedin a variety of ways; some are understandableto outsiders and others are more difficult tocomprehend. Symbols and artifacts repre-sent the major physical manifestations ofculture; rites and rituals represent behavioralmanifestations. Language and stories canalso be an important window into culture,both directly and for more symbolic reasons.Ultimately, culture is difficult to compre-hend and an outsider needs a long time todecipher it.

The culture of an organization may beshaped by a number of factors. For mostorganizations, the organizational founder(s)is the most important factor in initially shap-ing the organization. Over time, however,culture will also be impacted and shapedby the extent to which it facilitates organiza-tional adaptation and survival. Cultures tendto develop and ultimately persist over timebecause they have adaptive value for theorganization.

Studying organizational culture can bechallenging, but it is necessary in order tofully understand it. There are certainlyinstances of the use of self-report measuresof culture, although many organizationalculture researchers are wary of this method.As a result, the most typical method of study-ing organizational culture has been ethnog-raphy. Using a qualitative assessment of aculture is consistent with the notion thatmembers of a culture are not good at repor-ting their basic assumptions. This methodology is also consistent with the anthropologi-cal roots of this field. In the future, other meth-ods will probably be available for studyingculture as well.

Changing the culture of an organizationis difficult, given that culture is reflected inbasic assumptions. Nevertheless, organiza-tional cultures do change over time, and, inmost cases, the mechanisms responsible forchange depend on the life stage of the orga-nization. Clearly, though, organizational cul-ture change is not something that occursquickly or easily in organizations. True orga-nizational culture change usually occursonly in response to extreme environmentalconditions.

A final factor to consider in examiningorganizational culture is its impact onimportant outcomes. Not a great deal ofempirical research has been done on theeffects of organizational culture, most likelybecause multiple organizations are neededto do such research. Nevertheless, empiricalresearch has shown that organizational cul-ture may impact a number of importantoutcomes such as performance, attractionand recruitment of employees, employeeretention, and employee satisfaction andwell-being. Although a great deal of re-search is yet to be done in this area, itappears that there is no one type of culturethat is ideal. The most important factor

Chapter Summary 469

appears to be a match between organiza-tional culture and characteristics of employ-ees rather than what is considered to be theidealized culture.


Adkins, B., & Caldwell, D. (2004). Firmor subgroup culture: Where does fitting


JENNIFER CHATMAN AND MATCHING PEOPLE WITH ORGANIZATIONS

As a graduate student watching all my friends

go out and find their first ‘‘real jobs,’’ I became

intrigued with the idea that some people were

much more committed, that is, willing to go the

extra mile, on their jobs than others. Whatcould account for this difference? I didn’t think

it was a personality issue because all my friends

were smart (of course) and hard working. I

decided that it must be something about the

way the companies they worked for treated

them, or the match between the organization’s

culture and their own preferences. To study

this issue, I decided to look into the publicaccounting industry because with very few

firms, I could capture nearly the entire industry

(they were called ‘‘The Big Eight’’ back in the

late ‘80’s when I did my first study; after many

mergers and Arthur Andersen’s demise with

Enron, they are often referred to as ‘‘The Final

Four!’’). I also chose this industry because the

firms are so similar in structure, size, and tech-

nology so that any differences in people’s level

of commitment would have to be due to differ-

ences in the organization’s culture, that is, their

values and behavioral norms.Even within this very homogeneous

industry, there were striking differences in

organizational culture. One firm valued detail

orientation above all else, another valued

initiative, and a third cared most about devel-

oping a single culture across the entire, global

firm. These striking differences played out in

a number of important ways for employees.For example, those who fit well with the

culture stayed longer, were promoted faster,

and demonstrated greater commitment to the

firm. Interestingly, these outcomes of person-

culture fit mattered more for performance

than did the more typical person-job fit we

think of which is how well people’s knowl-

edge, skills, and abilities fit the specific job.What this implies is that, as a job seeker, it

may be more important to find an organiza-

tion whose culture matches your own pref-

erences (e.g., are you a team player or do you

like to focus more on individual achieve-

ment?) than to find the perfect job regardless

of the organization. That is, job selection is

more effectively viewed as organization selec-tion because, if you resonate with the culture

of the organization you join, chances are

you’ll be very successful there and hold many

different jobs over time.

Jennifer A. Chatman

Haas School of Business

University of California, Berkeley


matter most? Journal of Organizational Be-havior, 25, 969--978.

Denison, D. R., & Mishra, A. K. (1995).Toward a theory of organizational cultureand effectiveness. Organization Science, 6,204--223.

Mason, R. O. (2004). Lessons in organiza-tional ethics from the Columbia disaster:

Can culture be leathal? Organizational Dyna-mics, 33, 128--142.

VanVianen, A. E. M. (2000). Person-organ-ization fit: The match between newcomers’and recruiters’ preferences for organiza-tional cultures. Personnel Psychology, 53,113--150.


Chapter Fifteen

OrganizationalChange andDevelopmentI

t is often said that change is one of thefew constants in today’s organiza-tions. In this final chapter, we coverthe various ways in which organiza-tions seek to change through the use

of organizational development. Organiza-tional development is a process by whichorganizations use the theories and technol-ogy of the behavioral sciences to facilitatechanges that enhance their effectiveness.Since organizational development draws onmany of the theories and research findingsthat have been covered in previous chapters,it is fitting that organizational development isthe focus of this final chapter.

The chapter will begin by defining orga-nizational development and then focus onwhy organizations seek to change. Manyorganizations seek to change simply to beable to survive. If they don’t change, they’llgo out of business. Organizations, however,may also change for more proactive reasons.We will then discuss the theoretical modelsthat have been most influential in guidingthe work of organizational-developmentpractitioners. The discussion will be fol-lowed by descriptions of the most popularorganizational-development interventions.

After describing organizational-develop-ment interventions, the focus of the chapterwill shift to a more general discussion ofprocess issues in organizational develop-ment. Perhaps the most important of theseissues is the conditions necessary to sustainmeaningful organizational change, as well asconditions that often stand in the way ofchange. This will be followed by a discussionof the evaluation of organizational-develop-ment interventions—an important topic,

considering the cost of organizationalchange. The chapter will conclude withan exploration of some of the most com-mon ethical issues faced by organizational-development professionals.

WHAT IS ORGANIZATIONALDEVELOPMENT AND WHYIS IT USED?

Organizational development has beendefined in a variety of ways, by a numberof authors (French & Bell, 1995; Porras &

473

Robertson, 1992). Porras and Robertson,however, integrated the numerous defini-tions of organizational development andstated that: ‘‘Organizational development isa set of behavioral science–based theories,values, strategies, and technologies aimedat planned change of the organizational worksetting for the purpose of enhancing individ-ual development and improving organiza-tional performance, through the alterationof organizational members’ on-the-jobbehaviors’’ (p. 722).

Several aspects of this definition are note-worthy. First, the focus of organizational de-velopment is facilitating organizationalchanges that enhance both organizationalperformance and individual development.This distinguishes organizational develop-ment from organizational interventions thatfocus exclusively on enhancing either organi-zational performance or individual develop-ment. Second, organizational development isrooted in the theories and methodology of thebehavioral sciences. Therefore, organizationaldevelopment is different from approaches toorganizational change, which are based solelyon changes in manufacturing technology or,perhaps, information systems. It is true, how-ever, that organizational development canassist in the implementation of changes inmanufacturing or information technology(e.g., Barrett, Grant, & Wailes, 2005;Davidson, 2006). Finally, this definitionmakes it very clear that the key to organiza-tional change is changing the behavior ofemployees.

Now that organizational development hasbeen defined, the next issue to explore is whyorganizational development is used. One ofthe most common motivating factors behindorganizational-development programs canbest be described as survival. Organizationaldecision makers may realize (hopefully, not

too late) that change is necessary if the orga-nization is to remain competitive. Considerthe following example. One of the authorsonce worked for a large telecommunicationscompany that was engaged in a fairly compre-hensive organizational-development pro-gram. The motivating force behind thisprogram was actually quite simple. Prior tothe program, this organization (like others inthat industry) operated in a very stable, regu-lated environment. When this industry wasderegulated, top management recognized thatif the organization did not become more mar-keting oriented and customer focused, theywould not remain competitive.

Another powerful motivator of change ispoor organizational performance. When anorganization fails to show a profit over anextended period of time, or sees its marketshare being steadily eroded, it will oftenfacilitate organizational change. This factorwas undoubtedly at work at the ChryslerCorporation in the late 1970s. During thatperiod, Chrysler had lost market share andwas literally in danger of going out of busi-ness. The fact that Chrysler was performingso poorly made it easier for Lee Iaccoca toinstitute a number of organizational changesthat ultimately turned the company around(Iaccoca, 1984).

The desire for survival is often the moti-vating factor behind many organizational-development programs, but there are cer-tainly other factors as well. In some cases,relatively effective organizations will engagein programs of planned change, and theycould do so for a variety of reasons. For ex-ample, an effective organization may engagein change for strategic reasons (e.g., Buller,1988). A manufacturing organization thatdecides to enter the consumer products busi-ness, perhaps through an acquisition, mayhave to institute a number of organizational

474 Organizational Change and Development

changes in order to make this strategy workeffectively. An organization that competesprimarily on the basis of product qualitymay decide that it wants to put greateremphasis on customer service. Again, orga-nizational changes may very well be neces-sary if this strategy is successful.

Some organizations may simply antici-pate changes in the external environmentand proactively respond to those changes.For example, many fast-food restaurantsanticipated demographic changes in thepopulation, which led to greater utilizationof retirees as employees. As another example,several years ago, some universities putconsiderable resources into long-distanceeducation programs because they antici-pated changes in computer technology andan increase in the number of working adultswho would want to attend college. In bothinstances, organizational development wasthe tool of choice for facilitating thesechanges.

A final reason for organizations’ engagingin programs of planned change can simplybe described as self-improvement; that is, noexternal pressure for change exists, and thereare no concrete strategic reasons for chang-ing. Instead, a well-functioning organizationmay just want to improve itself in some way.An example of this motivation for changeoccurs when a major league baseball teamwins the World Series and then is very active,during the off-season, acquiring new playersand making changes in the coaching staff.This response may be due to the fact that,despite its success, the team recognizes thatit still has some weaknesses that need to beaddressed. Organizations that are alwaysstriving to get better will remain competitivelonger than organizations in which successleads to complacency (Ferrier, Smith, &Grimm, 1999).

A BRIEF HISTORY OFORGANIZATIONALDEVELOPMENT

While there is no recognized ‘‘father of orga-nizational development,’’ the person whowould probably be most deserving of thattitle would be Kurt Lewin, whose contribu-tions to the broader field of organizationalpsychology were described in Chapter 1.With respect to organizational development,Lewin made a number of contributions, buttwo stand out as being most important. First,Lewin was the first psychologist to providethe field of organizational development witha theoretical base. His Three-Step Model ofChange and Action Research continues toserve as an important theoretical guide toorganizational-development practitioners.Second, Lewin played a key role in the estab-lishment of both the Center for the Study ofGroup Dynamics, at Massachusetts Instituteof Technology (MIT), and the NationalTraining Laboratories (NTL), in Bethel,Maine.

The Center for the Study of Group Dy-namics was a fertile site for the study of manygroup processes that are important to orga-nizations. It also served as the trainingground for many individuals who playedimportant roles in shaping the emerging fieldof organizational development (French &Bell, 1995). Although Lewin died beforethe establishment of the National TrainingLaboratories (NTL), he was instrumentalin their development. The major purposeof the NTL was to provide laboratory or T-group training to managers and educators.T-groups are essentially unstructured groupsin which participants learn both throughtheir interactions and through the evolvinggroup dynamics. Although there has beenconsiderable criticism of T-groups over the

A Brief History of Organizational Development 475

years (e.g., Campbell & Dunnette, 1968),the purported goals of T-group training—enhancement of interpersonal skills andawareness of group dynamics—are certainlyworthwhile. In the early 1950s, T-groupswere synonymous with organizational devel-opment, and nearly all of the early practi-tioners of organizational development wereeither laboratory trainers, or they had gonethrough such training.

Along with Lewin and the developmentof T-groups, a major factor in the history oforganizational development was the applica-tion of survey research methods in organiza-tions. This was begun in 1946 when theSurvey Research Center (SRC) was estab-lished at the University of Michigan underthe direction of Rensis Likert. Likert, alongwith individuals such as Floyd Mann,devoted considerable attention to the devel-opment and refinement of survey researchmethodology. This technique became partof the field of organizational developmentin 1947, when Likert was able to interestDetroit Edison in conducting a survey onemployee attitudes, perceptions, reactions,and behaviors.

What made this project at Detroit Edisonunique, however, was that the SRC staff notonly conducted the survey and compiled theresults, but they also helped the companypresent the results back to the employees.The approach taken by the SRC researchers,which is now commonplace, was to use whatwas termed an interlocking chain of conferencesto feed the data back. The top managementteam would first receive the survey results.Each individual in this group would thenpresent the results back to his or her owngroup, with the assistance of a consultant.This process would then be repeated untilthe results of the survey were filtered downto everyone in all levels of the organization.From these beginnings, survey feedback

has become one of the most common andmost effective (Bowers, 1973) organizationaldevelopment interventions.

A third major historical foundation oforganizational development was the devel-opment and use of action research methods.Action research, as mentioned earlier, isattributed to Kurt Lewin and is often usedto describe the more general process bywhich organizations change. Action researchalso has an applied side and, during the mid-1940s and early 1950s, many actionresearch projects were conducted not onlyin business organizations but also in educa-tional institutions and community settings.In addition to the general process of actionresearch, one of the important principlesthat came out of the action research move-ment was the idea that research is a collab-orative effort between the researcher andthe members of an organization. Indeed,organizational-development practitionerstoday stress the importance of participationof members of the client system in the orga-nizational-development process (Halbesle-ben, Osburn, & Mumford, 2006).

The final historical foundation of organi-zational development was the sociotechnicalsystems and socioclinical work that emergedin Great Britian. The center of this activitywas the Tavistock Clinic, which was estab-lished in 1920 to provide psychotherapy andto treat the battle neuroses resulting fromWorld War I. The work at Tavistock thatcontributed the most to the field of organi-zational development was a series of experi-ments, by Trist and Bamforth (1951), in theredesign of work in coal mines. These re-searchers found a strong link between thedesign of work and the social structure andgroup dynamics within the mines. This re-search was important on its own merits, andit served as one of the first applications, in anactual industrial setting, of the ongoing U.S.


work on group dynamics. Trist conductedmany subsequent investigations of the inter-action between the design of work and thesocial environment. The Tavistock workhas served as a foundation for many work-redesign interventions, and it has helped toprovide more general insight into the rela-tionship between the technological andsocial environments.

As will be evident throughout thischapter, many of the interventions and ap-proaches in organizational development canbe traced to the historical foundations dis-cussed previously. Like any field, however,organizational development is not static;thus, a number of more recent trends inthe field will shape the future. One clearrecent trend in organizational developmentis the increased interest in interventionsaimed at improving team functioning. Smallteams represent the building blocks of manyorganizations today, and the success of entireorganizations depends heavily on the successof individual teams. It is no surprise, then,that team building has become by far one ofthemostpopularorganizational-developmentinterventions.

A second recent trend is the increased useof large-scale, broad-based organizational-change interventions (Mirvis, 2005; Steil &Gibbons-Carr, 2005). This trend towardmore large-scale change efforts is probablydue to a number of factors. For example,because of the volatile competitive environ-ments many organizations face, change oftenneeds to be comprehensive and must occurquickly. Thus, organizations often cannotwait for individual- or team-focused inter-ventions to facilitate wider organizationalchange. This trend also reflects, to somedegree, a maturing of the field of organiza-tional development. Over time, the accumu-lated wisdom of organizational-developmentresearchers and practitioners has shown

that, for real change to occur in an organiza-tion, the entire system must be involved(e.g., Beckhard, 1967). What better way toget total system involvement than to makethe whole organization the focus of an inter-vention?

Another current trend in organizationaldevelopment is that the extent of applicationhas grown widely. Organizational develop-ment is used in a wide variety of organiza-tions (see, e.g., Athey & Hautaluoma, 1988)and has become truly international in scope(e.g., Perlaki, 1994; Rao & Vijayalakshmi,2000). Perhaps the only negative fact asso-ciated with the growth of organizational de-velopment is that it makes the establishmentof common ethical and professional stand-ards quite difficult. Although many organi-zational-development practitioners aretrained specifically in organizational devel-opment and in closely related fields, such asindustrial/organizational psychology, orga-nizational behavior, and human resourcesmanagement, there is no specific ‘‘credential’’that one must have as a practitioner of orga-nizational development. As a result, few safe-guards exist to protect organizations againstincompetent or unethical organizational-development practitioners. However, in re-cent years, organizational-developmentprofessionals have taken some steps to estab-lish firm standards of professional practice(see Comment 15.1).

THE THEORY BASEOF ORGANIZATIONALDEVELOPMENT

In most comprehensive treatments of or-ganizational development, the author(s)inevitably point out that organizational de-velopment is a field without a strong the-oretical base (e.g., Beer, 1976; Burke, 1994;Porras & Robertson, 1992; Porras & Silvers,


1991). Often, consultants apply organi-zational-development interventions basedmuch more on empiricism, or even trial-and-error, than on solid theoretical ground-ing. This has led some observers to vieworganizational development more as a tech-nology than as a legitimate subtopic withinorganizational psychology. Such views arecertainly provocative and may have somemerit, but organizational-development prac-titioners do in fact have a theory base thatpractitioners can look to for guidance. In thissection, we review that theory base.

Lewin’s Three-Step Model

The oldest theory of the organizational-change process is referred to as Lewin’sThree-Step Model (Lewin, 1947). Lewin useda physical metaphor to explain the processby which social systems change. The use ofmetaphor in theory development was in-troduced in Chapter 13 on organizationaldesign (e.g., McKenna & Wright, 1992)and is also quite useful in describing organi-zational change. Like organizational design,organizational change is a highly abstract

COMMENT 15.1

THE ORGANIZATIONAL-DEVELOPMENT PROFESSION

ORGANIZATIONAL DEVELOPMENT HAS come a long

way as a profession during the past 50 years.

When the field was first developing, organi-

zational development was seen primarily as atechnology that was applied by those with

training in psychology and other behavioral

sciences. Over time, however, this field has

developed an identity that is distinct from

psychology and other behavioral sciences. For

example, many organizational-development

practitioners now belong to professional asso-

ciations that are designed to advance the inter-ests of the profession and provide some

guidelines for its practice. The two largest

associations are the Organizational Develop-

ment Network (http://www.odnetowrk.org/)

and the Organizational Development and

Change Division of the Academy of Manage-

ment (http://www.aom.pace.edu/odc/).

Another indication of the growth of theorganizational-development profession can be

found in the number of its academic programs.

According to the most recent estimates from

the Organizational Development and Change

Division of the Academy of Management, there

are now eight stand-alone doctoral programs

in organizational development. If one adds to

this the 14 masters’ programs and six certificate

programs, it is now quite possible for a personinterested in organizational development to

obtain advanced and specific training in this

field. In years past, those interested in organi-

zational development were forced to obtain

academic training in related fields (e.g., psy-

chology, sociology, communications) and

learn the specifics of organizational develop-

ment primarily through work experience.The emergence of professional associations

representing organizational-development pro-

fessionals is positive, for two reasons. First,

greater organization of the profession will lead

to higher professional standards, which will

ultimately lead to better practice. Second, given

the growth in academic programs in organi-

zational development, there will undoubtedlybe more empirical research and theorizing.

These will be crucial if organizational develop-

ment is truly going to be both a science and a

technology.


process that cannot be readily simulated ormodeled in a laboratory setting.

The three steps in Lewin’s model arepresented in Figure 15.1. In the first stepin the change process, the stage labeledunfreezing, an organization begins to recog-nize the need for change. This is a crucialstep in the change process, according toLewin (1947), because an organization can-not, and will not change unless there is somerecognition of the need for change. Manyfactors described earlier as motivators oforganizational change (e.g., loss in profits,major environmental changes) would qualifyas unfreezing events. That is, if an organi-zation is unprofitable or faces dramaticchanges in its operating environment, thiscould make its employees recognize the needfor change. The word could is italicizedbecause it is certainly possible for an organi-zation to recognize these unfreezing events,yet not connect them to the need for organi-zational change. Mirvis (2005) describes thecase of a Dutch food company that wanted toplace more emphasis on quality control. Inorder to begin the change process, this com-pany took all of its employees to a warehousethat was completely filled with wasted prod-ucts. These employees then watched as these

defective products were destroyed. By doingthis the company was emphasizing, in dra-matic fashion, the need for all employees tohave higher quality standards.

Assuming that a sufficient level of or-ganizational unfreezing has occurred, thesecond step in Lewin’s model is change ortransformation. This is also a crucial step inthe process because it represents tangiblechanges in the way an organization operates.For example, an organization may convert toa team-based structure, or redesign jobs toenhance customer satisfaction, or engage inany number of other changes. In addition tobeing a crucial step, this is also a very diffi-cult step. Changes introduced at this pointmay require that employees do things verydifferently than in the past. For some, thismay be invigorating; for most people, how-ever, changing the way they have donethings in the past is a very difficult process.

When an organization changes or trans-forms itself in some way, the next step isrefreezing. The changes that are enacted dur-ing the second step in the model become arelatively permanent part of an organiza-tion’s behavioral repertoire. Refreezing mayalso be a difficult step because employeesmay be resistant to the organizational chan-ges. For example, it is quite common foremployees to initially be very enthused aboutenacting organizational changes. However,after their initial enthusiasm wears off, em-ployees may revert to old ways of doingthings. For true refreezing to occur, employ-ees must see that it is in their best interests tomaintain the organizational changes that arecarried out in the transformation phase.

Perhaps the greatest strength of Lewin’sThree-Step model is its simplicity. This mod-el is easy to understand and, in fact, providesan organization with some useful guidancein carrying out organizational changes. Forexample, an organization must consider the

FIGURE 15.1Lewin’s Three-Step Model of the OrganizationalChange Process

Unfreezing

Transformation

Refreezing


need to prepare its employees prior to intro-ducing changes (e.g., unfreezing), and mustanticipate some degree of resistance beforethese changes become a permanent part ofthe culture (e.g., refreezing). Organizationsthat do not pay attention to these factors areunlikely to have successful change efforts.

Ironically, simplicity is also one of theprimary weaknesses of Lewin’s model. Morespecifically, the model has been criticized forhaving an overly simplistic view of thechange process (Purser & Petranker, 2005).Another problem with this model is that itdoes not provide a great deal of insight intohow, for example, the process of unfreezingactually works; for that matter, it provideslittle insight into how organizations canactually facilitate the processes of change orrefreezing. Thus, although Lewin’s model isgood as a description of organizationalchange, it falls somewhat short as a compre-hensive model of the organizational-changeprocess because it lacks explanatory power.

The Action Research Model

A second general theoretical model of theorganizational change process, also attrib-uted to Kurt Lewin, is the Action ResearchModel. The general idea behind action re-search is that organizational change is lik-ened to a cyclical research process. Actionresearch also emphasizes that throughout allphases of the research process, there is activecollaboration between the researcher and themembers of the client system. The majorsteps in the action research process, asdelineated by Lewin (1951), are presentedin Figure 15.2.

As can be seen, the first step in the actionresearch process is problem identification.For any research to be undertaken, or anychange to occur, there has to be some recog-nition of a problem that people care about. In

the most general sense, a problem exists anytime there is a difference between the currentstate of affairs and the desired state of affairs.For example, an organization’s profits maybe lower than they should be; the level ofemployee turnover may be higher than theorganization feels is desirable; or the numberof workers’ compensation claims filed byemployees may be higher than the organiza-tion feels it should be.

The second step in the Action ResearchModel is development of hypotheses. Obvi-ously, for any particular problem, therecould be a multitude of causes. Fortunately,based on prior theory and, perhaps, theexperience of organizational members, itis often possible to focus heavily on somecauses and pay less attention to others. Forexample, if the problem area identified isemployee turnover, many years of researchon turnover has shown that factors such asemployees’ job attitudes and the economicenvironment are important contributingfactors (Hulin, 1991). Based on this know-ledge, it may be useful to examine em-ployee attitudes and, possibly, explore the

FIGURE 15.2Lewin’s Action Research Model

Problem Identification

Development of Hypotheses

Hypothesis Testing

Data Interpretation


external job market to develop somehypotheses about turnover.

After hypotheses have been specified, thenext step in the action research process isdata collection. This is an important stepbecause it distinguishes action researchfrom less scientific forms of inquiry. Forexample, the previously described hypothe-ses could be tested by asking experts orperhaps even through introspective meth-ods. When action research is used, however,the scientific method is used, and it requiresthat hypotheses be tested by gathering em-pirical data.

After empirical data are collected, thenext step in the action research process isdata interpretation. The critical questionthe action researcher is trying to answerat this point is: Do the empirical data sup-port, or fail to support, the proposedhypotheses? The biggest dilemma for theresearcher at this point is essentially‘‘deciding how to decide’’ whether the datasupport the hypotheses. Fortunately, infer-ential statistical methods are available (e.g.,Hays, 1988) to assist the action researcherin making such decisions.

After the data are interpreted and a deci-sion is made about whether the hypothesesare supported, is the action research se-quence complete? The researcher may pos-sibly lose interest in a particular problem;but, more typically, the action research cyclewill repeat itself. For example, if the re-searcher collects data and finds that jobsatisfaction is inversely related to turnover(as suspected), this raises other importantquestions. What measures could the orga-nization take toward raising employees’levels of job satisfaction and, by inference,reducing turnover? Are certain facets of jobsatisfaction more important than others indetermining turnover? These are empiricalquestions that could certainly be tested

through further research projects. Put dif-ferently, the action research cycle can berepeated once again in order to answerthese questions.

Thus, according to the Action ResearchModel, the process of organizational changecan be characterized as a continuing cycli-cal process of hypothesis generation, datacollection, data evaluation, and, ultimately,intervention. Another important aspectof the Action Research Model was alludedto earlier: The research process is a collab-orative effort between the researchers andmembers of the client system. In a moretraditional research project, there is a dis-tinct power or status differential betweenresearchers and the ‘‘subjects’’ of studies.Although researcher–participant collabora-tion can pose some difficult dilemmas attimes (e.g., Mirvis & Seashore, 1979), itspositive benefit is that the client systemassumes ownership of the organizational-change process—an important ingredientin maintaining organizational change overtime.

To a certain extent, the advantages anddisadvantages associated with the Action Re-search Model are similar to those associatedwith Lewin’s Three-Step Model. That is, theaction research provides a very useful guideto understanding organizational change.Furthermore, there is empirical evidencethat action research can be a useful approachto solving problems in organizations (e.g.,Halbesleben et al., 2006). However, likethe Three-Step Model, it also does notdirectly explain the most important factorsthat are involved in the change process. Forexample: Does change begin with the lead-ership of an organization or with lower-levelemployees? What are the factors that lead toresistance to change? How does an organi-zation sustain organizational change overthe long term? In fairness to Lewin, the


Action Research Model was never reallyintended to be a theory of organizationalchange. Rather, it is probably most appro-priate to think of the Action Research Modelmore as a general ‘‘theory of intervention’’than as a theory of organizational change.

General Systems Theory

A third general theoretical base upon whichmuch of organizational development rests isGeneral Systems Theory. General SystemsTheory, which was briefly covered in Chap-ter 13, was developed by von Bartalanffy in1950 and made its way into organizationalpsychology in 1966, through the work ofKatz and Kahn. The basic idea of generalsystems, as applied to organizations, isthat organizations import material from theenvironment, transform that input, and ulti-mately return it to the environment in somealtered form. As a result, organizations areconstantly in a dynamic interaction withtheir external environments. More often thannot, organizations change in reaction to, orin anticipation of, changes in the externalenvironment. If organizations were ‘‘closedsystems’’ and therefore could ignore theexternal environment, there would be littleneed to engage in organizational change anddevelopment activities.

Another aspect of General SystemsTheory that has greatly influenced the fieldof organizational development is the ideathat any system is comprised of a series ofsmaller ‘‘subsystems.’’ As an analogy, thehuman body is composed of a number of‘‘subsystems’’ that guide functions such ascirculation, digestion, and so on. Katz andKahn (1966) pointed out that organizationsalso consist of a number of interrelated sub-systems that guide such functions as import-ing materials from the environment (e.g.,purchasing), transforming those materials

(e.g., production), and making sure thatthe transformed materials are returned tothe external environment (e.g., marketing).

The idea of interrelated subsystems isrelevant to organizational developmentbecause whenever a change is introducedin one part of an organization, those guidingthe change must be on the lookout for the‘‘system-wide’’ ramifications of that change.If an organization decided to change to ateam-based structure (a change that is, infact, becoming increasingly popular), therewould be many system-wide implications ofthis change. Within a team-based structure,supervisors become more like ‘‘advisers’’ or‘‘consultants’’ rather than bosses. It is alsoquite likely that working in teams mayrequire fundamentally different skills andabilities than are found in more traditionalstructures. Finally, for a team-based struc-ture to be effective, a vastly different com-pensation system is needed. Thus, onechange may require changes in the way thatleaders and other employees are selected andtrained, as well as changes in the organiza-tion’s compensation system.

Theories of Organizational Change

The theories covered to this point are usefuland provide some insight into the generalprocesses involved in organizational changeand development, but they do not describethe specific organizational factors that areinvolved in the change process. One of themajor reasons there are so few specific theo-ries of organizational change is that organi-zations are so diverse. As a result, it is quitedifficult to create a model that serves as ageneralized guide to change in all organiza-tions. Fortunately, some theories of this typehave been developed, and two are reviewedin this section.


Burke (1994) developed a theory of theorganizational-transformation process that isgeneral enough to apply to a great variety oforganizational types. As can be seen inFigure 15.3, the model proposes that organi-zational transformation is the result of inter-related factors. Starting at the top of themodel, the external environment is often akey factor in initiating organizational trans-formation because change is often motivatedby survival or by the desire to capitalize on anew opportunity. The model proposes thatthe external environment has a direct impacton leadership, the mission and strategy of anorganization, and organizational culture.Note, however, that in each of the threecases, these effects are reciprocal. For exam-ple, the external environment impacts lead-ership, but leadership could also shape the

external environment. The same could besaid for mission and strategy and for organi-zational culture. Note that all three of theseorganizational factors are interrelated.

The model further proposes that missionand strategy, leadership, and organizationalculture have a direct impact on individualand organizational performance. Note, how-ever, that, as in the first portion of the model,all relations are reciprocal. That is, missionand strategy, leadership, and organizationalculture impact performance but are alsoimpacted by the performance of individualsand the organization as a whole. The othernotable feature of the model is the dynamicfeedback loop between performance andthe external environment. It is possible forthe external environment to have a directimpact on individual and organizational

FIGURE 15.3Burke’s Theory of Organizational Change

Mission andStrategy

ExternalEnvironment

Individual andOrganizationalPerformance

LeadershipOrganization

Culture


performance. Also, the performance of indi-viduals and the organization as a wholeimpacts the external environment.

Burke’s (1994) model tells us severalimportant things about the process of organi-zational change. First, it clearly shows that theexternal environment plays a key role in theorganizational-change process. This is cer-tainly consistent with General SystemsTheory, and it reinforces the point that orga-nizations cannot afford to ignore the externalenvironment. The model also suggests thatthere are three key ‘‘levers of change’’ thatorganizations can use if they want to sustainmeaningful change over time. These levers arethe interrelated factors of mission and strat-egy, leadership, and organizational culture.Let’s consider each of these factors.

Mission and strategy dictate what an or-ganization’s purpose is, and how it plans toachieve this purpose. It follows, then, that achange in mission and strategy will serve as apowerful catalyst for broader organizationalchange. Furthermore, at the individual em-ployee level, mission and strategy are impor-tant because they help to provide individualswith a sense of purpose and coherence.Therefore, if mission and strategy arechanged, this will serve as a powerful mes-sage to employees that their behaviors mustalso change in some way.

The leadership within an organization isimportant for a number of reasons. Organi-zational leaders play a key role in developingorganizational mission and strategy, as wellas shaping the culture. They also make keycontributions to the development of internalpolicies and procedures and the representa-tion of the organization to the external envi-ronment. Thus, it should come as nosurprise that organizations are only as effec-tive as their leaders and that the involvementof top management is a key factor in sustain-ing successful organizational change.

Organizational culture, as discussed inthe previous chapter, represents the under-lying values and basic assumptions that arepresent in an organization. Culture can havepowerful effects, both positive and negative,on the performance of individuals and entireorganizations. It follows, then, that culturewould play a key role in organizationalchange. This might occur very directly; thatis, organizational change might be viewed asculture change.

Culture may also play a more indirect rolein organizational change. It is much morelikely that changes will be sustained or will‘‘take hold’’ if they are at least somewhatcompatible with an organization’s prevailingculture. For example, an organization thatdecides to redesign jobs to enhance employeeautonomy yet retains a highly authoritarianculture will probably not be successful insustaining this form of organizational change.In this case, the organization may have theoption of implementing less dramatic organi-zational change, or attempting to change to amore participative culture prior to imple-menting the job design intervention.

Porras and Robertson (1992) have pro-posed a theoretical framework from which toview the organizational change and develop-ment process. This model, presented inFigure15.4, is a bit more detailed than Burke’s(1994), although there are some similarities.For example, Porras and Robertson (1992)propose that the external environment is animportant factor in driving organizationalchange. They also propose that the overallpurpose of the organization (represented byvision) drives many of the tangible interven-tions that are designed to facilitate organiza-tional change. Based on the vision that is beingcarried out, several interrelated variables pro-posed in the model may be used as levers ofchange for an organization. These includeorganizational arrangements, social factors,


FIGURE 15.4Porras and Robertson’s Model of the Organizational Change Process

Environment

IndividualCognitions

Vision

OrganizationalPerformance

Organization

Organ

ization

Work Settin

g

Org

aniz

atio

n

Wor

k Se

ttin

gM

embers

IndividualDevelopment

On-the-JobBehavior

PhysicalSetting

SocialFactors

OrganizingArrangements Technology

Mem

bers

Source: J. I. Porras and P. J. Robertson. (1992). Organizational development: Theory, practice, and research. In M. D.

Dunnette and L. M. Hough (Eds.), Handbook of industrial and organizational psychology (2nd ed., Vol. 3, pp. 719–822). Palo

Alto, CA: Consulting Psychologists Press, Inc. Reprinted by permission.

the physical setting, and the organization’stechnology.

Changes in these aspects of the worksetting should ultimately lead to improvedorganizational performance and enhancedindividual development. Notice, in the model,that these changes are fully mediated by indi-vidual cognitions and the behavior of employ-ees. This is clearly the most important andfundamental proposition in Porras and Rob-ertson’s model: Behavior change is the keymediating variable in organizational change.At first glance, this would appear to be a ratherobvious statement, but it is also a very impor-tant one. Many times, when we speak of theorganizational-change process, we tend to forgetthatorganizationsarecollectionsofpeople.Anorganization becomes more innovative only ifits individuals come up with new productdesigns and creative solutions to problems;an organization becomes more customer-service oriented if individual employees placehigh priority on serving customers.

The major practical implication of Porrasand Robertson’s (1992) model is that organi-zations must pay attention to behavior changewhen instituting organizational change.Often, failure to consider individual behaviorchange amounts to organizational decisionmakers’ failure to see the perspective of em-ployees, which is ‘‘What’s in it for me?’’ Thereare cases when employees may stand to gainvery little by changing their behaviors, even iforganizational decision makers believe thatthe associated changes are needed.

The organizational-change theories pre-sented in this section are still quite general.However, compared to the more generalmodels of the change process presented ear-lier, they tell us a great deal more aboutspecific things that an organization can doto facilitate and sustain change. With that inmind, we now shift the focus to specificorganizational-development interventions.

ORGANIZATIONAL-CHANGEINTERVENTIONS

Generally speaking, organizational-changeinterventions differ in terms of (1) the levelin the organization at which they are aimedand (2) the process(es) they are designed toimpact. Organizational-development inter-ventions are typically aimed at individuals,groups, or the organization as a whole. Ofthese, the most popular level of interven-tion is groups, largely because most orga-nizations have realized how importantgroups are to the success of organizations.Despite the popularity of group-level inter-ventions, another trend is occurring inorganizational development: the use ofbroad, system-wide interventions (French& Bell, 1999; Mirvis, 2005). Individual-level interventions are also used fre-quently—although, in many organizations,these are presented simply as trainingprograms rather than as organizational-development interventions.

In terms of processes that interventionsare designed to impact, essentially anythingthat can impact the performance of indi-viduals, or the organization as a whole,could be the focus of an organizational-development intervention. Common exam-ples of processes on which organizational-development interventions are focused arethe roles that employees are asked to play;the goals that drive individual employeesand organizations as a whole; group andintergroup processes; organizational struc-ture; and organizational strategy.

Individual-Level Interventions

In reviewing the history of organizational de-velopment, one of the first organizational-development interventions was sensitivitytraining, also known as T-group training. Al-though T-group training is carried out in


groups, it is an intervention aimed at theindividual. This is because the primary goalsof T-group training are enhancement of inter-personal skills and competence, enhancedawareness of the impact of one’s behavior onothers, and a greater general understanding ofgroup dynamics (Forsyth, 1999).

Although T-group training was at onetime the most popular intervention in or-ganizational development, it is rarely usedby organizational-development practitionerstoday. Probably the major reason is that theeffectiveness of T-groups is doubtful (Camp-bell & Dunnette, 1968). It is also difficult totransfer back to the workplace what one haslearned in the T-group. The total honestyand authenticity that are the hallmarks ofthe T-group movement may not play wellin most real-world work settings. Also, someethical questions surround the use of T-groups—in particular, requiring that em-ployees participate.

Despite the fact that T-groups are rarelyused now as an organizational-developmentintervention, it’s the impact of this interven-tion on the field of organizational develop-ment that cannot be overestimated. Manyorganizational-development interventionsthat are popular today (e.g., process consul-tation, team building) are rooted in the T-group movement. Although team-buildingsessions typically do not focus on interper-sonal issues, the ground rules that governteam-building meetings quite often parallelthose of T-group sessions—for example, par-ticipant candor is encouraged, and efforts aremade to maintain a sense of ‘‘psychologicalsafety.’’

Another common individual-level orga-nizational-development intervention is jobredesign (Hackman & Oldham, 1980;Parker & Wall, 1998). Job redesign was dis-cussed in some detail in Chapter 9, so it willnot be covered in depth here. Job redesign

can be a powerful individual-level organiza-tional-development intervention becauseemployees typically spend more timeworking on their job tasks than on any otheractivity in the workplace. Thus, job redesigncan be a very powerful and efficient wayto change the behavior of employees. Theprimary limitation of job redesign is that itdoes not typically address more ‘‘macro’’issues in the work environment; that is, evenif jobs in an organization are successfullyredesigned this will not compensate for aculture of mistrust and hostility. Anotherdrawback of job redesign is that it is costly.The expense necessary to diagnose andchange jobs may be prohibitive for someorganizations, and the system-wide effectsof job redesign may be far reaching.

Another commonly used organizational-development intervention that is focused onindividual employees is Management byObjectives (MBO) (Carroll & Tosi, 1973).Although the specifics of MBO programsvary widely among different organizations,certain features are common to most. Specif-ically, most involve some level of joint goalsetting between employees and their super-visors. In addition, in most MBO programs,the performance of individual employees isassessed in relation to their progress inaccomplishing these objectives. In general,empirical research has shown that MBO pro-grams have a positive effect on employeeperformance (Rodgers & Hunter, 1991),particularly when they are supported bytop management. There is, however, evi-dence that the effects of MBO could be neg-ative if objectives are not clearly stated(Dahlsten, Styhre, & Willander, 2005).

Group-Level Interventions

As stated earlier, the most common levelat which organizational development-


interventions are focused is the group orteam. This is because more and more orga-nizations are adopting team-based structures(Gordon, 1992). It may also reflect the keyrole that T-groups played in the history oforganizational development. By far themost common group-level organizational-development intervention is team building(Covin & Kilman, 1991). According to Lie-bowitz and DeMeuse (1982), team building isdefined as ‘‘a long-term, data-based interven-tion in which intact work teams experien-tially learn, by examining their structures,purposes, norms, values, and interpersonaldynamics, to increase their skills for effectiveteamwork. It is a direct attempt to assist thegroup in becoming more adept at identifying,diagnosing, and solving its own problems, usu-ally with the aid of a behavioral scienceconsultant’’ (p. 2, italics added). This is obvi-ously a very general definition, and researchhas shown that there is a great deal of varia-tion in the way team building is carried outin organizations (Offermann & Spiros,2001).

The italicized portion of this definition—identifying, diagnosing, and solving its ownproblems—really captures the essence of theteam-building process. A work group thatsuccessfully engages in a team-building inter-vention is not necessarily going to be problemfree. Rather, this group will be capable ofrecognizing when things are not going well,diagnosing the root causes of the problem(s),and taking steps to solve the problem(s).

Although team-building interventions dovary considerably from organization to or-ganization, Liebowitz and DeMeuse (1982)describe a series of eight steps that are com-mon to most. These steps are presentedin Figure 15.5. The first step in the team-building process is labeled scouting. Scout-ing is a preliminary step that involves anexchange of information between a consul-

tant and a potential client organization. Theconsultant would typically describe his orher expertise, values, and style of operation.The organization, in turn, would normallydescribe the nature of its problem(s) as wellas its views regarding the potential causes ofthese problem(s). If it is determined thatteam building is an appropriate intervention,the process would move ahead to the nextstep. If team building is inappropriate, theconsulting relationship may be terminated,unless the consultant can offer other neededservices.

FIGURE 15.5Major Steps in the Team-Building Process

Scouting

Entry

Data Collection

Data Feedback

Diagnosis

Action Planning

Action Implementation

Evaluation


Assuming that team building is appropri-ate, the process then moves on to the nextstep: entry. This is the point at which theclient–consultant relationship begins to beformalized. Although this is typically donethrough written contracts, there is consider-able variation in the way consulting contractsare drawn up. Some consultants prefer con-tracts that are very specific about the natureof the work the consultant will perform;others prefer contracting to be more openended (e.g., Schein, 1987). Regardless ofhow the actual contract is worded, what isimportant at this point is that the consultantand the client organization forge a commonunderstanding on the major dimensions ofthe consulting relationship (e.g., activities ofthe consultant, fees, time frame of the work,method of billing). It is also important at thispoint that the consultant establishes credi-bility with the client organization, and thatthere is enough managerial support to sus-tain the project.

When the preliminary steps of scoutingand entry have been completed, the consul-tant begins the process of data collection. Thissimply involves collecting information, from avariety of sources, on the current functioningof the work groups in an organization. Thismay also involve personal interviews of groupmembers, although consultants often obtaindata from a variety of other sources, such assurveys, observations, and archival records(e.g., minutes of meetings, performancerecords). Surveys are probably the most com-mon form of data collection since there aremany measures of team functioning avail-able (see Wageman, Hackman, & Lehman,2005). Regardless of the data-collectionmethod(s) used, the major objective hereis to get as complete a picture of the func-tioning of groups as possible. Thus, themore methods and sources of informationthat can be used, the better.

After data on group functioning arecollected, the next steps are to summarizethese data and then engage in some form ofdata feedback. This typically involves hold-ing meetings with each of the work groupsin the client organization and presentingthe data that have been collected. Althoughat first glance this may appear to be just‘‘telling people what they already know,’’this may not always be the case. For exam-ple, some group members may have amuch more positive picture of theirgroup’s functioning than is indicated bythe data. Ideally, such discrepancies serveas ‘‘unfreezing experiences’’ for the groupmembers and prompt a desire for a closerexamination of the team’s functioning.Group members may deny these discrep-ancies by questioning the accuracy of thedata, or they may even turn hostile towardthe process or the consultant.

During data feedback, certain themes orareas of concern often surface. For example,group members may be dissatisfied with thelevel of communication or with the waydecisions are made within their group.Unfortunately, data are simply numbers;typically, they do not tell a group why aparticular issue continues to surface. Thus,at this point, the group needs to engage indiagnosis, which represents the group’sattempt to explain why the data came outthe way they did. This is a very crucial step inthe team-building process because, for manygroups, it may be the first time they have evertaken a serious look at their work methodsand internal processes (Hackman & Morris,1975).

Ultimately during the diagnosis stagegroups identify things that are preventingthem from performing up to their full capa-bilities. One could also think of these asbarriers to performance. Developing a list ofperformance barriers is an important first


step in the team-building process becauseonce these are identified the group can beginto address them. This is the purpose of thenext step: action planning. During thisphase, one or more action plans are devel-oped for each of the important problems orperformance barriers identified. For exam-ple, if one of the problems identified in thediagnosis stage is ‘‘Poor communicationamong group members,’’ a correspondingaction plan might be ‘‘Increase the numberof group meetings from once a month toonce every 2 weeks.’’

Action plans are very important becausethey represent the most tangible products ofa team-building meeting—particularly, whata group has committed to do (often in writ-ing) to make it function more effectively.One thing is important to consider, however:Not all action plans are of equal value. Actionplans are much more useful if they are spe-cific and measurable, have an identifiabletime frame associated with them, and holdan individual or individuals responsible forimplementing them. Typically, when one ormore of these attributes is missing, the actionplan will be forgotten soon after the team-building meeting adjourns.

After the team-building meeting is over,the group goes back to its normal routineand thus enters the next phase: action imple-mentation. This involves implementing theaction plans that were agreed upon duringthe team-building meeting. This is the mostchallenging part of the team-building pro-cess because even if the group’s action plansare very good (e.g., specific, measurable),they often require group members to giveup old habits and engage in new behaviors.For example, it may be easy for a leader totalk about letting group members have moreinput into decisions, but it’s quite anotherthing to actually relinquish a portion of his orher control over decision making.

The final stage in the team-building pro-cess is evaluation, which normally takesplace after some period of time has passedsince the team-building meeting. Evaluationinvolves the assessment of the group’s pro-gress on the implementation of the actionplans. This step is important because itmakes the group accountable for implement-ing its action plans. If there is no follow-up,groups will either fail to implement actionplans or will begin to implement them butlose interest. Often, the evaluation portion ofthe process comes in the form of a follow-upmeeting in which the groups, with the assis-tance of a consultant, review progress oneach of the action plans generated duringthe initial team-building meeting. Ideally,this meeting will reveal that the group isimplementing the action plans as written.In some cases, though, it may reveal thatthe action plans need to be revised, or thatthere are external factors preventing thegroup from implementing them as written.

As with many organizational-developmentinterventions described in this chapter, not agreat deal of empirical research exists on theeffectiveness of team-building interventions.The little empirical research that has beendone provides mixed support for the effec-tiveness of team building. Some empiricaldata have failed to support the effectivenessof team building (e.g., Eden, 1985), butthere is also evidence that team buildingcan be a very effective method of improvinga number of group processes (e.g., Cohen &Bailey, 1997). A key factor in determiningthe success of team building is whether thereis a need for it in the first place. If a team isfunctioning very well, or works under con-ditions of very low interdependence, teambuilding may have little effect on group pro-cesses or outcomes.

A second organizational-developmentintervention that may also be aimed at the


group level is process consultation (Schein,1969, 1987). Process consultation is actuallyboth a philosophy of consulting and an orga-nizational-development intervention. As aconsulting philosophy, process consultationemphasizes that the role of a consultant is toguide an organization through the process ofdiscovering what its problems are and helpin finding solutions to those problems. Indoing so, the emphasis is on how things aredone (e.g., process issues) rather than what isactually done. Readers eager to learn moreabout process consultation as a consultingphilosophy are strongly urged to read EdgarSchein’s various works on the topic (Schein,1987).

Process consultation also represents a setof interventions that may be used by a con-sultant. Although these interventions are notexclusive to group settings, this is the contextin which they are often used. Schein (1987)classified interventions on a continuum,based on how confrontational they are vis-a-vis the client. The least confrontationaltype of intervention that a process consultantcan use is an exploratory inquiry. For exam-ple, a process consultant may ask a group:‘‘How have things been going the past fewmonths?’’ or ‘‘Tell me a little bit about thingsthat are important to the group’s success.’’

The second general type of interventiondescribed by Schein (1987) is a diagnosticintervention. In this case the process consul-tant does not tell the client what to do,although the questions he or she may askare more focused than in the exploratoryinquiry stage. If a group decides that itsprimary problem is poor communicationamong group members, a diagnostic inquirymight be: ‘‘What do you think contributesto this poor communication within yourgroup?’’ or ‘‘What can I do to help you im-prove communication?’’ It is doubtful thatmembers of such a group would perceive

either of these questions as being confronta-tional. On the other hand, they do focus on aspecific aspect of group functioning, both interms of isolating the problem and in deter-mining whether the consultant will be able tohelp them solve it.

The third level proposed by Schein(1987) is action alternative interventions.While this type of intervention does notinvolve telling the client what to do, it ispotentially more confrontational than thefirst two types of interventions becausethe process consultant is asking the group:‘‘What have you done about the problem?’’So, for example, if a group feels that internalcommunication is a problem and that thenature of their work roles contributes to thisproblem, the process consultant might ask,as an action alternative intervention, ‘‘Whathave you done to improve the situation?’’ or‘‘If roles are the problem, have you doneanything to change the roles of group mem-bers in a way that would facilitate commu-nication?’’

Why does Schein (1987) consider thesequestions to be potentially confrontational?The key to answering this question really liesin the difference between the roles of clientand consultant. According to Schein, con-sulting is really all about helping, and there-fore consultants are really professionalhelpers. If one accepts this notion, a logicalcorollary is that the client organization is inthe position of seeking help, primarilybecause it cannot solve a particular problemon its own. The idea that an organizationneeds help potentially puts it in a dependentor lower status position vis-a-vis the consul-tant. Therefore, when the process consultantasks—‘‘What have you done about this prob-lem?’’—the client organization risks embar-rassment by saying that it has done thewrong thing or, worse yet, it has done noth-ing to solve the problem.


The final type of intervention describedby Schein (1987) is labeled confrontive inter-ventions. In interventions of this type, theprocess consultant makes recommendationsas to how to solve the problem. For example,a process consultant may say: ‘‘Why don’t youtry_____as a way to solve the problem?’’ or ‘‘Iwould recommend trying_____.’’ These state-ments are clearly more directive than theprevious three. They are also potentially moreconfrontational, because the client mayassume that the way he or she had been doingthings is incorrect. Intervening in this wayalso tends to reinforce the status differencesbetween the client and the process consultantthat were previously described.

According to Schein (1987), a key issue inprocess consultation is how far and howquickly one moves along this interventionfrom exploratory inquiries to confrontive in-terventions. Many consultants begin withconfrontive interventions or move to thatpoint very quickly, often because the clientexpects or even demands it. Many organiza-tions see consultants as bright experts towhom they can essentially ‘‘hand off’’ theirproblems. Unfortunately, handing off prob-lems is often counterproductive because evenextremely bright consultants rarely haveenough information to immediately makeconcrete recommendations to an organiza-tion. Most problems in organizations are toocomplex and are embedded in a cultural con-text that an outside consultant cannotcompletely understand.

Another reason problems are handed offis simply because it is reinforcing to a con-sultant. Both authors have experienced thepositive feelings associated with being seenas an expert by employees of an organizationand having one’s advice taken seriously.When members of an organization ask foradvice, consultants are often compelled tooffer something more concrete than ‘‘What

do you think the problem is?’’ The irony, ofcourse, is that, in the long run, a consultant isoften much more helpful to the client orga-nization by keeping his or her interventionsat the exploratory or diagnostic levels. Bydoing so, the consultant ultimately obtainsmuch more information about the organiza-tion and its problems than he or she wouldby immediately being confrontational.Therefore, if the consultant needs to makerecommendations in the future, these will bebased on much more solid information.

Staying at the exploratory or diagnosticlevels is also beneficial because these inter-ventions force members of a client organiza-tion to think. As a result, these types ofinterventions help to sharpen employees’ ana-lytical and diagnostic skills so that they ulti-mately develop the capacity to solve some ofthe problems on their own. It also increasesthe chances that, although the members of theclient organization are being helped by theprocess consultant, they will ultimately comeup with the solution to whatever problem(s)are identified. This is important becausemembers of the client organization are typi-cally in a much better position than the con-sultant to judge what will and what will notwork in that organization.

In addition to this general typology ofinterventions, Schein (1987) proposed anumber of more specific interventions thatfit into this typology. These are presented inTable 15.1. To reiterate a point made earlier,these interventions were not specifically de-signed to be applied to groups, but they areoften used to improve group functioning.Notice that many of these interventionsare so simple, and are used so frequently,that we don’t really think of them as beingorganizational-development interventions.Terms such as active interested listening, forc-ing historical reconstruction, forcing concretiza-tion, forcing process emphasis, and diagnostic


questions and probes are used all the time as away to help people figure out what theirproblems are.

Items 6 through 10 in Table 15.1 aremore specific to group settings and requirea bit more explanation. With process man-agement and agenda setting, the process con-sultant might provide suggestions as to howgroup meetings should be conducted or howthe agenda might be structured. Whilepotentially confrontational, these interven-tions may also be very helpful. For example,one of the problems that the authors haveobserved in group meetings over the years ispoor management of the group’s time—thatis, the group may consistently get throughonly one or two agenda items during itsmeetings. One way to deal with this problemis to be very structured in setting the meetingagenda and allocating a certain number ofminutes to each agenda item.

Providing feedback is also a potentiallyconfrontational intervention, yet it may alsobe quite useful to a group. A process con-sultant sitting in a group meeting canobserve many things about the group’s func-tioning. Probably the most visible thing is the

manner in which the group communicates.Is communication very formal, or do groupmembers feel free to jump in at any time? Doall members of the group communicate, ordo one or two group members dominate thediscussions? If a group’s discussions aredominated by a small number of vocal em-ployees, having a consultant point this outcould be highly confrontational. This is par-ticularly true if this behavior pattern is indi-cative of other problems in the group.Despite being potentially confrontational,this type of feedback may be highly usefulbecause a group may not even be aware thatthis is happening.

When a consultant makes content sug-gestions and recommendations, he or she isbeing even more directive and potentially‘‘confrontive.’’ A process consultant, for ex-ample, may recommend that a work groupallot time at the end of a meeting for groupmembers to discuss concerns they have withthe way the group functions. Recommenda-tions could also be made about the frequencyof team meetings—a process consultant maysuggest that a group that meets only once amonth should start meeting twice a month.

TABLE 15.1A Listing and Categorization of Process Consulting Interventions

Intervention Category

1. Active, interested listening Exploratory

2. Forcing historical reconstruction Diagnostic

3. Forcing concretization Diagnostic

4. Forcing process emphasis Diagnostic

5. Diagnostic questions and probes Diagnostic, action oriented

6. Process management and agenda setting Confrontive

7. Feedback Confrontive

8. Content suggestions and recommendations Confrontive

9. Structure management Confrontive

10. Conceptual inputs Potentially confrontive

Adapted from: E. H. Schein. (1987). Process consultation: Lessons for managers and consultants

(Vol. II). Reading, MA: Addison-Wesley.


This type of intervention is potentially con-frontational because, although the processconsultant is not explicitly saying that thingshave been done wrong, the recommenda-tions given to the work group may conveythat opinion.

With structure management, the processconsultant may make recommendationsregarding the design of a group’s task, orperhaps the manner in which a group carriesout its tasks. For example, in a group whereeach group member performs a highly spe-cialized task, a consultant may recommendcombining tasks in order to increase themeaningfulness of the work. This is againpotentially confrontational because the pro-cess consultant is directing the client toaccept a solution to a problem that he orshe, rather than the client, has proposed.

The final intervention listed in Table 15.1is labeled conceptual inputs. It is commonfor process consultants to present relevantconceptual material during group-related in-terventions. For example, a consultant whois helping a group improve its effectivenessmay provide the group with a 20 minute‘‘lecturette’’ in which models of group effec-tiveness are described. Although conceptualinputs can certainly be useful, they can alsobe counterproductive, particularly if they areoverused. As Schein (1987) points out, con-ceptual inputs may be confrontational if theyhighlight weaknesses in a group’s processes.For example, a group that does not encour-age a great deal of debate and open commu-nication when decisions are made maybecome quite uneasy if a process consultantwere to provide a brief lecture on Group-think. Recall from Chapter 11 that a lack ofopen debate and communication are two ofthe primary symptoms of Groupthink.

Given the generality of process consulta-tion, there is little empirical evidence as to

whether this approach is effective. It wouldseem, however, that there would be a num-ber of advantages in using this method, asopposed to more traditional forms of con-sultation. As Schein (1987) points out, orga-nizations often know that something is ‘‘notright,’’ but they are not sure where the prob-lems lie. Thus, more often than not organi-zations and work groups need help andguidance in discovering what those prob-lems are and how best to solve them. Processconsultation is ideal in this type of situationbecause the primary assumption behind thismodel is that the client owns the problemand is ultimately the one who knows theorganization well enough to figure out theproblems—and the solutions to the prob-lems.

Organization-Wide Interventions

Interventions that target the organization as awhole are becoming increasingly popular,for a number of reasons. For example, orga-nizational change is most likely to occur andbe sustained when the entire organization isinvolved in the change effort (Burke, 1994).Also, given the rapid rate of change in thebusiness environment today, many orga-nizations do not have the luxury of interven-ing with groups or individuals and thenwaiting for these interventions to facilitateorganization-wide change.

The most widely used organization-wideintervention is survey feedback. In a typicalsurvey feedback program, a survey is admin-istered to all employees. It may addressissues such as attitudes, perceptions oforganizational climate, perceptions of man-agement, and perceived level of organiza-tional effectiveness, among other things.After the survey is administered and


returned, the results are tabulated and thenpresented back to employees, typicallywithin their own work groups.

As with most organizational-developmentinterventions, the exact manner in which sur-vey feedback interventions are carried outvaries depending on the organization andconsultant involved. A common practice,though, is for consultants to first conductpersonal interviews or focus groups with atleast a random sample of employees. Based onthe themes that emerge from this initial stage,survey items are then developed (e.g., Gavin,1984). In other cases consultants may have astandardized instrument that they administerto employees. An advantage of using a stand-ardized approach is that an organization’sresults can be compared with those of otherorganizations in the same industry or thesame geographical location. A disadvantageof this type of survey instrument is that theitems are not specific to the particular orga-nization, so many issues that are important toemployees may be missed.

Although survey development is impor-tant, the feedback process is the key part ofsurvey feedback, and thus it will be dis-cussed in a bit more detail. In a model surveyfeedback program, the data are fed back toevery employee in the organization, in atrickle down procedure. That is, the topmanagement team of an organization firstreviews the survey results. Each member ofthe top management team then shares thedata with his or her subordinates. This pro-cess is repeated until everyone in the orga-nization has had an opportunity to reviewthe data. The feedback process is what makesit an organization-wide intervention, and itdistinguishes true survey feedback frommany employee opinion surveys conductedin organizational settings. For example, if aconsultant conducts a survey of organiza-

tional employees and prepares a writtensummary report for top management, butthe process essentially ends at this point, thisis not true survey feedback.

Research has shown that survey feedbackis an effective organizational-change inter-vention. In fact, Bowers (1973) comparedseveral organizational-development inter-ventions and found that survey feedbackwas the most effective. Other evidence hasalso shown that survey feedback is an effec-tive intervention (Neuman, Edwards, &Raju, 1989). Research has also investigatedmore specific factors that impact the feed-back process. For example, Klein, Kraut, andWolfson (1971) found that employees aremost responsive to feedback when it comesfrom their managers and is perceived as rel-evant to their actual work groups. Because ofthis, in many survey feedback interventions,two types of data are presented back to em-ployees: (1) data that represent the opinionsof the entire organization and (2) data thatrepresent the opinions of the department inwhich the work group may reside.

A second organization-wide interventionis changing the structure of an organization.As discussed in Chapter 13, organizationshave numerous options when it comes toorganizational structures. The rationalebehind structural change is quite simple. Ifall of the departments in an organization arerestructured, employees have no choice butto engage in at least some form of change.Unfortunately, though, the changes that arebrought about by structural changes mayultimately be only ‘‘pseudo’’ changes. Unlessa change in structure is accompanied bychanges in employees’ behaviors, this is arelatively ineffective method of organiza-tional change. For example, if an organiza-tion changes to a team-based structure, thiswill not result in meaningful organizational


change if individual employees continue toact primarily in their own self-interests.

Other common methods of facilitat-ing broad-based organizational change arethrough strategic planning and visioning in-terventions. Strategic planning is simply anorganization’s plan for what it is going toaccomplish and the mechanisms that willbe used to accomplish it. Strategic planningis an important function in an organizationbecause it drives important activities, such ashuman resources planning and compensa-tion, and it influences key decisions such asacquisitions. Given this importance, oneobvious way for an organization to effectchange would be to change its strategicobjectives. For example, the top manage-ment of an organization that had oncerelied primarily on getting its products tomarket quickly may decide that it wants tobe seen as a technological innovator. Thistype of strategic decision is highly signifi-cant, and an organization would likelyhave to change in a variety of ways in orderto carry out this new strategy. For example,this organization might decide to hire thebest technical talent available, or perhapsacquire a small but highly innovative com-petitor. It may also require a fundamentalchange in organizational culture, or per-haps even in structure.

The process of visioning simply requiresthe members of an organization (usually topmanagement) to decide (1) what the organi-zation would look like if it were ideal, (2)what the organization currently looks like,and (3) what strategic steps need to be takenin order to get the organization from where itis to where it could begin an ideal state. Anexample of this visioning process is depictedin Figure 15.6. This type of activity can bevery useful, particularly prior to strategicplanning, because it forces organizationaldecision makers to think about where they

are headed and, perhaps more importantly,how to get there.

Most typically, strategic planning andvisioning processes will represent the begin-ning of a long-term, large-scale organiza-tional-change process. These two macro-level interventions will necessitate manyother interventions at many other organiza-tional levels. At Harley-Davidson, for exam-ple, a long-term process of organizationalchange essentially began with a simplevisioning exercise (Teerlink & Ozley, 2000)and then branched out to a variety of otherinterventions needed to help that organiza-tion get from where it was to where it ulti-mately wanted to be (see Comment 15.2).When looked at in this way, it is probablymore accurate to think of very global inter-ventions (such as visioning) as precursors toorganizational change rather than interven-tions themselves.

A form of visioning that may be used inconjunction with strategic planning isreferred to as scenario planning (Schwartz &Ogilvy, 1998). In scenario planning, mem-bers of an organization come together anduse multiple scenarios of the future envi-ronment in order to facilitate discussion,and they develop plans for responding tothese scenarios. Steil and Gibbons-Carr(2005) describe how scenario planningwas used at a large conference of facultyfrom dental schools. Using this method,conference participants planned potentialresponses to future challenges in dentaleducation.

FIGURE 15.6Example of the Visioning Process

Organizationas It

CurrentlyExists

Visionof theIdeal

Organization

Steps Needed to Getfrom Currentto Ideal State


CONDITIONS NECESSARYFOR SUCCESSFULORGANIZATIONAL CHANGE

Up to this point, we have defined organiza-tional development, covered the theoreticalbase of this field, and described some of the

more commonly used organizational-development interventions. In this section,we shift the focus to the organizational-development process. More specifically,we examine organizational factors thatimpact whether or not organizationalchange and development interventions are

COMMENT 15.2

ORGANIZATIONAL CHANGE AT HARLEY-DAVIDSON

IN TERMS OF brand loyalty and identification,

few organizations have been as successful as

motorcycle manufacturer Harley-Davidson.

Its products not only provide transportation,but they have really come to symbolize a life-

style as well. Harley-Davidson endured some

very tough times in the early 1980s due to

Japanese competition and labor problems.

Remarkably, the company pulled together,

weathered this crisis, and by the late 1980s

was again very profitable. Unfortunately, to

get through this crisis, the organizationadopted a very rigid ‘‘command and control’’

culture in which employees obeyed orders but

took little personal initiative. This culture

helped save the company; but, unfortunately,

changes were needed if the company was to

continue to grow in the absence of an external

crisis.

In the book More than a Motorcycle: TheLeadership Journey at Harley-Davidson, Rich

Teerlink and Lee Ozley describe the difficult,

long-term process of moving from a ‘‘command

and control’’ culture to one in which employees

are empowered and feel a sense of ownership.

What’s striking about this process is that it

began with a very simple visioning exercise.

Specifically, the top executives at Harley-Davidson, with the assistance of a consultant,

discussed how the company was currently

functioning and how the company ideally

should be functioning. An examination of

the differences between these two (current

versus ideal) began a process that has gone

on for over 10 years and has led to significantchanges in organizational structure, perfor-

mance appraisal, compensation practices,

and labor relations.

A number of important lessons about orga-

nizational change can be learned from this

book. Perhaps most important is: For organi-

zational change to be meaningful, it must be

comprehensive. Another important lesson isthat organizational change must be supported

by top management in order to succeed. Rich

Teerlink, one of the authors, was the chief

executive officer of Harley-Davidson through

most of this change process, and he strongly

supported it. Without his support, it is doubt-

ful that this process would have survived.

Finally, in reading this book, it becomes veryobvious that this change process did not always

go smoothly. There were times when union

support of the change process was low and

other times in which more immediate crises

took higher priority. Ultimately, though, the

organization kept at it and produced mean-

ingful, positive organizational changes.

Source: R. Teerlink and L. Ozley. (2000). More than

a motorcycle: The leadership journey at Harley-Davidson.

Boston, MA: Harvard Business School Press.


successful. These include top managementsupport, the consultant who guides theorganization through the change process,general resistance to change, and organiza-tional ownership of the change and devel-opment process.

Top Management Support

For most initiatives in organizational settingsto succeed, support from top managementpersonnel is crucial; in fact, a lack of topmanagement support is almost a guaranteeof failure. Thus, it should come as no sur-prise that top management support is a keyfactor in determining whether or not organi-zational-development programs are success-ful. In this section, we examine why topmanagement support is so important to thesuccess of organizational change.

A very practical reason that top manage-ment support is crucial is that top manage-ment personnel have a great deal of controlover organizational resources. This is impor-tant because organizational change and de-velopment programs are very expensive. Notonly does an organization typically have topay the fees of outside consultants; in mostcases, organizational change and develop-ment programs require a great deal of em-ployees’ time. Employees’ time is a finiteresource; therefore, during the time thatemployees are engaged in organizational-development activities, they are not produc-ing products, serving customers, or doingother things that directly support the missionof the organization. If top managementpersonnel do not commit these types of re-sources to an organizational developmenteffort, it will likely fail.

Top management support is also crucialbecause, in most organizations, top man-agement personnel provide an organizationwith its strategic direction. Organizations

do not change in a random fashion—rather,they typically change for some logical rea-son. Top management’s involvement andsupport are essential in order to give anorganizational-change effort proper direc-tion. If an organizational-change effort wereto be conducted without this involvement,it is quite possible that change might occur,but the change might be counterproductive.

A final reason that top management’s sup-port is so crucial is that it has symbolic value.Although it is unlikely that most employees inorganizations feel a strong personal connec-tion to members of their top managementteam, they do look to these individuals forguidance. If members of the top managementteam are indifferent toward an organizational-development effort, this signals to employeesthat the program is unimportant. Conversely,if top managers enthusiastically support it,this communicates to employees that theorganizational effort is important and thatthey should be committed to it.

The Consultant Guiding the Process

Many students see organizational-develop-ment consulting as a glamorous professioncharacterized by high earnings, interestingwork, and exciting travel opportunities.While there is certainly some truth to this, itis also true that organizational-developmentconsulting is very hard work. Facilitating day-long meetings, traveling to confer with clients,and summarizing interview and survey datacan all be physically and mentally draining.Consultants must also constantly be aware ofthe ethical implications of their behaviors.

Organizational-development consultantshave a very strong impact on whether anorganizational-change program is successful.A highly skilled organizational-developmentconsultant will greatly enhance the odds thatan organizational-change program will be


successful, even if the organization may ini-tially be vague about what it is trying toaccomplish. On the other hand, a poorlytrained organizational-development consult-ant who is lacking in skill will most certainlyincrease the odds that an organizational-change program will fail.

Given the importance of the organiza-tional-development consultant, a questionthat naturally arises is: ‘‘What are theattributes of an effective organizational-development consultant?’’ This is a difficultquestion to answer because the success of aconsultant largely depends on the fit be-tween his or her skills and personality, andthe attributes of the client organization. Aconsultant may be a smashing success inone organization but a miserable failure inanother. Despite this situational specificity, itis possible to come up with a set of generalattributes that tend to characterize successfulconsultants.

One attribute that may seem obvious toreaders is that consultants need to have awell-developed knowledge of organizationaldevelopment and, more generally, behaviorin organizations. This is important becauseconsultants observe behavioral processes inorganizations and, based on these observa-tions, make recommendations about organi-zational-development interventions. Suchknowledge often (but not always) comesfrom graduate study in the behavioral scien-ces or related fields (e.g., organizational be-havior, industrial relations, human resourcesmanagement). A consultant, however, mayalso acquire such knowledge through severalyears of working in a corporate training anddevelopment department, or through con-tinuing education.

In addition to knowledge, several skillsare needed in order to develop a successfulconsulting relationship with an organization.Probably the most basic skill consultants

need to have is listening. Particularlywhen a consultant first makes contactwith an organization, it is important that heor she gathers as much information aboutthat organization as is possible. The only wayto do that is for the consultant to activelylisten as members of the organization explainwhy they want to embark on a program oforganizational change. Consultants mustalso listen to members of an organizationdescribe the history and culture of the orga-nization, since problems (and solutions toproblems) occur in a historical and culturalcontext.

It is also important for consultants topossess very well-developed communicationskills. Consulting involves a great deal ofverbal communication with the members ofa client organization, but it also requires thatthe consultant can communicate effectivelyin other ways (e.g., in writing, nonverbally).Consultants must be highly skilled commu-nicators because much of their communica-tion with the client organization comes in theform of feedback about problems and pro-cess issues. If an organization is going tochange successfully, the consultant’s feed-back must be frequent and easy for membersof an organization to understand.

Organizational-development consultantsalso need to have a solid grasp of researchmethodology and data analysis. Organiza-tional-development consulting almost alwaysinvolves the collection of empirical data aboutthe client organization, analysis of those data,and a summary of the data in a form thatmembers of the client organization can readilyunderstand. Consultants who do not possessat least minimal skills in data collection andanalysis, and who are unable to summarizedata for nontechnical audiences, are at adistinct disadvantage compared to thosewho can. Consultants who lack these skillsmay rely on less rigorous methods of data


gathering and ultimately may compromisethe quality of an organizational-change effort.

In addition to knowledge and skill, someintangible qualities are important for organi-zational-development consultants to possess.For example, any type of consulting demandsa very strong sense of ethics. Inevitably, con-sultants are told things ‘‘in confidence,’’ andthey may even detect that organizationalmembers are attempting to use their servicesfor political purposes. These are difficult sit-uations that require the highest ethical stand-ards. It is also important that consultantsclearly communicate these standards to cli-ents before accepting a consulting assign-ment.

Another highly important, yet intangible,quality necessary for organizational-develop-ment consulting is flexibility. Effective con-sultants are flexible enough to see anorganization and its problems as they reallyare, not as they are seen through the lens of aparticular theory or their favorite interven-tion. This type of flexibility may at times leada consultant to conclude that an organiza-tion’s problems do not fit his or her skill set;thus, a particular consulting assignment maynot be appropriate. In the short term, thismay result in lost consulting revenue; in thelong run, however, it helps to build a con-sultant’s credibility. Flexibility also requiresa willingness to see multiple sides of an issue,and occasionally to admit that one is wrong.If a consultant offers advice or an assessmentof a problem that ultimately turns out to bewrong, he or she needs to be flexible enoughto admit the error and learn from it.

A final intangible characteristic that iscrucial in effective consulting is credibility.How does a consultant establish and main-tain credibility with an organization? Whenconsultants initially come into contact withorganizations, they are often sought out

because of their academic credentials, priorwork they have done, or an influential bookthey have written. In such cases, the organi-zation certainly has a positive view of theconsultant, and positive expectations abouthis or her capabilities, but the consultantdoes not yet have credibility. This is becausecredibility is an earned commodity, not aprofessional entitlement. Credibility isearned over time as the consultant interactswith the organization.

Perhaps the most important factor indetermining credibility is whether the con-sultant keeps his or her word. When theconsultant tells an employee that what heor she says will be held in confidence, is thisinformation really held in confidence? Whenthe consultant tells an organization that areport summarizing an employee attitudesurvey will be completed by a certain date,is the report delivered on or before thatdate? When a client organization is toldthat a project will cost a certain amount,does the consultant bill for that amount?These are obviously hypothetical examples,but they serve to underscore the point thatcredibility is something that consultantsaccrue each time they meet (or, ideally,exceed) the expectations of the client orga-nization.

General Resistance to Change

Humans are basically creatures of habit and,as such, they find a great deal of comfort inroutine and familiarity. Consequently, evenconsidering a change often evokes a greatdeal of apprehension and anxiety. Thisgeneral principle certainly applies in theworkplace as well. As was shown in Chapter14, people in organizations develop routinesand rituals surrounding many behaviorsand these become highly ingrained in an


organization’s culture. Once behaviors be-come highly ingrained, it is very difficult tochange them, even if such changes are pos-itive. Fear of change often comes from a be-lief that changes will have a negative impact,or from just general fear of the unknown.

It is unlikely that organizations can evercompletely eliminate resistance to change.What can be done, though, is to introduceorganizational changes in a way thatdecreases the potential for resistance. Oneway of doing this is to provide the opportu-nity for employee participation in the imple-mentation of the organizational changes. Aswas shown over 50 years ago by Coch andFrench (1948), employees are much morereceptive to changes when they have theopportunity to participate in their implemen-tation. Autocratic methods of organizationalchange may appear to be much quicker andeasier than participative methods, but, in thelong run, organizations stand a much greaterchance of encountering employee resistanceif changes are imposed on employees. Re-search over the years has certainly affirmedthe value of allowing employees to partici-pate in organizational changes that mayimpact them (e.g., Piderit, 2000).

Change can also be made less threateningto employees if organizations maintain ahigh level of communication with employeesthroughout the change process. This isimportant for two reasons. First, a high levelof communication is likely to increase em-ployee perceptions of procedural justice withrespect to an organizational change. Morespecifically, employees will feel as thoughthe organizational changes will be handledin a way that is fair to all employees. Com-munication is also important because it pro-vides employees with a rational basis to assessthe costs and benefits of the change. Researchhas shown that employees are less likely to

resist organizational changes if they see thatthe benefits outweigh the costs (Giangreco &Peccei, 2005), so clearly communicatingthese provides employees with the informa-tion necessary to make this judgment.

A final thought to consider is that in somecases strong resistance to change among em-ployees may be a sign that an organizationalchange should be reconsidered. While resist-ance to change is typically treated by re-searchers as something that must be‘‘overcome’’ (Piderit, 2000), employee reac-tions to proposed changes may at times be avaluable ‘‘reality check’’ for an organization.This is especially true in organizations wheretop management has sought out little inputfrom rank-and-file employees prior to goingahead with organizational-change programs.

Organizational Ownership of theChange and Development Process

It is often tempting for organizations to abdi-cate responsibility for organizational changeand development programs to outside con-sultants. The primary reason for this is quitesimple: Changing an organization is veryhard work. Developing a new vision for anorganization, participating in team-buildingactivities, or conducting survey feedbackmeetings are all mentally and physicallydraining activities. Given this difficulty, it istempting for members of an organization tosit back and let a consultant take care of thesechores and submit a report to the organiza-tion. Ultimately, though, the solutions thatcome strictly from consultants with noinvolvement from members of an organiza-tion may not mesh well with the realities ofan organization’s culture (Schein, 1987). Em-ployees may also have little enthusiasm forimplementing changes recommended by aconsultant.


Organizations may also resist ownershipbecause organizational development can bevery threatening. During the course of anorganizational-development program, em-ployees are often faced with some harshrealities. For example, they may realize thattheir organization is not as effective as itcould be, or that it is not as ideal a place towork as they once thought it was. Even morethreatening for employees is the possibilitythat they may bear some of the responsibilityfor some of the organization’s problems.Viewing an organizational-development pro-gram as ‘‘something the consultant does’’ maybe a way of sparing employees the difficulttask of confronting the organization’s (and,by implication, their own) shortcomings.

One thing consultants can do to increaseownership is to stress its importance atthe beginning of the organizational-develop-ment process. Often, in an effort to win aconsulting contract with an organization, aconsultant is tempted to portray the role ofthe organization as being very minimal (e.g.,the consultant will take care of everything).However, in the long run, this model is oftencounterproductive because the change pro-cess rarely works without the participation ofmembers of the client organization.

Consultants can also increase organiza-tional ownership by striving to make the clienta true partner in the change process. It is oftentempting for consultants to give organizationsadvice in a very top-down or authoritativemanner, particularly when they see thingsbeing done in the organization that are obvi-ously counterproductive. If consultants canresist this temptation, however, and exploresuch counterproductive practices in a collab-orative manner, this will ultimately be ofgreaterhelp to theorganization. It is alsomuchmore empowering to the client organization,because when problems occur in the future,the organization will be able to solve them.

EVALUATION OFORGANIZATIONAL-DEVELOPMENT PROGRAMS

At this point in the chapter, some readersmay find the organizational-developmentprocess, and the various interventions usedby organizational-development consultants,quite interesting.Moreskeptical readers,how-ever, may be asking themselves: ‘‘Does any ofthis stuff actually work?’’ This is certainly animportant question, given the large amount ofresources that are required to implement anorganizational-development program.

In this section, we tackle the difficultquestion of whether organizational develop-ment actually ‘‘works’’ and, perhaps just asimportant, why it works. We do this by firstexamining the issue of evaluative criteria.One of the problems in evaluating anythingis that we are often vague in specifying thecriteria by which we judge effectiveness. Wethen discuss the most common researchdesigns that are used in the evaluation oforganizational-development programs andthe difficulty of measuring change. Finally,a brief review of the published literature onthe effectiveness of various organizational-development interventions is provided.

The Problem of Evaluative Criteria

Within the broader field of industrial/orga-nizational psychology, ‘‘the criterion prob-lem’’ is typically associated with personnelselection (e.g., Campbell, 1990). To be sure,when organizations attempt to systematicallyvalidate selection procedures, deciding onappropriate criteria is a major challenge.What measures indicate whether or not anemployee is effective? How do we assess thecontribution of an employee to an organiza-tion’s success? Both of these questions needto be answered if one is to truly assess thevalue of employee-selection procedures.


The criterion problem is also an impor-tant issue when evaluating organizational-development programs. If an organiza-tional-development program is viewed in aholistic manner (e.g., the diagnosis, inter-ventions, follow-up), the criterion problembecomes particularly acute. This is becausethe primary goals of most, if not all, organi-zational-development programs are thesame: enhanced organizational effectivenessand employee well-being (Porras & Robert-son, 1992; Kohler & Munz, 2006). How,then, can we show that an organization ismore effective and employee well-being hasimproved after the implementation of anorganizational-development program? Oneway this could be done is through standardfinancial measures such as stock price, salesrevenues, return on assets, market share,health care costs, or any number of othermeasures that have been developed throughcost-accounting procedures.

Financial data are appealing becausethey have the feel of objectivity. When anorganization improves its financial perfor-mance or employee health care costs arereduced following the implementation ofan organizational-development program, itis hard to argue that such results were dueto methodological artifacts. A disadvantageof the financial indexes mentioned previ-ously is that they represent a fairly narrowview of organizational effectiveness and em-ployee well-being. For example, is an orga-nization truly effective if its stock price goesup, but it lays off 30% of its workforce? Arean organization’s employees better off ifhealth care utilization goes down 20%, yetits rate of turnover is one of the highest in itsindustry?

Another way to assess organizational ef-fectiveness and employee well-being, whichcan be equally as narrow, is to do so primar-ily in terms of employee behaviors and atti-

tudes. For example, organizations might usethe results of employee opinion surveys,rates of turnover, or number of grievancesfiled to indicate in a general sense how wellthe organization is performing. These effec-tiveness criteria may be much more relevantto the aims of an organizational-developmentprogram than financial data alone. Unfortu-nately, though, these criterion measures maybe as narrow as the financial indexes de-scribed previously. Is an organization effec-tive if its employees are highly satisfied withtheir jobs, yet its stock price declines steadilyover a 3-year period? Is an organization effec-tive if its employees report low levels ofstress, yet it continues to lose market shareto its competitors?

The best approach to measuring overallorganizational effectiveness or employee well-being is to utilize multiple criteria measures.This provides the most complete picture oforganizational effectiveness and, in fact,yields the most comprehensive view of theimpact of organizational-development pro-grams. Also, if such data are collected over arelatively long period of time, they may beparticularly revealing. For example, in somecases, the initial impact of organizational-development programs on traditional finan-cial measures is actually negative (e.g., Griffin,1991), but, over time, may ultimately end upbeing positive. Conversely, with measures ofemployee attitudes and well-being, employ-ees’ initial reactions to organizational devel-opment may be very positive. Over time,however, the novelty of an organizational-development program may wear off, and suchmeasures may return to baseline levels.

Deciding on the proper criterion measureis a bit easier if evaluation efforts are focusedon specific organizational-development in-terventions, as opposed to a general programof organizational change. For example,evaluating an organizational-development


program that features team building wouldlogically call for some assessment of group orteam effectiveness as the criterion. Further-more, within the group-effectiveness litera-ture, there exists some excellent guidance asto the criteria that define group effectiveness(e.g., Guzzo & Shea, 1992; Hackman, 1987).Even with such guidance, however, comingup with the actual criterion measures may bemore easily said than done.

Evaluation Research Designs

A research design is simply a plan for datacollection (Cook, Campbell, & Perrachio,1990). When any organizational interven-tion is evaluated, some type of research de-sign is needed because some data needs to becollected in order to assess the impact of thatintervention. In organizational settings, re-searchers typically have much less controlover data collection than do laboratory re-searchers. As a result, when conducting eval-uation research, a number of methodological‘‘compromises’’ usually have to be made inorder to obtain any data showing the effec-tiveness of interventions. For example, inorganizational settings it is rare to be ableto randomly assign research participants toconditions, and researchers have little con-trol over either the implementation of theintervention or extraneous variables thatmay impact the intervention. That’s not tosay that this is always the case (see Shperling& Shirom, 2005), but it is typical of fieldsettings.

The most typical way that organizational-development practitioners deal with chal-lenges in data collection is simply not toconduct any form of systematic evaluationat all (Porras & Robertson, 1992). Often, ifan organizational-development interventionseems like it might be helping, this may begood enough for upper management. For

those trained in the behavioral sciences, thisapparent disregard for rigorous evaluation oforganizational-development programs istroubling. However, when one considers thatevaluation can be a very human and often apolitical process, this disregard for evaluationis often quite rational (see Comment 15.3).

One of the most common designs used inevaluation research—and, unfortunately,one of the least powerful—is the One Group,Posttest Only design. In this type of design,participants in an organizational-develop-ment intervention would simply indicatewhether or not they felt the interventionwas useful or effective. This is obviously avery easy design to implement. However, amajor disadvantage of this design is thatthere is no baseline from which to evaluatethe impact of the intervention. Also, becausethere is no control group there is no way ofknowing whether there would have been apositive change in the criterion measure if theorganizational-development intervention hadnot been implemented. Due to these weak-nesses, Cook and Campbell (1979) refer tothis design as generally uninterpretable.

Another commonly used design is theOne Group, Pretest–Posttest design. In thiscase, a baseline measure is obtained bothprior to and after an organizational-development intervention or program is im-plemented. For example, measures of groupprocess may be obtained prior to and 3months after the implementation of a team-building intervention. The advantage of thisdesign is that there is a baseline measure.However, because there is still no controlgroup, it is impossible to tell whether theobserved effects would be obtained inthe absence of the intervention. With theteam-building example, it is possible that,as team members get to know each otherbetter over time, their interactions and sub-sequent group processes would improve


even in the absence of a team-building inter-vention.

To address this issue, the researcheractually has several options. One is to adda control group and thus create a Pretest–Posttest with Control Group design. Bydoing this, the effects of the interventioncan be compared against those of a groupthat does not take part in the intervention,thus providing the researcher with a firmerbase for judging the effectiveness of the in-tervention. Unfortunately, in most organiza-tional settings, control groups are hard tocome by because of a limited number of

employees, and in many cases the unwill-ingness of management to withhold a poten-tially valuable intervention.

If a control group cannot be obtained,researchers can use the Pretest–Posttest withMultiple Dependent Measures design. Thisdesign is similar to the previously describedOne Group, Pretest–Posttest design exceptthat the researcher measures two sets ofdependent measures: one set that should beimpacted by the intervention and another setthat should not be. If, for example, a team-building intervention has a positive impacton group-process measures, yet it has no

COMMENT 15.3

THE POLITICS OF EVALUATION

IN MOST DISCUSSIONS of evaluation, including the

one in this chapter, the bulk of the material

deals with technical issues such as evaluative

criteria and research designs. Furthermore,these technical issues are important to con-

sider if organizational interventions are to be

evaluated properly. Unfortunately, however,

in focusing on the technical aspects of evalua-

tion, we often overlook the politics of the

evaluation process.

Why is evaluation often a political process?

The key to answering this question is to under-stand what is actually occurring when an eval-

uation takes place. When we evaluate

something, we are literally determining its

value. For example, if we are evaluating a

team-building program, what we are really

doing is determining whether this program

has any value to the organization. One of the

reasons that organizations often do not want toknow the true value of organizational-develop-

ment interventions is that they’re afraid of what

they might find out. Obviously, if an interven-

tion is evaluated and found to be very effective,

everyone’s happy.

However, what happens when an inter-

vention is evaluated and found to be very

ineffective? In this case, everyone is unhappy,

particularly if the organization has investedconsiderable resources in the intervention.

There may also be cases in which a particular

individual (e.g., a Human Resources director)

initiates an intervention. If this is the case, such

an individual has a great deal riding on the

success or failure of that intervention. Thus, for

political reasons, it may not be in this person’s

best interests to conduct a thorough, methodo-logically rigorous evaluation.

The major point is that, even though eval-

uation is a technical process, it is also a very

human process. Useful evaluation can only

take place in organizations in which people

are willing to accept negative outcomes, to

learn from them, and ultimately to improve

the implementation of future interventions.Unfortunately, based on my own experience

and the low prevalence of rigorous evaluation,

such organizations are probably more the

exception than the rule.


impact on participants’ satisfaction withfringe benefits, this suggests that the interven-tion was effective.

A second option is a One Group, Inter-rupted Time Series design. This is similarto the One Group, Pretest–Posttest designexcept that several measures of a dependentvariable are obtained prior to and after theintervention. By obtaining these multiplemeasures, the researcher is able to statisti-cally model the behavior of the dependentmeasure over time and, more importantly,determine whether the intervention has anyimpact on that measure. This design is alsouseful because it allows the researcher someassessment of whether the effects that areobserved persist over time.

The research designs briefly described inthis section represent only a small sample ofthose available for the evaluation of organi-zational-development interventions or entireprograms. Readers seeking more informationabout research designs are urged to consultCook and Campbell (1979) or Cook et al.(1990). More specific information aboutevaluation research can be found in Shaw,Greene, and Mark (2006) and Bamberger(2006). Given the large number of researchdesigns available, in most cases it is possibleto conduct some form of evaluation of orga-nizational-development programs even inthe most challenging of field situations.

The Challenge of Measuring Change

The primary aim of evaluation research is tomeasure change. Although much has beenwritten about the statistical issues surround-ing the measurement of change (e.g., Cron-bach & Furby, 1970), conceptual issuessurrounding the meaning of change alsoimpact the evaluation of organizational-development programs. In this section, we

focus on the conceptual meaning of changein organizational-development research.

Golembiewski, Billingsley, and Yeager(1976) proposed that change in responseto organizational-development interventionsand programs can come in three varieties,which they labeled as Alpha, Beta, andGamma change, respectively. Alpha changerepresents what we typically think of astrue change; that is, an organization trulyimproves in some way after the organiza-tional-development intervention. The mean-ing of the dependent measures used, as wellas the manner in which participants perceivethe measures, remain the same both prior toand after an organizational-development in-tervention. As an example, an organizationmay implement a survey feedback programand find that the levels of job satisfaction ofemployees truly increase as a result of thisprogram.

Beta change represents a situation inwhich the participants’ frames of referencechange with respect to the dependent meas-ure being assessed. Thus, beta change canbe thought of as a methodological artifactrather than a true form of change. As anexample, suppose an organization imple-mented a team-building intervention de-signed to improve communication withinwork groups. To evaluate this intervention,group members’ perceptions of communica-tion may be measured prior to and followingthe intervention. In this case the researchermight find that scores on the communicationmeasure indicate that levels of communica-tion have improved following the interven-tion. However, an alternative explanation isthat, by going through this intervention, par-ticipants develop a greater awareness of thevarious ways in which people in groupscommunicate. Thus, what seemed to partic-ipants like a ‘‘small’’ amount of communica-tion prior to the intervention, may seem like


an ‘‘above-average’’ amount following theintervention. Note that in this example theactual level of communication has not reallychanged. What has changed, however, isparticipants’ definition of the various pointson the scale measuring communication.

Gamma change represents a situation inwhich participants essentially redefine orreconceptualize key dependent or outcomevariables. Like beta change, this can also bethought of as a methodological artifact.Golembiewski et al. (1976), however, arguedthat because one of the goals of organiza-tional development is to change employeeperspectives, this may actually be a legiti-mate form of change. As an example of betachange, an organization may implement anintervention designed to increase employeeinput into organizational decision making.To evaluate this intervention, the level ofparticipation may be assessed prior to andafter the program has been implemented.Prior to the intervention, employees mayconceptualize participation as being some-what narrow and something that is not partof their role. As a result, they may rate thisvery low. After the intervention, however,they may view participation in a muchbroader sense, see it as something that ispart of their job, and, consequently, rate itmuch higher than they did in the pretest.Note that in this example the actual level ofparticipation has not changed. Rather, em-ployees have changed the way they defineparticipation.

How does the researcher distinguishamong alpha, beta, and gamma change whenevaluating an organization program? One wayis obviously to decrease reliance on self-reportcriterion measures, or at least supplementself-reports with other criterion measures(e.g., Liu, Spector, Jex, 2004; Spector & Jex,1991). Both beta and gamma change effectsare due to cognitive processes. Thus, these

artifacts can be avoided by utilizing at leastsome nonself-report measures in the evalua-tion process. For example, archival measuresof performance could be used in conjunctionwith survey measures of employee attitudeswhen evaluating the impact of an organiza-tional-development intervention.

To assess whether gamma change hasoccurred, one recommendation has been tocompare the dimensionality of scales used tomeasure key dependent measures at pretestand posttest (Armenakis, Bedeian, & Pond,1983; Armenakis & Zmud, 1979). Recall thatgamma change represents a change in respon-dents’ conceptualization of key dependentmeasures. Given this change in conceptu-alization, it is possible that the dimension-ality of scales may change from pretest toposttest measurement. While this is cer-tainly logical, at least from a statisticalpoint of view, in practice it is often difficultto apply. Individual scale items contain agood deal of measurement error (Nunnally& Bernstein, 1994), and evaluation studiesoften have sample sizes that severely limitthe usefulness of such analyses.

Evidence on the Effectivenessof Organizational Development

As was stated earlier, many organizational-development programs are conducted with-out the benefit of any formal evaluation.Fortunately, enough empirical evaluationsof organizational-development programshave been conducted over the years and theyhave yielded several summaries, using bothqualitative and quantitative methods. Themost widely cited qualitative summary of theeffectiveness of a number of organizational-development interventions was conductedby Bowers (1973). The most importantfinding from this review was that when anumber of organizational-development


interventions were compared, survey feed-back appeared to be the most effective.

Another widely cited qualitative review oforganizational-development studies was con-ducted by Terpstra (1981), who summarized52 published organizational-developmentstudies. Compared to Bowers’ (1973) review,Terpstra’s was more systematic because hecoded each study reviewed, in terms ofwhether the effects of the organizational-development program were positive, neutral,or negative. Based on this classification, hisconclusion was that the effects of organiza-tional development are generally positive.This effect, however, was found to be mod-erated by methodological rigor; positive find-ings were most likely to be found in studiesthat were lacking in methodological rigor.Subsequent studies, however, have not sup-ported Terpstra’s claim of a pervasive ‘‘posi-tive findings’’ bias in organizational-development evaluation (e.g., Bullock &Svyantek, 1983; Woodman & Wayne, 1985).

In more recent years, researchershave applied meta-analytic methods to theevaluation of organizational-developmentinterventions. For example, meta-analyticreviews have supported the effectivenessof MBO (Rodgers & Hunter, 1991), aswell as a number of other organizational-development interventions (e.g., Guzzo,Jette, & Katzell, 1985; Neuman et al.,1989). For the most part, these quantitativereviews suggest that many organizational-development interventions have a positiveimpact on a number of employee attitudesand behaviors. It is also true, however, thatin all of these meta-analyses, a portion ofvariance in the effects is left unexplainedafter accounting for statistical artifacts suchas sampling error, unreliability, rangerestriction, and so on.

This suggests that, while organizational-development interventions often have posi-

tive effects, these effects may vary consider-ably acrossorganizations. Suchvariability alsoprobably reflects the fact that many organiza-tional programs, as well as interventionswithin programs, fail. Relying on the pub-lished literature may slightly overestimatethe effectiveness of organizational-development programs and interventionsbecause studies that show positive effectsstandamuch better chanceofbeingpublishedthan those that show no effects or even nega-tive effects (see Dahlsten, Styhre, & Will-ander, 2005).

SPECIAL ISSUES IN CLIENT--CONSULTANT RELATIONSHIPS

At this point, a relatively comprehensivecoverage of organizational development hasbeen provided. However, one aspect of theorganizational-development process that hasnot been covered is the relationship betweenthe organizational-development consultantand the organization. This is an importanttopic because some readers may ultimatelyassume the role of organizational-develop-ment consultant at some point in theircareers. Therefore, in this final section, webriefly examine some of the importantissues that a consultant faces in facilitatingorganizational-development programs.

Balancing the Needs of MultipleClients

When providing consulting services to anorganization, the first client an organiza-tional-development consultant encountersis typically a member of the top managementteam or, in some cases, a human resourcesexecutive. Individuals at these levels ofteninitiate organizational-development activi-ties because they tend to focus on ‘‘big


picture’’ issues such as overall organizationaleffectiveness and change. In addition, indi-viduals at these levels are typically able toauthorize the budgetary resources neededfor consulting services.

Once a consultant is in an organization,however, the list of ‘‘clients’’ begins to growvery quickly. For example, depending on thesize of the organization, a consultant mayvery well come into contact with employeesranging from division heads to hourly work-ers. One of the challenges that consultantsoften face, particularly as they deal withemployees at lower levels of the organiza-tion, is the perception that they are onlyserving the interests of those in upper man-agement. This is a difficult challenge, espe-cially because consultants are typicallybrought into organizations by upper-management employees.

Probably the most effective way to bal-ance the needs of multiple clients is toaddress the issue in the earliest stages ofthe consulting relationship. That is, when aconsultant first enters an organization, heor she needs to make his or her initialcontact aware that there will likely be mul-tiple, and often competing, interests amongvarious employees in the organization.After making the client aware of this, thenext issue that needs to be confronted ishow the consultant will handle such com-peting interests.

In the authors’ opinion, consultants arebest served by maintaining a neutral stancewhen faced with such situations. If a con-sultant appears to be too ‘‘pro-management,’’he or she risks losing credibility withlower-level employees and may then findit very difficult to obtain their cooperationin carrying out the organizational-develop-ment program. Conversely, becoming tooaligned with lower-level employees mayput a consultant in the awkward position

of going against the individual(s) whofacilitated his or her entry into the organi-zation. Ultimately, consultants are bestserved by assuming that conflicts of inter-ests will occur and making it clear to allparties involved that the consultants willnot take sides.

Maintaining Confidentiality

A very simple rule of thumb for consultantsregarding confidentiality: If you assure some-one that the information he or she is givingyou is confidential, honor that commitment.In actuality, though, maintaining confiden-tiality in consulting relationships can be a verycomplex issue that poses some serious ethicaldilemmas. One of the reasons is that con-sultants are often tempted to violate it. Forexample, a member of upper managementmay want to know what a particular vicepresident said about her leadership styleduring a focus-group session. A group ofhourly employees may want to know whatupper management is planning to do withregard to a rumored merger. In both ofthese hypothetical cases, the consultantmay be tempted to reveal confidential in-formation, primarily because it wouldplease the party receiving the informationand is reinforcing to the consultant.

Confidentiality may also be compro-mised due to carelessness on the part of theconsultant or those working for the consul-tant (e.g., Gavin, 1984). For example, a con-sultant may provide the top managementgroup with written comments from a surveyand forget to first delete the names of therespondents. A consultant may unknowinglyleave the notes from an interview with anemployee in a place where anyone can seethem. Obviously, in both cases, the violationof confidentiality is completely uninten-tional. However, regardless of the intent,

Special Issues in Client–Consultant Relationships 509

such lapses are embarrassing to the consul-tant and could severely damage a consultingrelationship.

Perhaps the most typical reason forconsultants’ being pressured to revealconfidential information is because of mis-understandings regarding the limits of con-fidentiality. For example, the organization’stop echelon may feel that they will be able toview completed questionnaires that employ-ees have filled out, but the consultant plansto provide them with only summary data.Or, perhaps the manager of a group plansto find out ‘‘who said what’’ during thepersonal interviews of group membersprior to a team-building session, but theconsultant plans to report only the generalthemes that emerged during these interviews.In both of these cases, the problem is a lack ofclarity regarding the confidentiality of theinformation being collected by the consul-tant.

The best way to avoid such misunder-standings is for the consultant to clarify allissues surrounding confidentiality prior toformally entering into the consulting relation-ship. Often, when consultants begin workingwith an organization, there is a temptation to‘‘dive right in’’ and simply deal with issuessuch as confidentiality as they arise. In thelong run, though, consultants and organiza-tions are much better served by taking thetime to thoroughly address issues such asconfidentiality prior to beginning a project.

Terminating a ConsultingRelationship

Some consulting assignments are short-termin nature, so there are clear starting and end-ing points. However, consultants also developrelationships with organizations that are morelong term, and the services provided are lesswell defined in scope. A consultant may

essentially be ‘‘on retainer’’ and be utilizedby an organization in a variety of ways. Thisis helpful to an organization because, overtime, the consultant may develop consider-able knowledge of the organization and itsproblems. Long-term relationships with orga-nizations can also be very advantageous for aconsultant. They may be quite lucrative andmay also result in the development of somevery satisfying relationships with members ofthe organization. However, what happens ifrelationships like these go sour?

Deciding when to terminate a consultingrelationship is obviously one of the mostdifficult issues a consultant faces, and thereare no easy solutions. A common criterionthat consultants use for making this decisionis whether the consulting relationship isprofessionally satisfying. Particularly forpart-time consultants, providing consultingservices is done more out of interest in theactivity than for financial gain. Therefore,some consultants will decide to terminate aconsulting relationship when the consultingactivity is no longer of interest.

Another common basis on which a con-sultant may decide to terminate a consultingrelationship is whether he or she is helpingan organization. Recall from the discussionof process consultation (Schein, 1987) thatconsulting is essentially a form of profession-al helping. Given this definition, a consultantmay conclude that, although he or she istrying to be helpful, the services being pro-vided are not really doing the client organi-zation any good. This may be due to the factthat different services may be needed, orperhaps the organization has improved to acertain level and further improvement isunlikely.

Finally, a consultant may decide to ter-minate a consulting relationship due tophilosophical or value differences withthe client organization. A consultant may


terminate a relationship upon discoveringthat the organization condones sexual har-assment of its female employees, or that it‘‘looks the other way’’ in response to otherunethical behaviors. Making the decisionto terminate an otherwise satisfying con-sulting relationship for these types of rea-sons is obviously difficult. However, in thelong run, a consultant is probably betteroff terminating such a relationship andworking with organizations where suchconflicts do not exist.

CHAPTER SUMMARY

In this chapter, we examined the importantarea of organizational change and develop-ment. We began by defining organizationaldevelopment and describing the typical rea-sons that organizations engage in programsof planned change. The focus of the chapterthen shifted to a discussion of the historicalroots of organizational development. As wasshown, this field has a rich history and bene-fited from the contributions of many organi-zational psychologists.

The chapter then shifted to a discussionof the theoretical foundations of organiza-tional development. Some theories focus onthe general phenomenon of organizationalchange, and others serve as a foundationfor more specific interventions. Althoughtheories can be useful guides in the imple-mentation of organizational-developmentprograms, organizational-development prac-titioners must be careful not to depend tooheavily on theory.

The next section of the chapter focusedon organizational-development interven-tions. Given the number of interventions,this was not intended to be a comprehensivecoverage; rather, the goal was to highlight forreaders the most widely used organizational-development interventions. Typically, inter-

ventions focus on the individual, the group,or broad organizational levels. Of thesethree, interventions at the group level arepresently the most popular, although broadorganizational interventions are being usedmore frequently in recent years. Regardlessof the level at which interventions take place,however, some level of behavior change hasto occur if these interventions are to enhanceorganizational effectiveness.

A number of factors were discussed thatimpact the success of organizational-devel-opment programs. These included top man-agement support, consultant competence,and organizational ownership of the changeprocess. Top management personnel typicallyhave the power to authorize the resourcesneeded for an organizational-change effortto occur, and employees often look to topmanagement in deciding what is and what isnot important. The competence level of theconsultant is a crucial factor because even auseful intervention will not be effective if itis applied incompetently. Finally, if an orga-nization does not take ownership of theorganizational-development process, thensuch efforts are reduced to being passingfads rather than catalysts for change.

Evaluation of organizational-develop-ment programs is challenging, for a numberof reasons. Specifically, it is often difficult tocome up with appropriate criteria; a gooddeal of creativity may be needed to produceappropriate research designs; and it may bedifficult to assess whether the changesobserved are caused by the organizational-development program rather than artifacts.Despite these challenges, considerable eval-uation research has been conducted inthe field of organizational development.Generally speaking, the results of organiza-tional development have been shown tobe positive, although there is considerableorganization-to-organization variation.

Chapter Summary 511


PHILIP MURVIS AND LESSONS FROM STUDYING ORGANIZATIONAL

CHANGE

Two factors shaped my intellectual interest and

emotional energy for organization develop-

ment (OD). My college and doctoral years were

during an era of social change: the civil andwomen’s rights movement, anti-Vietnam war

protests, Watergate, and such. Early on I saw

myself as an activist and believed, like so many,

that knowledge from psychology could be used

to redress wrongs and make things right.

One issue that caught my attention con-

cerned the frustrations of youngerpeople enter-

ing the workforce in the 1970s. They were whatDaniel Yankelovich called the ‘‘new breed’’

worker with higher education levels than their

predecessors and more ‘‘anti-authority’’ atti-

tudes. And they were not well suited to the

routinized industrial and service job designs

of that era nor to the conformist, strictly hier-

archical business cultures of the time. As a

result, the ‘70s was an era of the ‘blue collarblues’ and ‘white collar woes’ and when the

quality of American-made goods began to

decline and then pale versus imports from

Japan.

I had studied OD in college, from the

vantage of scholarly Chris Argyris and also

radical Saul Alinsky. At the University of Mich-

igan, in the org. psych. program, I was involved

in several action research projects with EdLawler to, in my mind, ‘‘save the worker.’’ This

was stimulating because I got the chance to do

OD research in many different kinds of orga-

nizations and to work with lot of different

kinds of people. It was also sobering because

I learned that my ideas on work reform weren’t

always matched by the interests of many work-

ers and managers and that even when they wereit was damnably difficult to effect organization

change.

My first book on OD was called, aptly,

FailuresinOrganizationDevelopmentandChange,

andincludedcasestudiesbymanyleaders in the

field describing their failures and what they

learned from them. Since then, I’ve studied

OD and applied it to corporate mergers andacquisitions, the global transformation of com-

panies, and, most recently, in the advance of

corporate citizenship.

A few things to take away from my life and

work: (1) Get out in the field early and often—

book learning is fine but nothing teaches like

experience, thenreflectingon it, andcomparing

what you find with the field’s rich knowledge-base; (2) Look for the new—find what’s emerg-

ing in organizations, such as ‘‘bottom of the

pyramid’’ markets, new ways of managing em-

ployee diversity, new forms of customer

engagement—and see how OD can engage

those trends; and (3) Go global—travel, con-

nect virtually to students in Asia and else-

where, engage fellow students from foreignlands, go to international conferences and so

on. It’s challenging to discover that what you

know to be true takes different twists and

turns in other lands.

Philip H. Murvis

Bethesda, MD


The final section of the chapter discussedsome of the common challenges in client–consultant relationships. These includebalancing the needs of various clients withinan organization, maintaining confidentialityof information, and deciding when to termi-nate a consulting relationship. As with mostother difficult issues, there are no formulasor easy solutions. In most cases, though,taking the time to address these issues priorto formally beginning the consulting processis the best way to ensure that they do notbecome major problems down the road.


Detert, J. R., Schroeder, R. G., & Mauriel, J. J.(2000). A framework for linking culture and

improvement initiatives in organizations.Academy of Management Review, 25, 850–963.

Dyer, W. (1994). Team building: Currentissues and new alternatives (3rd ed.). Read-ing, MA: Addison-Wesley.

Mirvis, P. H. (2005). Large group interven-tions: Change as theater. The Journal ofApplied Behavioral Science, 41, 122–138.

Ryan, A. M., Chan, D., Ployhart, R. E., &Slade, L. A. (1999). Employee attitude sur-veys in a multinational organization: Consid-ering language and culture in assessingmeasurement equivalence. Personnel Psychol-ogy, 52, 37–58.

Wageman, R., Hackman, J. R., & Lehman,E. (2005). Team Diagnostic Survey: Devel-opment of an instrument. The Journal ofApplied Behavioral Science, 41, 373–398.


References

Abelson, M. A., & Baysinger, B. D. (1984).

Optimal and dysfunctional turnover:

Toward an organizational level model.

Academy of Management Review, 9, 331--

341.

Abramis, D. J. (1994). Work role ambigu-

ity, job satisfaction, and job performance:

Meta-analysis and review. Psychological

Reports, 75, 1411--1433.

Ackerman, P. L. (1989). Within-task corre-

lations of skilled performance: Implications

for predicting individual differences?

(Comment on Henry & Hulin, 1987).

Journal of Applied Psychology, 74, 360--

364.

Adams, G. A., & Beehr, T. A. (1998).

Turnover and retirement: A comparison of

their similarities and differences. Personnel

Psychology, 51, 641--665.

Adams, G. A., King, L. A., & King, D. W.

(1996). Relationship of job and family

involvement, family social support, and

work-family conflict with job and life sat-

isfaction. Journal of Applied Psychology, 81,

411--420.

Adams, J. S. (1965). Inequity in social

exchange. In L. Berkowitz (Ed.), Advances in

Experimental Social Psychology (Vol. 2, pp.

267--299). San Diego, CA: Academic Press.

Adkins, B., & Caldwell, D. (2004). Firm

or subgroup culture: Where does fitting

matter most? Journal of Organizational

Behavior, 25, 969--978.

Adler, S., Skov, R. B., & Salvemini, N. J.

(1985). Job characteristics and job satisfac-

tion: When cause becomes consequence.

Organizational Behavior and Human Decision

Processes, 35, 266--278.

515

Aguinis, H., & Stone-Romero, E. F.

(1997). Methodological artifacts in moder-

ated multiple regression and their effects

on statistical power. Journal of Applied Psy-

chology, 82, 192--206.

Aiello, J. R., & Kolb, K. J. (1995). Elec-

tronic performance monitoring: A risk

factor for workplace stress. In S. L. Sauter

& L. R. Murphy (Eds.) Organizational risk

factors for job stress (pp. 163--179). Wash-

ington, DC: American Psychological Asso-

ciation.

Ajzen, I. (1988). Attitudes, personality, and

behavior. Chicago: Dorsey.

Ajzen, I. (2001). Nature and functions of

attitudes. Annual Review of Psychology, 52,

27--58.

Albanese, R., & Van Fleet, D. D. (1985).

Rational behavior in groups: The free rid-

ing tendency. Academy of Management

Review, 10, 244--255.

Alderfer, C. P. (1969). An empirical test of

a new theory of human needs. Organiza-

tional Behavior and Human Performance, 4,

142--175.

Alderfer, C. P., & Smith, K. K. (1982).

Studying intergroup relations embedded in

organizations. Administrative Science Quar-

terly, 27, 35--65.

Aldred, C. (1994). U.K. ruling focuses

attention to job stress. Business Insurance,

Dec., 5, 55--56.

Aldrich, H. (1999). Organizations evolving.

London: Sage.

Allen, D. G. (2006). Do organizational

socialization tactics influence newcomer

embeddedness and turnover? Journal of

Management, 32(2), 237--256.

Allen, N. J., & Meyer, J. P. (1990). The

measurement and antecedents of affective,

continuance, and normative commitment

to the organization. Journal of Occupational


Allen, N. J., & Meyer, J. P. (1996). Affec-

tive, continuance, and normative commit-

ment to the organization: An examination

of construct validity. Journal of Vocational

Behavior, 49, 252--276.

Allen, R. W., Madison, L. W., Porter,

L. W., Renwick, P. A., & Mayes, B. T.

(1979). Organizational politics: Tactics and

characteristics of its actors. California Man-

agement Review, 12, 77--83.

Allport, G. W. (1954). The nature of preju-

dice. Reading, MA: Addison-Wesley.

Amabile, T. M. (1983). The social psychol-

ogy of creativity: A componential concep-

tualization. Journal of Personality and Social


Amabile, T. M., & Conti, R. (1999).

Changes in the work environment for cre-

ativity during downsizing. Academy of

Management Journal, 42, 630--640.

Ambrose, M. L., & Kulik, C. T. (1999).

Old friends, new faces: Motivation research

in the 1990s. Journal of Management, 25,

142--175, 231--292.

Ambrose, M. L., Seabright, M. A., &

Schminke, M. (2002). Sabotage in the

516 References

workplace: The role of organizational jus-

tice. Organizational Behavior and Human

Decision Processes, 89, 803--812.

Americans with Disabilities Act of 1990,

Publ. L. No. 101--336, 104 Stat. 328 (1990).

Codified at U.S.C.A. § 12101 et seq.

Andersson, L., & Pearson, C. M. (1999).

Tit for tat? The spiraling effect of incivility

in the workplace. Academy of Management

Review, 24, 452--471.

Armenakis, A., Bedeian, A. G., & Pond, S.

(1983). Research issues in OD evaluation:

Past, present, and future. Academy of Man-


Armenakis, A., & Zmud, R. W. (1979).

Interpreting the measurement of change in

organizational research. Personnel Psychol-

ogy, 32, 709--723.

Arvey, R. D., Bhagat, R. S., & Salas, E.

(1991). Cross-cultural and cross-national

issues in personnel and human resources

management: Where do we go from here?

In G. R. Ferris & K. M. Rowland (Eds.),

Research in personnel and human resources

management (Vol. 9, pp. 367--407). Green-

wich, CT: JAI Press.

Arvey, R. D., Bouchard, T. J., Segal, N. L.,

& Abraham, L. M. (1989). Job satisfaction:

Environmental and genetic components.

Journal of Applied Psychology, 74, 187--192.

Arvey, R. D., & Jones, A. P. (1985). The

use of discipline in organizational settings.

In B. M. Staw & L. L. Cummings (Eds.),

Research in organizational behavior (pp.

367--408). Greenwich, CT: JAI Press.

Ashford, S. J., & Black, J. S. (1996). Pro-

activity during organizational entry: The

role of desire for control. Journal of Applied


Ashforth, B. E., & Humphrey, R. H.

(1993). Emotional labor in service roles:

The influence of identity. Academy of Man-


Athey, T. R., & Hautaluoma, J. E. (1988).

Team building in religious organizations.

Organization Development Journal, 62--65.

Atkinson, J. W. (1964). An introduction to

motivation. Princeton, NJ: Van Nostrand.

Aube, C., & Rousseau, V. (2005). Team

goal commitment and team effectiveness:

The role of task interdependence and sup-

portive behaviors. Group Dynamics: Theory,

Research, and Practice, 9, 189--204.

Aust, B., & Ducki, A. (2004). Comprehen-

sive health promotion interventions at the

workplace: Experiences with health circles

in Germany. Journal of Occupational Health


Austin, J. T., Humphreys, L. G., & Hulin,

C. L. (1989). Another view of dynamic

criteria: A critical reanalysis of Barrett,

Caldwell, and Alexander. Personnel Psychol-

ogy, 42, 583--596.

Austin, J. T., & Vancouver, J. B. (1996).

Goal constructs in psychology: Structure,

process, and content. Psychological Bulletin,

120, 338--375.

Averill, J. E. (1973). Personal control over

aversive stimuli and its relationship to

stress. Psychological Bulletin, 80, 286--303.

References 517

Avolio, B. J., Gardner, W. L., Walumbwa,

F. O., Luthans, F., & May, D. R. (2004).

Unlocking the mask: A look at the process

by which authentic leaders impact follower

attitudes and behavior. The Leadership

Quarterly, 15, 801--823.

Aycan, Z., Kanungo, R. N., & Sinha, J. B.

(1999). Organizational culture and human

resource management practices: The model

of culture fit. Journal of Cross-Cultural Psy-

chology, 30, 501--526.

Bachrach, D. G., Bendoly, E., & Podsakoff,

P. M. (2001). Attributions of the ‘‘causes’’ or

group performance as an explanation of the

organizational citizenship behavior/organi-

zational performance relationship. Journal of

Applied Psychology, 86, 1285--1293.

Bachrach, D. G., & Jex, S. M. (2000).

Organizational citizenship and mood: An

experimental test of perceived job breadth.

Journal of Applied Social Psychology, 30,

641--663.

Bagozzi, R. P., & Yi, Y. (1990). Assessing

method variance in multitrait-multimethod

matrices: The case of self-reported affect

and perceptions at work. Journal of Applied


Bales, R.F. (1965). The equilibrium problem

in small groups. In A. P. Hare, E. F.

Bamberger, J. (2006). Current views on

creativity research. Psyccritiques,

Bandura, A. (1986). Social foundations of

thought and action: A social-cognitive theory.

Englewood Cliffs, NJ: Prentice-Hall.

Bandura, A. (1997). Self-efficacy: The exer-

cise of control. New York: Freeman.

Bandura, A. (2001). Social cognitive theory:

An agentic perspective. Annual Review of


Barkema, H. G., & Gomez-Mejia, L. R.

(1998). Managerial compensation and firm

performance: A general research frame-

work. Academy of Management Journal, 41,

135--145.

Barling, J. (1996). The prediction, experi-

ence, and consequences of workplace vio-

lence. In G. R. VandenBos & E. Q.

Bulatad (Eds.), Violence on the job: Identify-

ing risks and developing solutions (pp. 29--

49). Washington, DC: American Psycho-

logical Association.

Barling, J., Dupre, K. E., & Hepburn, C.

G. (1998). Effects of parents’ job insecurity

on children’s work beliefs and attitudes.


Barling, J., & Griffiths, A. (2003). A his-

tory of occupational health psychology. In

J. C. Quick & L. E. Tetrick (Eds.), Hand-

book of occupational health psychology (pp.

19--31). Washington DC: American Psy-

chological Association.

Baron, R. A., & Bell, P. A. (1976). Aggres-

sion and heat: The influence of ambient

temperature, negative affect, and a cooling

drink on physical aggression. Journal of

Personality and Social Psychology, 33, 245--

255.

Baron, R., Handley, R., & Fund, S. (2006).

The impact of emotional intelligence on

518 References

performance. In V. U. Druskat, F. Sala, &

G. Mount (Eds.), Linking emotional intelli-

gence and performance at work: Current

research evidence with individuals and groups

(pp. 3--19). Mahwah, NJ: Lawrence Erl-

baum Associates.

Baron, R. M., & Kenny, D. A. (1986). The

moderator-mediator variable distinction in

social psychological research: Conceptual,

strategic and statistical considerations. Jour-

nal of Personal and Social Psychology, 51,

1173--1182.

Barrett, G. V., Caldwell, M., & Alexander,

R. (1985). The concept of dynamic crite-

ria: A critical reanalysis. Personnel Psychol-

ogy, 38, 41--56.

Barrett, M., Grant, D., & Wailes, N.

(2005). ICT and organizational change:

Introduction to the special issue. Journal of

Applied Behavior Science, 42(1), 6--22.

Barrick, M. R., & Mount, M. K. (1991).

The big five personality dimensions and

job performance: A meta-analysis. Person-

nel Psychology, 44, 1--26.

Barrick, M. R., & Mount, M. K. (2005).

Yes, personality matters: Moving on to more

important matters. Human Performance, 18,

359--372.

Barrick, M. R., Mount, M. K., & Strauss,

J. P. (1993). Conscientiousness and per-

formance of sales representatives: Test of

the mediating effects of goal setting. Jour-

nal of Applied Psychology, 78, 715--722.

Barrick, M. R., Stewart, S. L., Neubert, M.

J., & Mount, M. K. (1998). Relating mem-

ber ability and personality to work-team

processes and team effectiveness. Journal of


Barrick, M. R., Stewart, G. L., & Piotrow-

ski, M. (2002). Personality and job per-

formance: Test of the mediating effects of

motivation among sales representatives.


Bass, B. M. (1998). Transformational leader-

ship: Industrial, military, and educational

impact. Mahwah, NJ: Erlbaum.

Bass, B. M., & Avolio, B. J. (1993). Trans-

formational leadership: A response to cri-

tiques. In M. M. Chemers & R. Ayman

(Eds.), Leadership theory and research: Per-

spectives and directions (pp. 49--80). San

Diego, CA: Academic Press.

Bates, C. A., Bow Bass, B. M., Avolio, B. J.,

Jung, D. I., & Berson, Y. (2003). Predict-

ing unit performance by assessing transfor-

mational and transactional leadership.


218.

Bauer, T. N., & Green, S. G. (1998). Test-

ing the combined effects of newcomer

information seeking and manager behavior

on socialization. Journal of Applied Psychol-

ogy, 83, 72--83.

Baugh, S. G., & Roberts, R. M. (1994). Pro-

fessional and organizational commitment

among professional engineers: Conflicting

or complementing? IEEE Transactions on

Engineering Management, 41, 108--114.

Baumeister, R. F., & Leary, M. R. (1995).

The need to belong: Desire for interpersonal

References 519

attachments as a fundamental human

motivation. Psychological Bulletin, 117, 497--

529.

Beach, L. R. (1993). Making the right deci-

sion. Englewood Cliffs, NJ: Prentice-Hall.

Beard, K. M., & Edwards, J. R. (1995).

Employees at risk: Contingent work and

the psychological experience of contingent

workers. In C. L. Cooper & D. M. Rous-

seau (Eds.), Trends in organizational behav-

ior (Vol. 2., pp. 109--126). Chichester,

England: Wiley.

Becker, H. S. (1960). A note on the con-

cept of commitment. American Journal of

Sociology, 66, 32--42.

Becker, T. E. (1992). Foci and bases of

commitment: Are they distinctions worth

making? Academy of Management Journal,

35, 232--244.

Beckhard, R. (1967). The confrontation

meeting. Harvard Business Review, 45, 149--

153.

Beehr, T. A. (1986). The process of retire-

ment: A review and recommendations for

future investigation. Personnel Psychology,

39, 31--55.

Beehr, T. A. (1995). Psychological stress in

the workplace. London: Routledge.

Beehr, T. A. (1998). Research on occupa-

tional stress: An unfinished enterprise. Per-

sonnel Psychology, 51, 835--844.

Beehr, T. A., & Franz, T. M. (1987). The

current debate about the meaning of job

stress. In J. M. Ivancevich & D. C. Ganster

(Eds.), Job stress: From theory to suggestion

(pp. 5--18). New York: Haworth Press.

Beehr, T. A., Jex, S. M., & Ghosh, P.

(2001). The management of occupational

stress. In C. M. Johnson, W. K. Redmon,

& T. C. Mawhinney (Eds.), Handbook of

organizational performance: Behavior analysis

and management (pp. 225--254). New

York: Haworth Press.

Beehr T. A., Jex, S. M., Stacy, B. A., &

Murray, M. A. (2000). Work stress and

co-worker support as predictors of individ-

ual strains and performance. Journal of

Organizational Behavior, 21, 391--405.

Beehr, T. A., & Newman, J. E. (1978). Job

stress, employee health, and organizational

effectiveness: A facet analysis, model, and

literature review. Personnel Psychology, 31,

665--699.

Beehr, T. A., Walsh, J. T., & Taber, T. D.

(1980). Relationship of stress to individu-

ally and organizationally valued states:

Higher order needs as a moderator. Journal

of Applied Psychology, 61, 35--40.

Beer, M. (1976). The technology of organiza-

tion development. In M. D. Dunnette (Ed.),

Handbook of industrial and organizational psy-

chology (pp. 937--993). Chicago: Rand

McNally.

Bellarosa, C., & Chen, P. Y. (1997). The

effectiveness and practicality of stress man-

agement interventions. Journal of Occupa-

tional Health Psychology, 2, 247--262.

520 References

Bem, D. J. (1972). Self-perception theory.

In L. Berkowitz (Ed.), Advances in experi-

mental social psychology (Vol. 6, pp. 1--62).

New York: Academic Press.

Bemmels, B., & Foley, J. R. (1996). Griev-

ance procedure research: A review and

theoretical recommendations. Journal of

Management, 22, 359--384.

Bennett, J. B., & Lehman, W. E. K. (1998).

Workplace drinking climate, stress, and

problem indicators: Assessing the influence

of teamwork (group cohesion). Journal of

Studies on Alcohol, 59, 608--618.

Bennett, R. J., & Robinson, S. L. (2000).

Development of a measure of workplace

deviance. Journal of Applied Psychology,

85(3), 349--360.

Bentler, P. M. (1990). Comparative fit

indices in structural models. Psychological

Bulletin, 107, 238--246.

Berdahl, J. L., & Moore, C. (2006). Work-

place harassment: Double jeopardy for

minority women. Journal of Applied Psychol-

ogy, 91, 426--436.

Berkowitz, L., & Donnerstein, E. (1982).

External validity is more than skin deep.

American Psychologist, 37, 245--257.

Bernichon, T., Cook, K. E., & Brown, J. D.

(2003). Seeking self-evaluative feedback:

The interactive role of global self-esteem

and specific self-views. Journal of Personality

and Social Psychology, 84, 194--204.

Bertua, C., Anderson, N., & Salgado, J. F.

(2005). The predictive validity of cognitive

ability tests: A UK meta-analysis. Journal of

Organizational and Occupational Psychology,

78, 387--409.

Bettencourt, L. A., Gwinner, K. P., &

Meuter, M. L. (2001). A comparison of

attitude, personality, and knowledge pre-

dictors of service-oriented organizational

citizenship behaviors. Journal of Applied


Bialek, H., Zapf, D., & McGuire, W.

(1977). Personnel turbulence and time uti-

lization in an infantry division (Hum RRO

FR-WD-CA 77--11). Alexandria, VA:

Human Resources Research Organization.

Blake, R. R., & Mouton, J. (1964). The

managerial grid. Houston, TX: Gulf.

Blake, R. R., Shephard, H. A., & Mouton,

J. S. (1964). Managing intergroup conflict in

industry. Houston, TX: Gulf.

Blakely, G. L., Andrews, M. C., & Moor-

man, R. H. (2005). The moderating effects

of equity sensitivity on the relationship

between organizational justice and organi-

zational citizenship behaviors. Journal of

Business and Psychology, 20, 259--273.

Blau, G., & Andersson, L. (2005). Testing

a measure of instigated workplace incivil-

ity. Journal of Occupational and Organiza-

tional Psychology, 78(4), 595--614.

Bliese, P. D. (2000). Within-group agree-

ment, non-independence, and reliability:

Implications for data aggregation and anal-

ysis. In K. J. Klien & S. W. Kozlowski

(Eds.), Multilevel theory, research, and methods

References 521

in organizations (pp. 349--381). San Fran-

sisco: Jossey-Bass.

Bliese, P. D., & Britt, T. W. (2001). Social

support, group consensus, and stressor-strain

relationships: Social context matters. Journal

of Organizational Behavior, 22, 425--436.

Bliese, P. D., & Jex, S. M. (1999). Incorpo-

rating multiple levels of analysis into occu-

pational stress research. Work and Stress,

13, 1--6.

Bliese, P. D., & Jex, S. M. (2002). Incor-

porating a mulitilevel perspective into

occupational stress research: Theoretical,

methodological, and practical implications.

Journal of Occupational Health Psychology,

7(3), 265--276.

Bobko, P. (1995). Editorial. Journal of


Bolino, M. C. (1999). Citizenship and

impression management: Good soldiers or

good actors? Academy of Management Review,

24, 82--98.

Bolino, M. C., & Turnley, W. H. (1999).

Measuring impression management in orga-

nizations: A scale development based on the

Jones and Pittman taxonomy. Organizational

Research Methods, 2, 187--206.

Bond, J. T., Galinsky, E., & Swanberg, J.

(1998). The 1997 National Study of The

Changing Workforce. New York: Families

and Work Institute.

Bono, J. E., & Judge, T. A. (2003). Core

self-evaluations: A review of the trait and

its role in job satisfaction and job perfor-

mance. European Journal of Personality, 17,

S5--S18.

Bono, J. E., & Judge, T. A. (2004). Person-

ality and transformational and transac-

tional leadership: A meta-analysis. Journal


Boode, G. (2005). Boundaries on the

move: The impact of cultural values and

language on organizational design and

communication within an organization in

Thailand. Asian Pacific Business Review, 11,

519--533.

Borman, W. C. (1991). Job behavior, per-

formance, and effectiveness. In M. D. Dun-

nette & L. M. Hough (Eds.), Handbook of

industrial and organizational psychology (2nd

ed., Vol. 2, pp. 271--398). Palo Alto, CA:

Consulting Psychologists Press.

Borman, W. C., & Motowidlo, S. J. (1993).

Expanding the criterion domain to include

elements of contextual performance. In N.

Schmitt & W. C. Borman (Eds.), Personnel

selection in organizations (pp. 71--98). San

Francisco: Jossey-Bass.

Bouchard, T. D., Jr. (1976). Field research

methods: Interviewing, questionnaires,

participant observation, systematic obser-

vation, unobtrusive measures. In M. D.

Dunnette (Ed.), Handbook of industrial and

organizational psychology (pp. 363--413).

Chicago: Rand McNally.

Bourrier, M. (2005). The contribution of

organizational design to safety. European


522 References

Bowen, D. E., Ledford, G. E., & Nathan,

B. R. (1991). Hiring for the organization,

not the job. Executive, 4, 35--51.

Bowers, D. G. (1973). OD techniques and

their results in 23 organizations: The

Michigan ICL study. Journal of Applied

Behavioral Science, 9, 21--43.

Bowling, N. A., & Beehr, T. A. (2006).

Workplace harassment from the victim’s

perspective: A theoretical model and meta-

analysis. Journal of Applied Psychology,

91(8), 998--1012.

Braddy, P. W., Meade, A. W., & Krousta-

lis, C. M. (2006). Organizational recruit-

ment website effects on viewers’

perceptions of organizational culture. Jour-

nal of Business and Psychology, 20(4), 525--

543.

Bramel, D., & Friend, R. (1981). Haw-

thorne, the myth of the docile worker, and

class bias in psychology. American Psychol-

ogist, 36, 867--878.

Brannick, M. T., Roach, R. M., & Salas, E.

(1993). Understanding team performance:

A multimethod study. Human Performance,

6(4), 287--308.

Brannick, M. T., & Spector, P. E. (1990).

Estimation problems in the block diago-

nal model of the multitrait-multimethod

matrix. Applied Psychological Measurement.

14, 325--339.

Breaugh, J. A., & Colihan, J. P. (1994).

Measuring facets of job ambiguity: Con-

struct validity evidence. Journal of Applied


Brehm, J. W. (1966). A theory of psycholog-

ical reactance. Oxford, England: Academic

Press.

Brett, J. M., & Rognes, J. K. (1986). Inter-

group relations in organizations. In P. S.

Goodman (Ed.), Designing effective work

groups (pp. 202--236). San Francisco: Jos-

sey-Bass.

Bridges, W. (1994). JobShift: How to prosper

in a workplace without jobs. Reading, MA:

Addison-Wesley.

Britt, T. W., Adler, A. B., & Bartone, P. T.

(2001). Deriving benefits from stressful

events: The role of engagement in mean-

ingful work and hardiness. Journal of Occu-

pational Health Psychology, 6(1), 53--63.

Britt, T. W., Castro, C. A., & Adler, A. B.

(2005). Self-engagement, stressors, and

health: A longitudinal study. Personality and

Social Psychology Bulletin, 31(11), 1475--

1486.

Britt, T. W., Davison, J. M., Bliese, P. D.,

& Castro, C. A. (2004). How leaders can

influence the health consequences of stres-

sors. Military Medicine, 169, 541--545.

Britt, T. W., Dickinson, J. M., Moore,

D. M., Castro, C. A., & Adler, A. B.

(2007). Correlates and consequences of

morale versus depression under stressful

conditions. Journal of Occupational Health


Britt, T. W., Thomas, J. L., & Dawson,

C. R. (2006). Self-engagement magnifies the

relationship between qualitative overload

References 523

and performance in a training setting. Jour-

nal of Applied Social Psychology, 36(9),

2100--2114.

Broadbent, D. E. (1985). The clinical

impact of job design. British Journal of Clin-

ical Psychology, 24, 33--44.

Brockner, J., Grover, S., Reed, T. F., &

DeWitt, R. L. (1992). Layoffs, job insecur-

ity, and survivors’ work effort: Evidence of

an inverted-U relationship. Academy of


Brockner, J., Grover, S., Reed, T. F., DeW-

itt, R. L., O’Malley, M. (1987). Survivors’

reactions to layoffs: We get by with a little

help from our friends. Administrative Sci-

ence Quarterly, 32, 526--542.

Brotheridge, C., & Grandey, A. (2002).

Emotional labor and burnout: Comparing

two perspectives of ‘‘people work.’’ Journal

of Vocational Behavior, 60, 17--39.

Brown, R., Condor, S., Matthews, A.,

Wade, G., & Williams, J. A. (1986).

Explaining intergroup differentiation in

industrial organization. Journal of Occupa-

tional Psychology, 59, 273--286.

Brown, R., Maras, P., Massar, B., Vivian,

J., & Hewstone, M. (2001). Life on the

ocean wave: Testing some intergroup

hypotheses in a naturalistic setting. Group

Processes and Intergroup Relations, 4, 81--

97.

Bryant, H. (2005). Juicing the game: Drugs,

power, and the fight for the soul of major

league baseball. New York: Viking.

Buch, K., & Aldrich, J. (1991). O. D.

under conditions of organizational decline.

Organizational, Development Journal, 9, 1--5.

Buckley, M. R., Fedor, D. B., Veres, J. G.,

Weise, D. S., & Carraher, S. M. (1998).

Investigating newcomer expectations and

job-related outcomes. Journal of Applied


Buell, P., & Breslow, L. (1960). Mortality

from coronary heart disease in California

men who work long hours. Journal of

Chronic Diseases, 11, 615--626.

Buller, P. F. (1988). For successful strategic

organizational change: Blend OD practices

with strategic management. Organizational

Dynamics, 16, 42--55.

Bullock, R. J., & Svyantek, D. J. (1983).

Positive-findings bias in positive-findings

bias research: An unsuccessful replication.

Academy of Management Proceedings, 221--

224.

Buono, A. F., & Bowditch, J. L. (1989).

The human side of mergers and acquisitions:

Managing collision between people, cultures,

and organizations. San Francisco: Jossey-

Bass.

Burke, W. W. (1994). Organization devel-

opment (2nd ed.). Reading, MA: Addison-

Wesley.

Burns, T., & Stalker, G. M. (1961). The

management of innovation. London: Tavi-

stock.

Buss, D. M. (1996). The evolutionary psy-

chology of human social strategies. In E.

524 References

T. Higgins & A. W. Kruglanski (Eds.),

Social psychology: Handbook of basic princi-

ples (pp. 3--38). New York: Guilford.

Bussing, A., & Hoge, T. (2004). Aggres-

sion and violence against home care

workers. Journal of Occupational Health

Psychology, 9(3), 206--219.

Byrk, A., & Raudenbush, S. (1992). Hier-

archical linear models: Applications and

data analysis methods. Newbury Park, CA:

Sage.

Byrne, D. (1971). The attraction paradigm.


Byrne, J. E. (1988). Caught in the middle.

Business Week, Sept., 12, 80--88.

Cable, D. M., & Parsons, C. K. (2001).

Socialization tactics and person-organiza-

tion fit. Personnel Psychology, 54(1), 1--23.

Cable, D. M. & Turban, D. (2001).

Recruitment Image Equity: Establishing the

dimensions, sources and value of job

seekers’ organizational beliefs. In G. Ferris

(Ed.), Research in personnel/human resource

management (Vol. 20), pp. 115--163

Greenwich, CT: JAI Press.

Cable, D. M., & Yu, K. Y. T. (2006). Man-

aging job-seekers organizational image

beliefs: The role of media richness and

media credibility. Journal of Applied Psy-

chology, 91(4), 828--840.

Caldwell, D. E., & O-Reilly, C. A. (2003).

The determinants of team-based innovation

in organizations: The role of social influ-

ence. Small Group Research, 34, 497--517.

Campbell, D. T. (1958). Common fate,

similarity, and other indices of the status

of aggregates of persons as social entities.


Campbell, D. T., (1999). Innovative strat-

egies to thrive in today’s health care

environment. ADVANCE for Speech-

Language Pathologists and Audiologists, 9,

9--10.

Campbell, D. T., & Fiske, D. W. (1959).

Convergent and discriminant validation by

the multi-trait-multimethod matrix. Psycho-

logical Bulletin, 56, 81--105.

Campbell, J. P. (1990). Modeling the per-

formance prediction problem in industrial

and organizational psychology. In M. D.

Dunnette & L. M. Hough (Eds.), Handbook

of industrial and organizational psychology

(2nd ed., Vol. 1, pp. 687--732). Palo Alto,

CA: Consulting Psychologists Press.

Campbell, J. P. (1994). Alternative models

of job performance and their implications

for selection and classification. In M. G.

Rumsey, C. B. Walker, & J. H. Harris

(Eds.), Personnel selection and classification

(pp. 33--51). Hillsdale, NJ: Erlbaum.

Campbell, J. P., & Dunnette, M. D. (1968).

Effectiveness of t-group experiences in man-

agerial training. Psychological Bulletin, 70,

73--104.

Campion, M. A. (1993). Editorial. Article

review checklist: A criterion checklist for

reviewing research articles in applied psy-

chology. Personnel Psychology, 46, 705--

718.

References 525

Campion, M. A. (1996). A message from

your president: How to publish the results

of applied projects (or how two needles

can find each other in a haystack).

Industrial-Organizational Psychologist, 33,

9--10.

Campion, M. A., & Berger, C. J. (1990).

Conceptual integration and empirical test

of job design and compensation relation-

ships. Personnel Psychology, 43, 525--554.

Campion, M. A., Cheraskin, L., & Stevens,

M. J. (1994). Career-related antecedents

and outcomes of job rotation. Academy of


Campion, M. A., & McClelland, C. L.

(1991). Interdisciplinary examination of

the costs and benefits of enlarged jobs: A

job design quasi-experiment. Journal of


Campion, M. A., & McClelland, C. L.

(1993). Follow-up and extension of the

interdisciplinary costs and benefits of

enlarged design. Journal of Applied Psychol-

ogy, 70, 29--43.

Campion, M. A., Medsker, G. J., & Higgs,

A. C. (1993). Relations between work

group characteristics and effectiveness:

Implications for designing effective work

groups. Personnel Psychology, 46, 823--

850.

Campion, M. A., Mumford, T. V., Morge-

son, F. P., & Nahrhang, J. D. (2005).

Work redesign: Eight obstacles and oppor-

tunities. Human Resource Management, 44,

367--390.

Campion, M. A., Papper, E. M., &

Medsker, G. J. (1995). Relations between

work team characteristics and effective-

ness: Replication and extension. Personnel


Campion, M. A., & Thayer, P. W. (1985).

Development and field evaluation of an

interdisciplinary measure of job design.


Cannon, W. B. (1914). The interrelations

of emotions as suggested by recent physio-

logical researches. American Journal of Psy-

chology, 25, 256--282.

Caplan, R. D. (1987). Person-environment

fit in organizations: Theories, facts, and

values. In A. W. Riley & S. J. Zaccaro

(Eds.), Occupational stress and organizational

effectiveness (pp. 103--140). New York:

Praeger Press.

Caplan, R. D., Cobb, S., French, J. R. P.,

Jr., Harrison, R. V., & Pinneau, S. R.

(1975). Job demands and worker health:

Main effects and occupational differences.

Washington, DC: U.S. Government Print-

ing Office.

Caplan, R. D., & Jones, K. W. (1975).

Effects of workload, role ambiguity, and

Type A personality on anxiety, depression,

and heart rate. Journal of Applied Psychol-

ogy, 60, 713--719.

Carey, A. (1967). The Hawthorne studies:

A radical criticism. American Sociological

Review, June, 403--417.

Carey, J. M., & Kacmar, C. J. (1997). The

impact of communication mode and task

526 References

complexity on small group performance

and member satisfaction. Computers in

Human Behavior, 13, 23--49.

Carroll, S. J., & Tosi, H. L. (1973). Man-

agement by objectives. New York: Macmil-

lan.

Carsten, J. M., & Spector, P. E. (1987).

Unemployment, job satisfaction, and

employee turnover: A meta-analytic test of

the Muchinsky model. Journal of Applied


Cartwright, S., & Cooper, C. L. (1997).

Managing workplace stress. Thousand Oaks,

CA: Sage.

Carver, C. S., & Scheier, M. F. (1981).

Attention and self-regulation: A control theory

approach to human behavior. New York:

Springer.

Cascio, W. F. (1998). Applied psychology in

human resource management (5th ed.).

Upper Saddle River, NJ: Prentice-Hall.

Casper, W. J., Fox, K. E., Sitzmann, T. M.,

& Landy, A. L. (2004). Supervisor refer-

rals to work-family programs. Journal of

Occupational Health Psychology, 9(2), 136--

151.

Castano, E., Yzerbyt, V., Paladino, M. P.,

& Sacchi, S. (2002). I belong, therefore, I

exist: Ingroup identification, ingroup enti-

tativity, and ingroup bias. Personality and

Social Psychology Bulletin, 28, 135--143.

Chadwick-Jones, J., Nicholson, N., &

Brown, C. (1982). Social psychology of

absenteeism. New York: Praeger.

Champoux, J. E. (1991). A multivariate

test of the job characteristics theory of

work behavior. Journal of Occupational

Behavior, 12, 431--446.

Chan, D. (1998). Functional relations

among constructs in the same content

domain at different levels of analysis: A

typology of composition models. Journal of


Chao, G. T., O’Leary-Kelly, A. M., Wolf,

S., Klein, H. J., & Gardner, P. D. (1994).

Organizational socialization: Its content

and consequences. Journal of Applied Psy-

chology, 79, 730--749.

Chapman, D. S., & Webster, J. (2003).

The use of technologies in recruiting,

screening, and selection process for job

candidates. International Journal of Selection

and Assessment, 11, 113--120.

Chatman, J. (1991). Matching people and

organizations: Selection and socialization

in public accounting firms. Administrative

Science Quarterly, 36, 459--484.

Chattopadhyay, P. (1998). Beyond direct

and symmetric effects: The influence of

demographic dissimilarity on organiza-

tional citizenship behavior. Academy of


Chemers, M. M. (1983). Leadership theory

and research: A systems/process integra-

tion. In R. R. Paulus (Ed.), Group process

(pp. 9--39). New York: Springer-Verlag.

Chemers, M. M., Hays, R. B., Rhodewalt,

F., & Wysocki, J. (1985). A person-

References 527

environment analysis of job stress: A con-

tingency model explanation. Journal of Per-

sonality and Social Psychology, 49, 628--635.

Chen, G. (2005). Newcomer adaptation in

teams: Multilevel antecedents and out-

comes. Academy of Management Journal, 48,

101--116.

Chen, P. Y., & Spector, P. E. (1992). Rela-

tionships of work stressors with aggres-

sion, withdrawal, theft, and substance use:

An exploratory study. Journal of Occupa-

tional and Organizational Psychology, 65,

177--184.

Cialdini, R. B. (2001). Influence: Science

and practice. Needham Heights, MA: Allyn

& Bacon.

Clarke, S. (2006). Workplace harassment

from the victim’s perspective: A theoretical

model and meta-analysis. Journal of Occu-

pational Health Psychology, 11, 315--327.

Cleyman, K. L., Jex, S. M., & Love, K. G.

(1995). Employee grievances: An applica-

tion of the leader-member exchange

model. International Journal of Organiza-

tional Analysis, 3, 156--174.

Cober, R. T., Brown, D. J., Levy, P. E.,

Cober, A. B., & Keeping, L. M. (2003).

Organizational web sites: Web site content

and style as determinants of organizational

attraction. International Journal of Selection

and Assessment, 11(2--3), 158--169.

Coch, L., & French, J. R. P., Jr. (1948).

Overcoming resistance to change. Human

Relations, 1, 512--532.

Cohen, A. (1974). Two dimensional man:

An essay on the anthropology of power and

symbolism in complex society. London: Rout-

ledge and Kegan Paul.

Cohen, J. (1992). A power primer. Psycho-

logical Bulletin, 112, 155--159.

Cohen, J., & Cohen, P. (1983). Applied

multiple regression/correlation for the behavior-

al sciences. Hillsdale, NJ: Erlbaum.

Cohen, S., & Wills, T. A. (1985). Stress,

social support, and the buffering hypothe-

sis. Psychological Bulletin, 98, 310--357.

Cohen, S. G., & Bailey, D. E. (1997). What

makes teams work: Groups effectiveness

research from the shop floor to the executive

suite. Journal of Management, 20, 239--290.

Cohen-Charash, Y., & Spector, P. E. (2000).

The role of justice in organizations: A meta-

analysis. Organizational Behavior and Human

Decision Processes, 86, 287--321.

Colarelli, S. M., Dean, R. A., & Konstans,

C. (1987). Comparative effects of personal

and situational influences on job outcomes

of new professionals. Journal of Applied


Collins, J. M., & Schmidt, F. L. (1993).

Personality, integrity, and white collar

crime: A construct validity study. Personnel

Psychology, 46(2), 295--311.

Colquitt, J. A., Conlon, D. E., Wesson, M.

J., Porter, C., & Ng, K. Y. (2001). Justice

at the millennium: A meta-analysis of 25

years of organizational justice research.


528 References

Connolly, J. J., & Viswesvaran, C. (2000).

The role of affectivity in job satisfaction: A

meta-analysis. Personality and Individual

Differences, 29, 265--281.

Conway, J. M. (1999). Distinguishing con-

textual performance from task performance

for managerial jobs. Journal of Applied Psy-

chology, 84, 3--13.

Cook, T. D., & Campbell, D. T. (1979).

Quasi-experimentation: Design and analysis

issues for field settings. Chicago: Rand

McNally.

Cook, T. D., Campbell, D. T., & Perrachio,

L. (1990). Quasi experimentation. In M. D.





Cooper, C. L., & Dewe, P. (2004). Stress:

A brief history. Malden, MA: Blackwell

Publishing.

Cooper, C. L., Dewe, P., & O’Driscoll, M.

(2003). Employee assistance programs. In

J. C. Quick & L. E. Tetrick (Eds.), Hand-

book of occupational health psychology (pp.

289--304). Washington, DC: American

Psychological Association.

Cooper, M. L., Russell, M., & Frone, M.

R. (1990). Work stress and alcohol

effects: A test of stress-induced drinking.

Journal of Health and Social Behavior, 31,

260--276.

Cooper-Hakim, A., & Viswesvaran, C.

(2005). The construct of work commit-

ment: Testing an integrative framework.

Psychological Bulletin, 131, 241--259.

Cotton, J. L. (1995). Participation’s effect

of performance and satisfaction: A recon-

sideration of Wagner. Academy of Manage-

ment Review, 20, 276--278.

Covin, T., & Kilman, R. (1991). Profiling

large scale change efforts. Organizational

Development Journal, 9, 1--8.

Cox, M. H., Shephard, R. J., & Corey, P.

(1981). Influence of employee fitness, pro-

ductivity and absenteeism. Ergonomics, 24,

795--806.

Crampton, S. M., & Wagner, J. A., III.

(1994). Percept-percept inflation in micro

organizational research: An investigation of

prevalence and effect. Journal of Applied


Cronbach, L. J., & Furby, L. (1970). How

should we measure ‘‘change’’---or should

we? Psychological Bulletin, 74, 68--80.

Cropanzano, R., & James, K. (1990). Some

methodological considerations for the

behavioral genetic analysis of work atti-

tudes. Journal of Applied Psychology, 75,

433--439.

Crowne, D. P., & Marlowe, D. (1964). The

approval motive: Studies in evaluative depend-

ency. New York: Wiley.

Crystal, G. S. (1991). In search of excess.

New York: Norton.

Crystal, G. S. (1995). Turner’s compensation

stirs executive pay criticism. CompFlash,

Dec 3.

References 529

Cunningham, C. J. L. (2005). New applicant

decision making: Understanding the influence of

salary, family-friendly/life-friendly policies, and

culture as influential organizational attributes.

Unpublished master’s thesis, Bowling Green

State University, Bowling Green, OH.

Daily, C. M., Johnson, J. L., Ellstrand,

A. E., & Dalton, D. R. (1998). Compensa-

tion committee composition as a determi-

nant of CEO compensation. Academy of


Dahlsten, F., Styhre, A., & Willander, M.

(2005). The unintended consequences of

management by objectives: The volume

growth target at Volvo Cars. Leadership &

Organization Development Journal, 26(7),

529--541.

Damanpour, F. (1991). Organizational

innovation: A meta-analysis of effects of

determinants and moderators. Academy of


Dane, F. C. (1990). Research methods.

Pacific Grove, CA: Brooks-Cole.

Dansereau, F., Alutto, J. A., & Yammarino,

F. J. (1984). Theory testing in organizational

behavior: The variant approach. Englewood

Cliffs, NJ: Prentice-Hall.

Dansereau, F., Graen, G., & Haga, W.

(1975). A vertical dyad approach to leader-

ship within formal organizations. Organiza-


46--78.

Davidson, E. (2006). A technology frames

perspective on information technology and

organizational change. The Journal of

Applied Behavioral Science, 42, 23--39.

Davis, S. M., & Lawrence, P. R. (1977).

Matrix. Reading, MA: Addison-Wesley.

Davis-Blake, A., & Pfeffer, J. (1989). Just a

mirage: The search for dispositional effects

in organizational research. Academy of

Management Review, 14, 385--400.

Dawis, R. V. (1990). Vocational interests,

values, and preferences. In M. D. Dunnette

& L. M. Hough (Eds.), Handbook of indus-

trial and organizational psychology (2nd ed.,

Vol. 1, pp. 833--871). Palo Alto, CA: Con-

sulting Psychologists Press.

Day, A. L., & Catano, V. M. (2006). Screen-

ing and selecting out violent employees. In

K. E. Kelloway, J. Barling, & J. J. Hurrell

(Eds.), Hankbook of workplace violence (pp.

549--577). St. Mary’s University: Sage Pub-

lications, Inc.

Day, D. V., & Bedeian, A. G. (1991). Pre-

dicting job performance across organiza-

tions: The interaction of work orientation

and psychological climate. Journal of Man-

agement, 17, 589--600.

Day, D.V., Sin, H-P., & Chen, T.T.

(2004). Assessing the burdens of leader-

ship: Effects of formal leadership roles on

individual performance over time. Person-


Deadrick, D. L., & Madigan, R. (1990).

Dynamic criteria revisited: A longitudi-

nal study of performance stability and

530 References

predictive validity. Personnel Psychology, 43,

717--744.

DeChurch, L. A., & Marks, M. A. (2006).

Leadership in multi-system teams. Journal


Deci, E. L. (1975). Intrinsic motivation.

New York: Plenum Press.

Deci, E. L., & Ryan, R. M. (1985). Intrinsic

motivation and self-determination in human

behavior. New York: Plenum.

DeFrank, R. S. & Ivancevich, J. M. (1998).

Stress on the job: An executive update.

Academy of Management Update, 12(3), 55--

66.

De Hoogh, A. H. B., Den Hartog, D. N., &

Koopman, P. L. (2005). Linking the Big

Five-Factors of personality to charismatic

and transactional leadership; perceived

work environment as a moderator. Journal

of Organizational Psychology, 26, 839--865.

de Laat, P. B. (1994). Matrix management

of projects and power struggles: A case

study of an R&D laboratory. Human Rela-

tions, 47, 1089--1119.

De La Rose, G. M. (2006). Towards an

understanding of individual ratings of cohesion

within work units: A multilevel study.

Unpublished master’s thesis. Bowling

Green State University, Bowling Green,

OH.

DeMatteo, J. S., Eby, L. T., & Sundstrom,

E. (1998). Team-based rewards: Current

empirical evidence and directions for

future research. Research in Organizational

Behavior, 20, 141--183.

Denison, D. R. (1984). Bringing corporate

culture to the bottom line. Organizational

Dynamics, 13, 5--22.

Denison, D. R. (1990). Corporate culture

and organizational effectiveness. New York:

Wiley.

Denison, D. R., Cho, H. J., & Young, J.

(2000). Diagnosing organizational culture:

Validating a model and method. Working

paper, International Institute for Manage-

ment Development, Lausanne, Switzerland.

Denison, D. R., Haaland, S., & Goelzer, P.

(2004). Corporate culture and organiza-

tional effectiveness: Is Asia different from

the rest of the world? Organizational

Dynamics, 33(1), 98--109.

Denison, D. R., & Mishra, A. K. (1995).

Toward a theory of organizational culture

and effectiveness. Organization Science, 6,

204--223.

Denison, D. R., & Sutton, R. I. (1990).

Operating room nurses. In J. R. Hackman

(Ed.), Groups that work (and those that don’t)

(pp. 293--308). San Francisco: Jossey-Bass.

Dentler, R. A., & Erikson, K. T. (1959).

The functions of deviance in groups. Social

Problems, 7, 98--107.

Depret, E. F., & Fiske, S. T. (1999). Per-

ceiving the powerful: Intriguing individu-

als versus threatening groups. Journal of

Experimental Social Psychology, 35, 461--480.

References 531

DeShon, R. P., Kozlowski, S. W. J.,

Schmidt, A. M., Milner, K. R., & Wiech-

mann, D. (2004). A multiple-goal, multi-

level model of feedback on the regulation

of individual and team performance in

training. Journal of Applied Psychology, 89,

1035--1056.

Devine, D. J., & Phillips, J. L. (2000,

April). Do smarter teams do better? A meta-

analysis of team-level cognitive ability and

team performance. Paper presented at the

15th annual conference of the Society for

Industrial-Organizational Psychology, New

Orleans, LA.

Dickinson, T. L. & McIntyre, R. M. (1997).

A conceptual framework for teamwork

measurement. In M. T. Brannick, E. Salas,

& C. Prince (Eds.), Team performance assess-

ment and measurement (pp. 19--43). Mah-

wah, NJ: Lawrence Erlbaum Associates.

Dickter, D. N., Roznowski, M., & Harri-

son, D. A. (1996). Temporal tempering:

An event history analysis of the process of

voluntary turnover. Journal of Applied Psy-

chology, 81, 705--716.

Diefendorff, J. M., Brown, D. J., & Kamin,

A. M. (2002). Examining the roles of job

involvement and work centrality in pre-

dicting organizational citizenship behaviors

and job performance. Journal of Organiza-

tional Behavior, 23, 93--108.

Dillman, D. A. (2000). Mail and Internet

surveys: The tailored design method (2nd

ed.). New York: Wiley.

Dipboye, R. L., & Flanagan, M. F. (1979).

Research settings in industrial and organi-

zational psychology: Are findings in the

field more generalizable than in the labora-

tory? American Psychologist, 34, 141--151.

Driskell, J. E., Goodwin, G. F., Salas, E.,

& O’Shea, P. G. (2006). What makes a

good team player? Personality and team

effectiveness. Group Dynamics: Theory,

Research, and Practice, 10, 249--271.

Driskell, J. E., & Salas, E. (1991). Group

decision making under stress. Journal of


Driskell, J. E., Salas, E., & Johnston, J.

(1999). Does stress lead to a loss of team

perspective? Group Dynamics: Theory,

Research, and Practice, 3(4), 291--302.

Driskell, J. E., Salas, E., & Johnston, J.

(2001). Stress management: Individual and

team training. In E. Salas, C. A. Bowers, &

E. Edens (Eds.), Improving teamwork in

organizations: Applications of resource man-

agement training (pp. 55--72). Mahwah, NJ:

Erlbaum.

Driskell, J. E., Salas, E., & Johnston, J. H.

(2006). Decision making and performance

under stress. In T. W. Britt, C. A. Castro,

& A. B. Adler (Eds.), Military life: The psy-

chology of serving in peace and combat (pp.

128--154). Westport, CT: Praeger Security

International.

DuBrin, A. J. (1993, Fall). Deadly political

sins. National Business Employment Weekly,

11--13.

532 References

Dubrovsky, V. J., Keisler, S., & Sethna, B.

N. (1991). The equalization phenomenon:

Status effects in computer-mediated and

face-to-face decision making groups.

Human-Computer Interaction, 6, 119--146.

Dudley, N. M., Orvis, K. A., Lebiecki,

J. E., & Cortina, J. M. (2006). A meta-

analytic investigation of conscientiousness

in the prediction of job performance:

Examining the intercorrelations and the

incremental validity of narrow traits. Jour-


Durkheim, E. (1951). Suicide (J. A. Spauld-

ing & G. Simpson, Trans.). Glencoe, IL:

Free Press.

Dyer, W. G. (1987). Team building: Issues

and alternatives (2nd ed.). Reading, MA:

Addison-Wesley.

Eagly, A. H., & Chaiken, S. (1993). The

psychology of attitudes. Fort Worth, TX:

Harcourt, Brace, & Jovanovich.

Eastman, K. K. (1994). In the eyes of the

beholder: An attributional approach to

ingratiation and organizational citizenship

behavior. Academy of Management Journal,

37, 1379--1391.

Eden, D. (1985). Team development: A true

field experiment at three levels of rigor. Jour-


Eden, D., & Aviram, A. (1993). Self-efficacy

training to speed reemployment: Helping

people help themselves. Journal of Applied


Edwards, J. R. (1992). A cybernetic

theory of stress, coping, and well-being

in organizations. Academy of Management

Review, 17, 238--274.

Edwards, J. R. (1994). The study of congru-

ence in organizational behavior research:

Critique and a proposed alternative. Organ-

izational Behavior and Human Decision Proc-

esses, 58, 51--100.

Edwards, J. R., & Parry, M. E. (1993). On

the use of polynomial regression equations

as an alternative to difference scores in

organizational research. Academy of Man-

agement Journal, 36, 1577--1613.

Edwards, J. R., Scully, J. A., & Brtek, M.

D. (1999). The measurement of work:

Hierarchical representation of the multi-

method job design questionnaire. Personnel


Ekman, P. (1973). Cross-cultural studies

of facial expression. In P. Ekman (Ed.),

Darwin and facial expression: A century of

research in review (pp. 169--222). New

York: Academic Press.

Ellickson, M. C. (2002). Determinants of

job satisfaction of municipal government

employees. Public Personnel Management,

31, 343--358.

Ellis, S., & Arieli, S. (1999). Predicting

intentions to report administrative and

disciplinary infractions: Applying the rea-

soned action model. Human Relations, 52,

947--967.

References 533

Elsayed-Ekhouly, S. M., & Buda, R.

(1996). Organizational conflict: A compar-

ative analysis of conflict styles across

cultures. International Journal of Conflict

Management, 7(1), 71--80.

Enchautegui-de-Jesus, N., Hughes, D.,

Johnston, K. E., & Hyun, J. O. (2006).

Well-being in the context of workplace

ethnic diversity. Journal of Community Psy-

chology, 34(2), 211--223.

Equal Employment Opportunity Commis-

sion. (1980). Guidelines on sexual harass-

ment in the workplace [No. 45 FR 74676--

74677]. Washington, DC: Author.

Erez, A., & Judge, T. A. (2001). Relation-

ship of core self-evaluations to goal setting,

motivation, and performance. Journal of


Eskew, D., & Hennemen, R. L. (1996).

Survey of merit pay plan effectiveness: End

of the line for merit pay or hope for

improvement. Human Resource Planning,

19, 12--19.

Evans, M. G., & Ondrack, D. A. (1991).

The motivational potential of jobs: Is a

multiplicative model necessary? Psychologi-

cal Reports, 69, 659--672.

Fahr, J. L., Dobbins, G. H., & Cheng,

B. S. (1991). Cultural relativity in action:

A comparison of self-ratings made by Chi-

nese and U.S. workers. Personnel Psychol-

ogy, 44, 129--147.

Falbe, C. M., & Yukl, G. (1992). Conse-

quences for managers of using single influ-

ence tactics and combinations of tactics.

Academy of Management Journal, 35, 638--

652.

Falkenberg, L. W. (1987). Employee

fitness programs: Their impact on the

employee and organization. Academy of


Farrell, D., & Stamm, C. L. (1988). Meta-

analysis of the correlates of employee

absence. Human Relations, 41, 211--227.

Fayol, H. (1984). General and industrial

management. Belmont, CA: Lake Publishing.

Feldman, D. C. (1976). A contingency

theory of socialization. Administrative Sci-


Feldman, D. C. (1981). The multiple social-

ization of organizational members. Academy

of Management Review, 6, 309--318.

Ferrier, W., Smith, K., & Grimm, C.

(1999). The role of competitive action in

market share erosion and industry

dethronement: A study of industry leaders

and challengers. Academy of Management

Journal, 42, 372--388.

Ferris, G. R., Gilmore, D. C., & Kacmar,

K. M. (1990, April). Potential moderators of

the organizational politics-job anxiety rela-

tionship. Paper presented at the annual

Society for Industrial and Organizational

Psychology Convention, Miami, FL.

Festinger, L. (1954). A theory of social

comparison processes. Human Relations, 7,

114--140.

534 References

Fiedler, F. E. (1967). A theory of leader

effectiveness. New York: McGraw-Hill.

Fiedler, F. E., & Garcia, J. E. (1987). New

approaches to leadership: Cognitive resources

and organizational performance. New York:

Wiley.

Field, R. H. G., & House, R. J. (1990). A

test of the Vroom-Yetton model using

manager and subordinate reports. Journal


Finkelstein, M. A., & Penner, L. A.

(2004). Predicting organizational citizen-

ship behavior: Integrating the functional

and role identity approaches. Social Behav-

ior and Personality, 32, 383--398.

Finkelstein, S. (1992). Power in top man-

agement teams: Dimensions, measurement,

and validation. Academy of Management

Journal, 35, 505--538.

Finkelstein, S., & Boyd, B. K. (1998).

How much does the CEO matter? The role

of managerial discretion in setting of CEO

compensation. Academy of Management

Journal, 41, 179--199.

Finkelstein, S., & Hambrick, D. C. (1988).

Chief executive compensation: A synthesis

and reconciliation. Strategic Management

Journal, 9, 543--558.

Fischer, R. & Ury, W. (with B. Patton,

Ed.) (1981). Getting to YES: Negotiating

agreement without giving in. Boston:

Houghton-Mifflin.

Fishbein, M. (1979). A theory of rea-

soned action: Some applications and

implications. In H. Howe & M. Page

(Eds.), Nebraska symposium on motivation

(pp. 65--116). Lincoln: University of

Nebraska.

Fisher, C. D., & Gitelson, R. R. (1983). A

meta-analysis of the correlates of role con-

flict and role ambiguity. Journal of Applied


Fiske, S. T. (2004). Social beings: A core

motives approach to social psychology. Hobo-

ken, NJ: Wiley.

Fitzgerald, L. F. (1993). Sexual harass-

ment: Violence against women in the work-

place. American Psychologist, 48, 1070--

1076.

Fitzgerald, L. F., Drasgow, F., Hulin, C.

L., Gelfand, M. J., & Magley, V. J. (1997).

Antecedents and consequences of sexual

harassment in organizations: A test of an

integrated model. Journal of Applied Psy-

chology, 82, 359--378.

Flannery, T. P., Hofrichter, D. A., & Plat-

ten, P. E. (1996). People, performance, and

pay: Dynamic compensation for changing

organizations. New York: Free Press.

Fleishman, E., Harris, E. F., & Burtt, H. E.

(1955). Leadership and supervision in indus-

try. Columbus, OH: Bureau of Educational

Research, Ohio State University.

Florkowski, G. W. (1987). The organiza-

tional impact of profit sharing. Academy of


Florkowski, G. W., & Schuster, M. H.

(1992). Support for profit sharing and

References 535

commitment: A path analysis. Human Rela-

tions, 45, 507--523.

Folger, R., & Cropanzano, R. (1998).

Organizational justice and human resource

management. Thousand Oaks, CA: Sage.

Ford, R. N. (1973). Job enrichment les-

sons from AT&T. Harvard Business Review,

Sept/Oct, 96--106.

Forsyth, D. R. (1999). Group dynamics

(3rd ed.). Belmont, CA: Brooks/Cole-

Wadsworth.

Forsyth, D. R. (2006). Group dynamics (4th

ed.). Belmont, CA: Wadsworth/Thompson.

Foti, R. J., & Rueb, J. (1990, April). Self-

monitoring, traits, and leadership emergence.

Paper presented at the annual meeting of

the Society for Industrial and Organiza-

tional Psychology, Miami, FL.

Fowler, F. J., Jr. (1984). Survey research

methods. Beverly Hills, CA: Sage.

Fox, M. L., Dwyer, D. J., & Ganster, D. C.

(1993). Effects of stressful demands and

control on physiological and attitudinal

outcomes in a hospital setting. Academy of


Fox, S., & Spector, P. E. (1999). A

model of work frustration-aggression.

Journal of Organizational Behavior, 20(6),

915--931.

Frambach, R. T., & Schillewaert, N.

(2002). Organizational innovation adop-

tion: A multi-level framework of determi-

nants and opportunities for future

research. Journal of Business Research, 55,

163--176.

Frank, L. L., & Hackman, J. R. (1975). A

failure of job enrichment: The case of the

change that wasn’t. Journal of Applied


Frankenhaeuser, M. (1979). Psychoneuro-

endocrine approaches to the study of

emotion as related to stress and coping.

In H. E. Howe & R. A. Diensbier (Eds.),

Nebraska symposium on motivation (pp.

123--161). Lincoln, NE: University of

Nebraska Press.

French, J. R. P., Caplan, R. D., & Harrison,

R. V. (1982). The mechanisms of job stress

and strain. Chichester, England: Wiley.

French, J. R. P., Jr., & Kahn, R. L. (1962).

A programmatic approach to studying the

industrial environment and mental health.

Journal of Social Issues, 18, 1--47.

French, J. R. P., Jr., & Raven, B. H.

(1959). The bases of social power. In D.

Cartwright (Ed.), Studies in social power

(pp. 150--167). Ann Arbor, MI: Institute

for Social Research.

French, W. L., & Bell, C. H. (1995).

Organization development: Behavioral science

interventions for organization improvement

(5th ed.). Englewood Cliffs, NJ: Prentice-

Hall.

Frese, M. (1985). Stress at work and psy-

chosomatic complaints: A causal interpre-

tation. Journal of Applied Psychology, 70,

314--328.

536 References

Frese, M., Kring, W., Soose, A., & Zempel,

J. (1996). Personal initiative at work: Dif-

ferences between East and West Germany.


63.

Fried, Y., & Ferris, G. R. (1987). The val-

idity of the job characteristics model: A

review and meta-analysis. Personnel Psy-

chology, 40, 287--322.

Fried, Y., Rowland, K. M., & Ferris, G. R.

(1984). The physiological measurement of

work stress: A critique. Personnel Psychol-

ogy, 37, 583--615.

Friedland, N., Keinan, G., & Regev, Y.

(1992). Controlling the uncontrollable:

Effects of stress on illusory perceptions of

controllability. Journal of Personality and

Social Psychology, 63, 923--931.

Frone, M. R. (2003). Predictors of overall

and on-the-job substance use among

young workers. Journal of Occupational

Health Psychology, 8(1), 39--54.

Frone, M. R. (2004). Alcohol, drugs, and

workplace safety outcomes: A view from a

general model of employee substance use

and productivity. In J. Barling, & M. R.

Frone (Eds.), The psychology of workplace

safety (pp. 127--156). Washington, DC:

American Psychological Association.

Frone, M. R. (2006). Prevalence and distri-

bution of alcohol use and impairment in

the workplace: A U.S. national survey.

Journal of Studies on Alcohol, 67(1), 147--

156.

Fuller, J. A., Stanton, J. M., Fisher, G. G.,

Spitzmuller, C., Russel, S. S., & Smith, P.

C. (2003). A lengthy look at the daily

grind: Time series analysis of events,

mood, stress, and satisfaction. Journal of

Applied Psychology, 88(6), 1019--1033.

Galbraith, J. R. (1995). Designing organiza-

tions: An executive briefing on strategy, struc-

ture, and process. San Francisco: Jossey-Bass.

Gallagher, D. G. (2005). Part-time and

contingent employment. In J. Barling, E.

K. Kelloway, & M. R. Frone (Eds.), Hand-

book of work stress (pp. 517--541). Thou-

sand Oaks, CA: Sage.

Gallupe, R. B., Dennis, A. R., Cooper, W.

H., Valacich, J. S., Bastianutti, L. M., &

Nunamaker, J. F., Jr. (1992). Electronic

brainstorming and group size. Academy of


Gangreco, A., & Pecccie, R. (2005). The

nature and antecedents of middle manager

resistance to change: Evidence from an

Italian context. International Journal of

Human Resources Management, 16(10),

1812--1829.

Gannon, M. J. (1971). Sources of referral

and employee turnover. Journal of Applied


Ganster, D. C., & Schaubroeck, J. (1991).

Work stress and employee health. Journal

of Management, 17, 235--271.

Gavin, J. F. (1984). Survey feedback: The

perspectives of science and practice. Group

and Organization Studies, 9, 29--70.

References 537

Geertz, C. (1973). The interpretation of cul-

tures. New York: Basic Books.

Gelade, G. A., Dobson, P., & Gilbert, P.

(2006). National differences in organiza-

tional commitment: Effect of economy,

product of personality, or consequence of

culture? Journal of Cross-Cultural Psychol-

ogy, 37, 542--556.

George, J. M. (1990). Personality, affect,

and behavior in groups. Journal of Applied


George, J. M., & Bettenhausen, K. (1990).

Understanding prosocial behavior, sales

performance, and turnover: A group-level

analysis. Journal of Applied Psychology, 75,

689--709.

George, J. M., & Brief, A. P. (1992). Feel-

ing good-doing good: A conceptual analy-

sis of the mood at work-organizational

spontaneity relationship. Psychological Bul-

letin, 112, 310--329.

Gerhart, B. (1987). How important are dis-

positional factors as determinants of job

satisfaction? Implications for job design

and other personnel programs. Journal of


Gerhart, B. (1990). Voluntary turnover

and alternative job opportunities. Journal


Gerhart, B., & Milkovich, G. T. (1990).

Organizational differences in compensation

and financial performance. Academy of


Gerhart, B., & Milkovich, G. T. (1992).

Employee compensation: Research and

practice. In M. D. Dunnette & L. M.

Hough (Eds.), Handbook of industrial and

organizational psychology (2nd ed., Vol. 3,

pp. 481--569). Palo Alto, CA: Consulting


Gersick, C. J. G. (1988). Time and transi-

tion in work teams: Toward a new model

of group development. Academy of Manage-

ment Journal, 31, 9--41.

Gersick, C. J. G. (1989). Marking time:

Predictable transitions in task groups.


309.

Gerstner, C. R., & Day, D. V. (1997).

Meta-analytic review of leader-member

exchange theory: Correlates and construct

issues. Journal of Applied Psychology, 82,

827--844.

Gilliland, S. W., & Landis, R. S. (1992).

Quality and quantity goals in a complex

decision task: Strategies and outcomes.


681.

Gladstein, D. (1984). Groups in context: A

model of task group effectiveness. Adminis-

trative Science Quarterly, 29, 499--517.

Glick, W. H., Jenkins, G. D., Jr. & Gupta,

N. (1986). Method versus substance: How

strong are underlying relationships,

between job characteristics and attitudinal

outcomes? Academy of Management Journal,

29, 441--464.

538 References

Goff, S. J., Mount, M. K., & Jamison, R. L.

(1990). Employer supported childcare,

work/family conflict, and absenteeism: A

field study. Personnel Psychology, 43, 793--

810.

Golden, T. D., Veiga, J. F., & Simsek, Z.

(2006). Telecommuting’s differential impact

on work-family conflict: Is there no place

like home? Journal of Applied Psychology,

91(6), 1340--1350.

Goldstein, I. L. (1993). Training in organi-

zations (3rd ed.). Pacific Grove, CA:

Brooks/Cole.

Goldstein, I. L., & Ford, J. K. (2002).

Training in organizations: Needs assessment,

development, and evaluation (4th ed.). Bel-

mont, CA: Wadsworth.

Golembiewski, R. T., Billingsley, K., &

Yeager, S. (1976). Measuring change and

persistence in human affairs: Types of

change generated by OD designs. Journal

of Applied Behavioral Science, 23, 295--313.

Gomez-Mejia, L. R. (1994). Executive

compensation: A reassessment and future

research agenda. In G. R. Ferris (Ed.),

Research in personnel and human resource

management (Vol. 12, pp. 161--222).


Gomez-Mejia, L. R., & Welbourne, T. M.

(1989). The strategic design of executive

compensation programs. In L. R. Gomez-

Mejia (Ed.), Compensation and benefits (pp.

216--260). Washington, DC: Bureau of

National Affairs.

Goodale, J. G., & Aagaard, A. K. (1975).

Factors relating to varying reactions to the

four-day work week. Journal of Applied Psy-

chology, 60, 33--35.

Goodman, P. S. (1986). The impact of task

and technology on group performance. In

P. S. Goodman (Ed.), Designing effective

work groups (pp. 120--167). San Francisco:

Jossey-Bass.

Gordon, J. (1992). Work teams: How far

have they come? Training, 29, 59--65.

Graen, G. (1976). Role making process

within complex organizations. In M. D.

Dunnette (Ed.), Handbook of industrial and



Grandey, A. A. (2000). Emotional regula-

tion in the workplace: A new way to con-

ceptualize emotional labor. Journal of


110.

Grandey, A. A. (2003). Managing emotions

in the workplace. Personnel Psychology,

56(2), 563--566.

Grandey, A. A., Fisk, G. M., & Steiner, D.

D. (2005). Must ‘‘service with a smile’’ be

stressful? The moderating role of personal

control for American and French employ-

ees. Journal of Applied Psychology, 90(5),

893--904.

Grant, A. M. (2007). Relational job design

and the motivation to make a prosocial

difference. Academy of Management Review,

32(2), 393--417.

References 539

Grant, A. M., Campbell, E. M., Chen, G.,

Cottone, K., Lapedis, D., & Lee, K.

(2007). Impact and the art of motivation

maintenance: The effects of contact with

beneficiaries on persistence behavior.


Processes, 103(1) 53--67.

Graziano, W. G., Jensen-Campbell, L. A.,

& Hair, E. C. (1996). Perceiving interper-

sonal conflict and reacting to it: The case

for agreeableness. Journal of Personality and


Greenberg, J. (1990). Employee theft as a

reaction to underpayment inequity: The

hidden cost of pay cuts. Journal of Applied


Greenberg, J. (2002). Who stole the

money, and when? Individual and situa-

tional determinants of employee theft.


Processes, 89, 985--1003.

Greenhalgh, T., Robert, G., & MacFarlane,

F. (2004). Diffusion of innovations in serv-

ice organizations: Systematic review and

recommendations. Milbank Quarterly, 82,

581--629.

Greenhaus, J., & Buetell, N. (1985).

Source of conflict between work and fam-

ily roles. Academy of Management Review,

10, 76--88.

Griffeth, R. W., & Hom, P. W. (1987).

Some multivariate comparisons of multina-

tional managers. Multivariate Behavioral

Research, 22, 173--191.

Griffeth, R. W., Hom, P. W., & Gaertner,

S. (2000). A meta-analysis of antecedents

and correlates of employee turnover:

Update, moderator tests, and research

implications for the next millennium. Jour-

nal of Management, 26(3), 463--488.

Griffin, M. A., & Neal, A. (2000). Percep-

tions of safety at work: A framework for

linking safety climate to safety perfor-

mance, knowledge, and motivation. Journal

of Occupational Health Psychology, 5(3),

347--358.

Griffin, R. W. (1991). Effects of work

redesign on employee perceptions, atti-

tudes, and behavior: A long-term investiga-

tion. Academy of Management Journal, 34,

425--435.

Griffin, R. W., & McMahan, G. C. (1994).

Motivation through job design. In J.

Greenberg (Ed.), Organizational behavior:

State of the science (pp. 23--44). New York:

Erlbaum and Associates.

Gruber, J. E. (1998). The impact of male

work environments and organizational pol-

icies on women’s experiences of sexual

harassment. Gender and Society, 12, 301--

320.

Gruman, J. A., Saks, A. M., & Zweig, D. I.

(2006). Organizational socialization tactics

and newcomer proactive behaviors: An

integrative study. Journal of Vocational

Behavior, 69(1), 90--104.

Guion, R. M., & Highhouse, S. (2004).

Essentials of personnel assessment and

540 References

selection. New Jersey: Lawrence Erlbaum

Associates Publishers.

Gully, S. M., Incalcaterra, K. A., Joshi, A.,

& Beaubien, J. M. (2002). A meta-analysis

of team-efficacy, potency, and performance:

Interdependence and level of analysis as

moderators of observed relationships. Jour-


Gundling, E. (2000). The 3M way to inno-

vation: Balancing people and profit. Tokyo:

Kodansha International.

Gupta, N., & Jenkins, D. G. (1985). Dual

career couples: Stress, stressors, strains, and

strategies. In T. A. Beehr, & R. S. Bhagat

(Eds.), Human stress and cognition in organi-

zations (pp. 141--176). New York: Wiley.

Gutek, B., Cohen, A., & Konrad, A.

(1990). Predicting social-sexual behavior at

work: A contact hypothesis. Academy of


Guzzo, R. A., & Campbell, R. J. (1990,

August). Conditions for team effectiveness in

management. Paper presented at the annual

meeting of the Academy of Management,

San Francisco.

Guzzo, R. A., Jette, R. D., & Katzell, R. A.

(1985). The effects of psychologically

based intervention programs on worker

productivity: A metaanalysis. Personnel Psy-

chology, 38, 275--291.

Guzzo, R. A., & Shea, G. P. (1992). Group

performance and intergroup relations in

organizations. In M. D. Dunnette & L. M.





Guzzo, R. A., Yost, P. R., Campbell, R. J.,

& Shea, G. P. (1993). Potency in groups:

Articulating a construct. British Journal of


Hackett, R. D. (1989). Work attitudes and

employee absenteeism: A synthesis of the

literature. Journal of Occupational Psychol-

ogy, 62, 235--248.

Hackett, R. D., & Guion, R. M. (1985). A

reevaluation of the absenteeism-job satis-

faction relationship. Organizational Behavior

and Human Decision Processes, 35, 340--

381.

Hackman, J. R. (1987). The design of

work teams. In J. W. Lorsch (Ed.), Hand-

book of organizational behavior (pp. 315--

342). Englewood Cliffs, NJ: Prentice-Hall.

Hackman, J. R. (Ed.). (1990). Groups that

work (and those that don’t): Creating the con-

ditions for effective teamwork. San Francisco:

Jossey-Bass.

Hackman, J. R. (1992). Group influences

on individuals in organizations. In M. D.





Hackman, J. R., Brousseau, K. R., &

Weiss, J. A. (1976). The interaction of

task design and group performance strat-

egies in determining group effectiveness.

References 541

Organizational Behavior and Human Per-

formance, 16, 350--365.

Hackman, J. R., & Lawler, E. E. (1971).

Employee reactions to job characteristics.


Hackman, J. R., & Morris, C. G. (1975).

Group tasks, group interaction process,

and group performance effectiveness: A

review and proposed integration. In L.

Berkowitz (Ed.), Advances in experimental

social psychology (Vol. 9, pp. 47--87). New

York: Academic Press.

Hackman, J. R., & Oldham, G. R. (1975).

Development of the Job Diagnostic Survey.



Motivation through the design of work:

Test of a theory. Organizational Behavior

and Human Performance, 16, 250--279.


Work redesign. Reading, MA: Addison-

Wesley.

Haines, V. Y., & Taggar, S. (2006). Ante-

cedents of team reward attitude. Group

Dynamics: Theory, Research, and Practice,

10, 194--205.

Halbesleben, J. R. B., Osburn, H. K., &

Mumford, M. D. (2006). Action research

as a burnout intervention: reducing burn-

out in the federal fire service. Journal

of Applied Behavioral Science, 42(2), 244--

266.

Hall, D. T., & Nougaim, K. E. (1968). An

examination of Maslow’s need hierarchy in

an organizational setting. Organizational

Behavior and Human Performance, 3, 12--

35.

Hallam, G., & Campbell, D. (1997). The

measurement of team performance with a

standardized survey. In M. T. Brannick,

E. Salas, & C. Prince (Eds.), Team perform-

ance assessment and measurement (pp. 155--

171). Mahwah, NJ: Lawrence Erlbaum

Associates.

Hallock, D. E., Salazar, R. J., & Venne-

man, S. (2004). Demographic and attitudi-

nal correlates of employee satisfaction with

an ESOP. British Journal of Management, 15,

321--333.

Hammer, T. H., & Landau, J. (1981).

Methodological issues in the use of

absence data. Journal of Applied Psychology,

66, 574--581.

Hannan, M. T., Polos, L., & Carroll, G. R.

(2003). The fog of change: Opacity and

asperity in organizations. Administrative

Science Quarterly, 48(3), 399--432.

Hansen, C. P. (1988). Personality charac-

teristics of the accident involved employee.

Journal of Business and Psychology, 2, 346--

365.

Hansen, C. P. (1989). A causal model of

the relationships among accidents, biodata,

personality, and cognitive factors. Journal


Harrison, D. A., Newman, D. A., & Roth,

P. L. (2006). How important are job

attitudes? Meta-analytic comparisons of

542 References

integrative behavioral outcomes and time

sequences. Academy of Management Journal,

49, 305--325.

Hatch, M. J. (1993). The dynamics of

organizational culture. Academy of Manage-

ment Review, 18, 657--693.

Hatcher, L., & Ross, T. L. (1991). From

individual incentives to an organization-

wide gain sharing plan: Effects on team-

work and product quality. Journal of


Hatton, C., Rivers, M., Mason, H., Mason,

L., Emerson, E., Kiernan, C., et al. (1999).

Organizational culture and staff outcomes

in services for people with intellectual dis-

abilities. Journal of Intellectual Disability

Research, 43, 206--218.

Hays, W. L. (1988). Statistics (4th ed.).

New York: Holt, Rinehart and Winston.

Heinisch, D. A., & Jex, S. M. (1997). Neg-

ative affectivity and gender as moderators

of the relationship between work-related

stressors and depressed mood at work.

Work and Stress, 11, 46--57.

Hellervik, L. W., Hazucha, J. F., &

Schneider, R. J. (1992). Behavior change:

Models, methods, and a review of evidence.

In M. D. Dunnette & L. M. Hough (Eds.),


chology (2nd ed., Vol. 3, pp. 823--895). Palo


Hellstrom, C., & Hellstrom, T. (2002).

Highways, alleys, and by-lanes: Charting

the pathways for ideas and innovation in

organizations. Creativity and Innovation


Henkoff, R. (1990). Cost cutting: How to

do it right. Fortune, Apr, 9, 40--46.

Henry, R., & Hulin, C. L. (1987). Stability

of skilled performance across time: Some

generalizations and limitations on utilities.


Henry, R., & Hulin, C. L. (1989). Chang-

ing validities: Ability-performance relations

and utilities. Journal of Applied Psychology,

74, 365--367.

Hertel, G., Konradt, U., & Orlikowski, B.

(2004). Managing distance by interdepend-

ence: Goal setting, task interdependence,

and team-based rewards in virtual teams.

European Journal of Work and Organiza-

tional Psychology, 13, 1--28.

Herzberg, F. (1968). One more time: How

do you motivate employees? Harvard Busi-

ness Review, Jan/Feb, 52--62.

Herzberg, F., Mausner, B., & Snyderman,

B. (1959). The motivation to work. New

York: Wiley.

Hickson, D. J., Pugh, D. S., & Pheysey, D.

(1969). Operations technology and organiza-

tion structure: An empirical reappraisal.

Administrative Science Quarterly, 14, 378--397.

Highhouse, S., & Hoffman, J. R. (2001).

Organizational attraction and job choice.

In C. L. Cooper & I. T. Robertson (Eds.),

International Review of Industrial and Orga-

nizational Psychology 2001 (Vol. 16, pp.

37--64). Chichester, England: Wiley.

References 543

Hirschfeld, R. R., Jordan, M. H., Field, H.

S., Giles, W. F., & Armenakis, A. A.

(2006). Becoming team players: Team

members’ mastery of teamwork knowledge

as a predictor of team task proficiency and

observed teamwork effectiveness. Journal of


Hochschild, A. R. (1979). Emotion work,

feeling rules, and social structure. American

Journal of Sociology, 85, 551--575.

Hochschild, A. R. (1983). The managed

heart: Commercialization of human feeling.

Berkeley: University of California Press.

Hochschild, A. R. (1989). The second shift:

Working parents and the revolution at home.

New York: Viking.

Hofmann, D. A., & Morgeson, F. P.

(1999). Safety-related behavior as a social

exchange: The role of perceived organiza-

tional support and leader-member

exchange. Journal of Applied Psychology, 84,

286--296.

Hofmann, D. A., & Stetzer, A. (1998). The

role of safety climate and communication

in accident interpretation: Implication

from negative events. Academy of Manage-


Hofmann, D. A., & Tetrick, L. E. (2003).

The etiology of the concept of health:

Implications for ‘‘organizing’’ individual

and organizational health. In D. A. Hof-

mann and L. E. Tetrick (Eds.), Health

and safety in organizations: A multilevel per-

spective (pp. 1--28). San Francisco: Jossey-

Bass.

Hofstede, G. (1980). Culture’s consequences:

International differences in work-related val-

ues. Beverly Hills, CA: Sage.

Hofstede, G. (1984). Culture’s consequences:

International differences in work-related values

(Abridged ed.). Newbury Park, CA: Sage.

Hogan, E. A., & Overmeyer-Day, L.

(1994). The psychology of mergers and

acquisitions. In C. L. Cooper & I. T. Rob-

ertson (Eds.), International review of indus-

trial and organizational psychology 1994

(Vol. 9, pp. 247--280). Chichester, Eng-

land: Wiley.

Hogan, J., & Hogan, R. (1989). How to

measure employee reliability. Journal of


Hollander, E. P. (1971). Principles and

methods of social psychology (2nd ed.). New

York: Oxford University Press.

Hollenbeck, J. R., Ilgen, D. R., & Cramp-

ton, S. M. (1992). Lower back disability in

occupational settings: A review of the liter-

ature from a human resource management

view. Personnel Psychology, 45, 247--278.

Holtom, B. C., Mitchell, T. R., Lee, T. W.,

& Interrieden, E. J. (2005). Shocks as

causes of turnover: What they are and how

organizations can manage them. Human

Resources Management, 44(3), 337--352.

Hom, P. W., Caranikas-Walker, F., Prus-

sia, G. E., & Griffeth, R. W. (1992). A

meta-analytic structural equations analysis

of a model of employee turnover. Journal


544 References

Homans, G. C. (1958). Social behavior as

exchange. American Journal of Sociology, 63,

597--606.

Hornstein, H. A., & deGuerre, D. W.

(2006). Bureaucratic organizations are bad

for our health. Ivey Business Journal, March/

April, 1--4.

House, R. J. (1971). A path-goal theory of

leader effectiveness. Administrative Science

Quarterly, 16, 321--339.

House, R. J. (1977). A 1976 theory of

charismatic leadership. In J. G. Hunt &

L. L. Larson (Eds.), Leadership: The cutting

edge. Carbondale: Southern Illinois Univer-

sity Press.

House, R. J. (1996). Path-goal theory of lead-

ership: Lessons, legacy, and a reformulated

theory. Leadership Quarterly, 7, 323--352.

House, R. J., Hanges, P. J., Javidan, M.,

Dorfman, P. W., & Gupta, V. (Eds.).

(2004). Culture, leadership, and organiza-

tions: The GLOBE study of 62 societies.

Thousand Oaks, CA: Sage Publications.

House, R. J., & Mitchell, T. R. (1974).

Path-goal theory of leadership. Contempo-

rary Business, 3, 81--98.

Hovorka-Mead, A. D., Ross, W. H.,

Whipple, T., & Renchin, M. B. (2002).

Watching the detectives: Seasonal student

employee reactions to electronic monitor-

ing with and without advance notification.

Personnel Psychology, 55(2), 329--362.

Huang, X., & Van de Vliert, E. (2003).

Where intrinsic job satisfaction fails to

work: National moderators of intrinsic

motivation. Journal of Organizational Behav-

ior, 24, 159--179.

Huang, X., & Van de Vliert, E. (2004). Job

level and national culture as roots of job

satisfaction. Applied Psychology: An Interna-

tional Review, 53, 329--348.

Hugick, L., & Leonard, J. (1991). Job dis-

satisfaction grows; ‘‘moonlighting’’ on the

rise. Gallop Poll News Service, 56, 1--11.

Hulin, C. L. (1991). Adaptation, commit-

ment, and persistence in organizations. In

M. D. Dunnette, & L. M. Hough (Eds.),


chology (2nd ed., Vol. 2, pp. 445--505). Palo


Hunter, J. E., Schmidt, F. L., & Huy, L.

(2006). Implications of direct and indirect

range restriction for meta-analyis methods

and findings. Journal of Applied Psychology,

91, 594--612.

Hunter, J. E., Schmidt, F. L., & Judiesch,

M. K. (1990). Individual differences in

output variability as a function of job com-

plexity. Journal of Applied Psychology, 75,

28--42.

Hurrell, J. J., Jr. (1995). Commentary:

Police work, occupational stress, and indi-

vidual coping. Journal of Organizational

Behavior, 16, 27--28.

Iacocca, L. (1984). Iacocca: An autobiogra-

phy. New York: Bantam Books.

Iaffaldano, M. T., & Muchinsky, P. M.

(1985). Job satisfaction and job performance:

References 545

A meta-analysis. Psychological Bulletin, 97,

251--273.

Ilgen, D. R., & Davis, C. A. (2000). Bear-

ing bad news: Reaction to negative per-

formance feedback. Applied Psychology: An

International Review, 49, 550--565.

Ilgen, D. R., Hollenbeck, J. R., Johnson,

M., & Jundt, D. (2005). Teams in organi-

zations: From input-process-output models

to IMOI models. Annual Review of Psychol-

ogy, 56, 517--543.

Ilgen, D. R., Mitchell, T. R., & Frederickson,

J. W. (1981). Poor performers: Supervisors’

and subordinates’ responses. Organizational


410.

Ilies, R., & Judge, T. A. (2003). On the

heritability of job satisfaction: The media-

ting role of personality. Journal of Applied


International Telework Association and

Council. (2001). Telework America report.

Retrieved November 20, 2007, from http://

www.workingfromanywhere.org/telework/

twa2001.htm.

Ironson, G. H., Smith, P. C., Brannick, M.

T., Gibson, W. M., & Paul, K. B. (1989).

Constitution of a job in general scale: A

comparison of global, composite, and spe-

cific measures. Journal of Applied Psychol-

ogy, 74, 193--200.

Ivancevich, J. M., & Matteson, M. T.

(1980). Stress and work: A managerial per-

spective. Glenview, IL: Scott, Foresman.

Ivancevich, J. M., Schweiger, D. M., &

Power, F. R. (1987). Strategies for manag-

ing human resources during mergers and

acquisitions. Human Resource Planning, 10,

19--35.

Jackofsky, E. F. (1984). Turnover and per-

formance: An integrated process model.

Academy of Management Review, 9, 74--83.

Jackson, S. E., & Schuler, R. S. (1985). A

meta-analysis and conceptual critique of

research on role ambiguity and role conflict

in work settings. Organizational Behavior

and Human Decision Processes, 36, 16--78.

Jackson, S. E., Stone, V. A., & Alvarez, E.

B. (1992). Socialization amidst diversity:

The impact of demographics on work team

oldtimers and newcomers. In L. L. Cum-

mings & B. M. Staw (Eds.), Research in

organizational behavior (Vol. 15, pp. 45--

109). Greenwich, CT: JAI Press.

James, L. R., & Brett, J. (1984). Mediators,

moderators, and tests for mediation. Jour-


James, L. R., Demaree, R. G., & Wolf, G.

(1984). Estimating within-group interrater

reliability with and without response bias.


James, L. R., & Jones, A. P. (1974).

Organizational climate: A review of theory

and research. Psychological Bulletin, 81,

1096--1112.

James, L. R., Mulaik, S., & Brett, J.

(1982). Causal analysis: Assumptions, mod-

els, and data. Beverly Hills, CA: Sage.

546 References

Janis, I. L. (1982). Groupthink: Psychological

studies of policy decisions and fiascos (2nd

ed.). Boston: Houghton Mifflin.

Janson, N. (1994). Safety culture: A study of

permanent way staff at British Rail. Amster-

dam: Vrije University.

Jarzabkowski, P., & Searle, R. H. (2004).

Harnessing diversity and collective action

in the top management team. Long Range

Planning, 37, 399--419.

Jaskyte, K. (2005). The impact of organiza-

tional socialization tactics on role ambigu-

ity and role conflict of newly hired social

workers. Administration in Social Work,

29(4), 69--87.

Jeanneret, P. R. (1991). Growth trends in

I/O psychology. Industrial-Organizational

Psychologist, 29, 47--52.

Jehn, K. A. (1994). Enhancing effective-

ness: An investigation of advantages and

disadvantages of value-based intragroup

conflict. International Journal of Conflict


Jenkins, G. D., Jr., Mitra, A., Gupta, N., &

Shaw, J. D. (1998). Are financial incentives

related to performance? A meta-analytic

review of empirical research. Journal of


Jex, S. M. (1991). The psychological bene-

fit of exercise in work settings: A review,

critique, and alternative model. Work and

Stress, 5, 133--147.

Jex, S. M. (1998). Stress and job perfor-

mance: Theory, research, and implications for

managerial practice. Thousand Oaks, CA:

Sage.

Jex, S. M., Adams, G. A., Bachrach, D. G.,

& Sorenson, S. (2003). The impact of sit-

uational constraints, role stressors, and

commitment on employee altruism. Journal

of Occupational Health Psychology, 8(3),

171--180.

Jex, S. M., Adams, G. A., Elacqua, T. C.,

& Bachrach, D. J. (1998, May). Type A as

a moderator: An examination using compo-

nent measures. Paper presented at the 1998

annual conference of the American Psy-

chological Society, Washington, DC.

Jex, S. M., & Beehr, T. A. (1991). Emerging

theoretical and methodological issues in the

study of work-related stress. In G. R. Ferris

& K. M. Rowland (Eds.), Research in person-

nel and human resources management (Vol. 9,

pp. 311--364). Greenwich, CT: JAI Press.

Jex, S. M., Beehr, T. A., & Roberts, C. K.

(1992). The meaning of occupational

‘‘stress’’ items to survey respondents. Jour-


Jex, S. M., & Bliese, P. D. (1999). Efficacy

beliefs as a moderator of the impact of

work-related stressors: A multilevel study.


Jex, S. M., & Elacqua, T. C. (1999). Self-

esteem as a moderator: A comparison of

global and organization-based measures.

Journal of Occupational and Organizational


Jex, S. M., & Heinisch, D. A. (1996).

Assessing the relationship between exercise

References 547

and employee mental health: Some meth-

odological concerns. In J. Kerr, A. Grif-

fiths, & T. Cox (Eds.), Workplace health:

Employee fitness and exercise (pp. 55--67).

London: Taylor & Francis.

Jex, S. M., & Spector, P. E. (1988, April).

Is social information processing a laboratory

phenomenon? Paper presented in M.

Deselles (Chair), Job attitude measure-

ment: Variations on a theme. Symposium

presented at the 1988 Society for Indus-

trial and Organizational Psychology Con-

vention, Dallas, TX.

Jex, S. M., & Spector, P. E. (1989). The

generalizability of social information proc-

essing to organizational settings: A sum-

mary of two field experiments. Perceptual

and Motor Skills, 69, 883--893.

Jex, S. M., & Spector, P. E. (1996). The

impact of negative affectivity on stressor-

strain relations: A replication and exten-

sion. Work and Stress, 10, 36--45.

Johns, G. (1991). Substantive and meth-

odological constraints on behavior and

attitudes in organizational research. Orga-

nizational Behavior and Human Decision

Processes, 49, 80--104.

Johns, G., & Xie, J. L. (1998). Perceptions

of absence from work: People’s Republic

of China versus Canada. Journal of Applied


Johns, G., Xie, J. L., & Fang, Y. (1992).

Mediating and moderating effects in job

design. Journal of Management, 18, 657--

676.

Johnson, J. G. (1990). Selecting ethno-

graphic informants. Newbury Park, CA:

Sage.

Johnson, W. B., & Packer, A. H. (1987).

Workforce 2000: Work and workers for the

21st century. Indianapolis, IN: Hudson

Institute.

Jones, B., Flynn, D. M. & Kelloway, E. K.

(1995). Perception of support from the

organization in relation to work stress, sat-

isfaction, and commitment. In S. L. Sauter

& L. R. Murphy (Eds.), Organizational risk

factors for job stress (pp. 41--52). Washing-

ton, DC: American Psychological Associa-

tion.

Jones, E. E., & Pittman, T. S. (1982).

Toward a general theory of strategic self-

presentation. In J. Suls (Ed.), Psychological

perspectives on the self (pp. 231--261).

Hillsdale, NJ: Lawrence Erlbaum.

Jones, J. W., & Boye, M. W. (1992). Job

stress and employee counterproductivity.

In J. C. Quick, J. J. Hurrell, & L. R. Mur-

phy (Eds.), Stress and well-being at work

(pp. 239--251). Washington, DC: Ameri-

can Psychological Association.

Jonsson, D., Rosengren, A., Dotevall, A.,

Lappas, G., & och Wilhelmensen, L.

(1999). Job control, job demands, and

social support at work in relation to cardi-

ovascular disease risk factors in MONICA

1995, Goteborg. Journal of Cardiovascular

Risk, 6, 379--385.

Jordan, M., Herriot, P., & Chalmers, C.

(1991). Testing Schneider’s ASA theory.

548 References

Applied Psychology: An International Review.

40, 47--54.

Joshi, A., Liao, H., & Jackson, S. E.

(2006). Cross-level effects of workplace

diversity on sales performance and pay.

Academy of Management, 49(3), 459--481.

Joyce, W. F. (1986). Matrix organization:

A social experiment. Academy of Manage-


Judge, T. A., Bono, J. E., Eriz, A., &

Locke, E. A. (2005). Core self-evaluations

and job and life satisfaction: The role of

self-concordance and goal attainment. Jour-


Judge, T. A., & Cable, D. M. (1997).

Applicant personality, organizational cul-

ture, and organizational attraction. Person-


Judge, T. A., & Colquitt, J. A. (2004).

Organizational justice and stress: The medi-

ating role of work-family conflict. Journal of

Applied Psychology, 89(3), 395--404.

Judge, T. A., Heller, D., & Mount, M. K.

(2002). Five-factor model of personality

and job satisfaction: A meta-analysis. Jour-


Judge, T. A., & Piccolo, R. F. (2004).

Transformational and transactional leader-

ship: A meta-analytic test of their relative

validity. Journal of Applied Psychology, 89,

755--768.

Judge, T. A., Piccolo, R. F., & Ilies, R.

(2004). The forgotten ones? The validity of

consideration and initiating structure in

leadership research. Journal of Applied Psy-

chology, 89, 36--51.

Judge, T. A., Scott, B. A., & Ilies, R.

(2006). Hostility, job attitudes, and work-

place deviance: Test of a multilevel model.

Journal of Applied Psychology, 91(1), 126--

138.

Judge, T. A., Thoresen, C. J., Bono, J. E.,

& Patton, G. K. (2001). The job satisfac-

tion-job performance relationship: A quali-

tative and quantitative review. Psychological

Bulletin, 127, 376--407.

Judge, T. A., Van Vianen, A. E. M., & De

Pater, I. E. (2004). Emotional stability, core

self-evaluations, and job outcomes: A review

of the evidence and an agenda for future

research. Human Performance, 17, 325--346.

Kahn, R. L. & Byosiere, P. (1992). Stress

in organizations. In M. D. Dunnette & L.

M. Hough (Eds.), Handbook of industrial

and organizational psychology (2nd ed., Vol.

2, pp. 571--650). Palo Alto, CA: Consult-

ing Psychologists Press.

Kahn, R. L., Wolfe, D. M., Quinn, R. P.,

Snoek, J. D., & Rosenthal, R. A. (1964).

Organizational stress: Studies in role conflict

and ambiguity. New York: Wiley.

Kammeyer-Mueller, J. D., Wanberg, C. R.,

Glomb, T. M., & Ahlburg, D. (2005). The

role of temporal shifts in the turnover

process: It’s about time. Journal of Applied


Kanfer, R. (1990). Motivation theory and

industrial/organizational psychology. In

References 549

M. D. Dunnette & L. M. Hough (Eds.),

Handbook of industrial and organizational

psychology (2nd ed., Vol. 1, pp. 75--170).

Palo Alto, CA: Consulting Psychologists

Press.

Kanfer, R., Ackerman, P. L., Murtha, T.

C., Dugdale, B., & Nelson, L. (1994). Goal

setting, conditions of practice, and task

performance: A resource allocation per-

spective. Journal of Applied Psychology, 79,

826--835.

Kanigel, R. (1997). The one best way: Fred-

erick Winslow Taylor and the enigma of effi-

cency. New York: Viking.

Karambayya, R. (1989). Contexts for orga-

nizational citizenship behavior: Do high per-

forming and satisfying units have better

‘‘citizens’’? York University Working Paper,

North York, Ontario, Canada.

Karasek, R. A. (1979). Job demands, job

decision latitude, and mental strain: Impli-

cations for job redesign. Administrative Sci-


Karasek, R. A., Baker, D., Marxer, F., Ahl-

bom, A., & Theorell, T. (1981). Job decision

latitude, job demands, and cardiovascular

disease: A prospective study of Swedish

men. American Journal of Public Health, 71,

694--705.

Katz, D., & Kahn, R. L. (1966). The social

psychology of organizations. New York:

Wiley.

Katz, D., & Kahn, R. L. (1978). The social

psychology of organizations (2nd ed.). New

York: Wiley.

Katzell, R. A., & Austin, J. T. (1992).

From then to now: The development of

industrial-organizational psychology in the

United States. Journal of Applied Psychology,

77, 803--835.

Kauffeld, S., Jonas, E., & Grote, S.

(2004). Climate for innovation: Creation

and first psychometric validation of an

instrument for measuring the climate for

innovation in organizations. Diagnostica,

50, 153--164.

Keenan, A., & Newton, T. J. (1985). Stress-

ful events, stressors and psychological strains

in young professional engineers. Journal of

Occupational Behavior, 6, 151--156.

Kelley, H. H. (1973). The process of causal

attribution. American Psychologist, 28, 107--

128.

Kelley, H. H., & Thibaut, J. W. (1978).

Interpersonal relations: A theory of interde-

pendence. New York: Wiley.

Kelman, H. C. (1958). Compliance, identi-

fication, and internalization: Three proc-

esses of attitude change. Journal of Conflict

Resolution, 2, 51--60.

Kemp, C. F., Zaccaro, S. J., Jordan, M., &

Flippo, S. (2004, April). Cognitive, social,

and dispositional influences on leader adapt-

ability. Poster presented at the 19th annual

meeting of the Society for Industrial and

Organizational Psychology, Dallas, TX.

Keppel, G. (1982). Design and analysis: A

researcher’s handbook (2nd ed.). Englewood

Cliffs, NJ: Prentice-Hall.

550 References

Keppel, G., & Zedeck, S. (1989). Data

analysis for research designs: Analysis of var-

iance and multiple regression/correlation

approaches. New York: Henry Holt &

Company.

Kerr, J. H. (1975). On the folly of reward-

ing A, while hoping for B. Academy of


Kerr, J. H., & Vos, M. C. (1993).

Employee fitness programs, absenteeism,

and general well-being. Work and Stress, 7,

179--190.

Kets de Vries, M. F. R., & Miller, D.

(1986). Personality, culture, and organiza-

tion. Academy of Management Review, 11,

266--279.

Kim, T. Y., Cable, D. M., Kim, S. P.

(2005). Socialization tactics, employee pro-

ductivity, and person-organization fit. Jour-

nal of Applied Psychology, 90(2), 232--241.

King, L. A., & King, D. W. (1990). Role

conflict and role ambiguity: A critical

assessment of construct validity. Psycholog-

ical Bulletin, 107, 48--64.

Kipnis, D. (1984). The use of power in

organizations and interpersonal settings.

Applied Social Psychology Annual, 5, 179--

210.

Kirkman, B. L., Rosen, B., Tesluk, P. E., &

Gibson, C. B. (2004). The impact of team

empowerment on virtual team perfor-

mance: The moderating role of face-to-face

interaction. Academy of Management Jour-

nal, 47(2), 175--192.

Kirsch, I. (2006). Medication and sugges-

tion in the treatment of depression. In M.

D. Yapko (Ed.), Hypnosis and treating

depression: Applications in clinical practice

(pp. 271--280). New York: Routledge/Tay-

lor & Francis Group.

Klaas, B. S. (1989). Determinants of griev-

ance activity and the grievance system’s

impact on employee behavior: An integrated

perspective. Academy of Management Review,

14, 445--458.

Klein, H. J. (1989). An integrated control

theory model of work motivation. Academy


Klein, K. J. (1987). Employee stock owner-

ship and employee attitudes: A test of

three models [Monograph]. Journal of


Klein, K. J., Dansereau, F. J., & Hall, R. J.

(1994). Levels issues in theory develop-

ment, data collection, and analysis. Acad-

emy of Management Review, 19, 195--229.

Klein, S. M., Kraut, A. I., & Wolfson, A.

(1971). Employee reactions to attitude sur-

vey feedback: A study of the impact of

structure and process. Administrative Sci-


Koestner, R., Zuckerman, M., & Olsson, J.

(1990). Attributional style, comparison

focus of praise, and intrinsic motivation.

Journal of Research in Personality, 24, 87--

100.

Kohler, J. M., & Munz, D. C. (2006).

Combining individual and organizational

References 551

stress interventions: An organizational

development approach. Consulting Psychol-

ogy Journal: Research and Practice, 58(1), 1--

12.

Kohler, S. S., & Mathieu, J. E. (1993).

Individual characteristics, work percep-

tions, and affective reactions influences on

differentiated absence criteria. Journal of


Kokotovich, M., Jex, S. M., & Adams, G. A.

(2000, April). Leader-member exchange: A

moderator of the stressor-satisfaction relation-

ship. Paper presented at the annual meeting

of the Society for Industrial and Organiza-

tional Psychology, New Orleans, LA.

Komaki, J., Barwick, K. D., & Scott, L. R.

(1978). A behavioral approach to occupa-

tional safety: Pinpointing and reinforcing

safe performance in a food manufacturing

plant. Journal of Applied Psychology, 63,

434--445.

Komaki, J., Heinzmann, A. T., & Lawson,

L. (1980). Effective training and feedback:

Component analysis of a behavioral safety

program. Journal of Applied Psychology, 65,

261--270.

Konovsky, M. A., & Cropanzano, R.

(1991). Perceived fairness of employee

drug testing as a predictor of employee

attitudes and health. Journal of Applied Psy-

chology, 76, 698--707.

Konovsky, M. A., & Pugh, S. D. (1990).

Citizenship behavior and social exchange.


669.

Kossek, E. E., & Ozeki, C. (1998). Work-

family conflict, policies, and the job-life sat-

isfaction relationship: A review and direc-

tions for organizational behavior-human

resources research. Journal of Applied Psy-

chology, 83, 139--149.

Kotter, J. P., & Heskett, J. L. (1992). Cor-

porate culture and performance. New York:

Free Press.

Kozlowski, S. W. J. & Bell, B. S. (2003).

Work groups and teams in organization.

In W. C. Borman, D. R. Ilgen, & R. J.

Klimoski (Eds.), Handbook of psychology:

Industrial and organizational psychology (pp.

333--375). Hoboken, NJ: Wiley.

Kozlowski, S. W. J., Chao, G. T., Smith, E.

M., & Hedlund, J. (1993). Organizational

downsizing: Strategies, interventions, and

research implications. In C. L. Cooper &

I. T. Robertson (Eds.), International review


(Vol. 8, pp. 263--332). London: Wiley.

Kraizberg, E., Tziner, A., & Weisberg, J.

(2002). Employee stock options: Are they

indeed superior to other incentive com-

pensation schemes? Journal of Business and


Kraus, S. J. (1995). Attitudes and the pre-

diction of behavior: A meta-analysis of the

empirical literature. Personality and Social

Psychology Bulletin, 21, 58--75.

Kristof, A. L. (1996). Person-organization

fit: An integrative review of its conceptual-

ization, measurement, and implications.

Personnel Psychology, 49, 1--50.

552 References

Kuhn, R. (1988). Psychological tests reduce

counterproductive acts by employees. Assets

Protection, 9, 9--12.

Kujacic, M., & Bojovic, N. (2003). Organi-

zational design of post corporation structure

using fuzzy multicriteria decision making.

Computation & Mathematical Organizational

Theory, 9, 5--18.

Kunin, T. (1955). The construction of a

new type of attitude measure. Personnel


LaBianca, G., Brass, D. J., & Gray, B.

(1998). Social networks and perceptions of

intergroup conflict: The role of negative

relationships and third parties. Academy of


Landy, F. J., & Farr, J. L. (1980). Perfor-

mance rating. Psychological Bulletin, 87, 72--

107.

Lapierre, L. M., & Allen, T. D. (2006).

Work-supportive family, family-supportive

supervision, use of organizational benefits,

and problem-focused coping: Implications

for work-family conflict and employee

well-being. Journal of Occupational Health

Psychology, 11(2), 169--181.

Latack, J. C., & Foster, L. W. (1985).

Implementation of compressed work

schedules: Participation and job redesign

as critical factors. Personnel Psychology, 38,

75--92.

Latack, J. C., & Halvovic, S. J. (1992).

Coping with job stress: A conceptual eval-

uation framework for coping measures.

Journal of Organizational Behavior, 13(5),

479--508.

Latane, B., Williams, K., & Harkins, S.

(1979). Many hands make light the work:

The causes and consequences of social

loafing. Journal of Personality and Social


Latham, G. P., & Huber, V. L. (1992).

Schedules of reinforcement: Lessons from

the past and issues for the future. Journal

of Organizational Behavior Management, 12,

125--149.

Latham, G. P., & Locke, E. A. (1991).

Self-regulation through goal setting. Orga-

nizational Behavior and Human Decision

Processes, 50, 212--247.

Latham, G. P., & Pinder, C. G. (2005).

Work motivation and research at the dawn

of the twenty-first century. Annual Review

of Psychology, 56, 485--516.

Lawler, E. E. (1990). Strategic pay: Aligning

organizational strategies and pay systems.

San Francisco: Jossey-Bass.

Lawler, E. E., & Jenkins, D. G. (1992).

Strategic reward systems. In M. D. Dun-



ed., Vol. 3, pp. 1009--1035). Palo Alto,


Lawler, E. E., Koplin, C. A., Young, T. F.,

& Fadem, J. A. (1968). Inequity reduction

over time in an overpayment situation.


formance, 3, 253--268.

References 553

Lawler, E. E., Mohrman, S., & Ledford, G.

(1995). Creating high performance organiza-

tions: Practices and results of, employee

involvement and TQM in Fortune 1000 com-

panies. San Francisco: Jossey-Bass.

Lawrence, P. R., & Lorsch, J. W. (1967).

Organization and environment: Managing

differentiation and integration. Boston:

Harvard Business School, Division of

Research.

Lazarus, R. S. (1966). Psychological stress

and the coping process. New York: McGraw-

Hill.

Leana, C. R., & Feldman, D. D. (1992).

Coping with job loss. New York: Lexington

Books.

Leavitt, H. J. (1951). Some effects of cer-

tain communication patterns on group

performance. Journal of Abnormal and Social


Lee, K., Allen, N. J., Meyer, J. P., & Rhee,

K. (2001). The three-component model of

organizational commitment: An application

to South Korea. Applied Psychology: An

International Review, 50, 596--614.

Lee, T. W., & Mitchell, T. R. (1994). An

alternative approach: The unfolding model

of voluntary employee turnover. Academy


Lee, T. W., Mitchell, T. R., Holtom, B. C.,

McDaniel, L. S., & Hill, J. W. (1999). The

unfolding model of voluntary turnover: A

replication and extension. Academy of Man-


Lee, T. W., Mitchell, T. R., Wise, L., &

Fireman, S. (1996). An unfolding model of

employee turnover. Academy of Manage-


Lee, T. W., Mitchell, T. R., Sablynski, C.

J., Burton, J. P., & Holto, B. C. (2004).

The effects of job embeddedness on orga-

nizational citizenship, job performance,

volitional absences, and voluntary turn-

over. Academy of Management Journal, 47,

711--722.

Lehman, W. E., Farabee, D. J., Holcom,

M. L., & Simpson, D. D. (1995). Predic-

tion of substance use in the workplace:

Unique contributions of personal back-

ground and work environment variables.

Journal of Drug Issues, 25, 253--274.

LePine, J. A., Colquitt, J. A., & Erez, A.

(2000). Adaptability to changing task con-

texts: Effects of general cognitive ability,

conscientiousness, and openness to experi-

ence. Personnel Psychology, 53, 563--593.

Levin, I., & Stokes, J. P. (1989). Disposi-

tional approach to job satisfaction: Role of

negative affectivity. Journal of Applied Psy-

chology, 74, 752--758.

Levine, J. M., & Moreland, R. L. (1990).

Progress in small group research. In M. R.

Rosenzweig & L. W. Porter (Eds.), Annual

Review of Psychology, 41, 585--634.

Levine, J. M., Moreland, R. L., & Choi, H.

S. (2001). Group socialization and new-

comer innovation. In M. A. Hogg & R. S.

Tindale (Eds.), Blackwell handbook of social

554 References

psychology: Group processes (pp. 86--106).

Malden, MA: Blackwell.

Levy, P. E., Cober, A. B., & Keeping, L.

M. (2003). Organizational web sites: Web

site style and content and determinants of

organizational attraction. International Jour-

nal of Selection and Assessment, 11, 158--

169.

Lewin, K. (1947). Frontiers in group dynam-

ics. Human Relations, 1, 26--41.

Lewin, K. (1951). Field theory in social sci-

ence. New York: Harper.

Lewis, M. (2004). Moneyball: The art of

winning an unfair game. New York: W. W.

Norton & Company.

Lickel, B., Hamilton, D. L., Wieczorkow-

ska, G., Lewis, A., Sherman, S. J., &

Uhles, A. N. (2000). Varieties of groups

and the perception of group entitativity.

Journal of Personality and Social Psychology,

78, 223--246.

Liden, R. C., & Maslyn, J. M. (1998). Multi-

dimensionality of leader-member exchange:

An empirical assessment through scale

development. Journal of Management, 24,

43--72.

Liden, R. C., Wayne, S. J., Stilwell, D.

(1993). A longitudinal study on the early

development of leader-member exchanges.


674.

Liebowitz, S. J., & DeMeuse, K. P. (1982).

The application of team building. Human

Relations, 35, 1--18.

Likert, R. (1961). New patterns of manage-

ment. New York: McGraw-Hill.

Likert, R. (1967). The human organization.

New York: McGraw-Hill.

Likert, J. G., & Araki, C. T. (1986). Man-

aging without a boss: System 5. LODJ, 7,

17--20.

Lim, S., & Cortina, L. M. (2005). Interper-

sonal mistreatment in the workplace: The

interface and impact of general incivility

and sexual harassment. Journal of Applied

Psychology, 90(3), 483--496.

Linville, P. W., Fischer, G. W., & Salovey,

P. (1989). Perceived distributions of the

characteristics of ingroup and outgroup

members: Empirical evidence and a com-

puter simulation. Journal of Personality and


Liu, C. (2003). A comparison of job stressors

and job strains among employees holding

comparable jobs in Western and Eastern soci-

eties. Unpublished doctoral dissertation,

University of South Florida, Tampa, FL.

Liu, C., & Spector, P. E. (2005). Interna-

tional and cross-cultural issues. In J. Barl-

ing, E. K. Kelloway, & M. R. Frone

(Eds.), Handbook of occupational health psy-

chology (pp. 487--516). Thousand Oaks,

CA: Sage.

Liu, C., Spector, P. E., & Jex, S. M.

(2005). The relation of job control with

job strains: A comparison of multiple data

sources. Journal of Occupational and Orga-

nizational Psychology, 78(3), 325--336.

References 555

Lievens, F., & Highhouse, S. (2003). A

relation of instrumental and symbolic

attributes to a company’s attractiveness as

an employer. Personnel Psychology, 56, 75--

102.

Locke, E. A. (1968). Toward a theory of

task motivation and incentive. Organiza-


157--189.

Locke, E. A. (1976). The nature and

causes of job satisfaction. In M. D. Dun-

nette (Ed.), Handbook of industrial and



Locke, E. A. (1982). The ideas of Freder-

ick W. Taylor: An evaluation. Academy of


Locke, E. A. (1986). Generalizing from

laboratory to field settings: Ecological val-

idity or abstraction of essential elements.

In E. A. Locke (Ed.), Generalizing from lab-

oratory to field studies (pp. 3--9). Lexington,

MA: Heath.

Locke, E. A., & Henne, D. (1986). Work

motivation theories. In C. L. Cooper & I.

T. Robertson (Eds.), International review of

industrial and organizational psychology

1986 (pp. 1--35). Chichester, England:

Wiley.

Locke, E. A., & Latham, G. P. (1990a). A

theory of goal setting and task performance.

Englewood Cliffs, NJ: Prentice-Hall.

Locke, E. A., & Latham, G. P. (1990b).

Work motivation and satisfaction: Light at

the end of the tunnel. Psychological Science,

4, 240--246.

Locke, E. A., & Latham, G. P. (2004). What

should we do about motivation theory? Six

recommendations for the twenty-first cen-

tury. Academy of Management Review, 29,

388--404.

Locke, E. A., & Latham, G. P. (2006).

New directions in goal-setting theory. Cur-

rent Directions in Psychological Science, 15,

265--268.

Lok, P., & Crawford, J. (2001). Antece-

dents of organizational commitment and

the mediating role of job satisfaction.

Journal of Managerial Psychology, 16, 594--

613.

Long, R. G., Bowers, W. P., Barnett, T., &

White, M. C. (1998). Research productiv-

ity in graduates in management: Effects of

academic origin and academy affiliation.


714.

Lord, R. G., & Hanges, P. J. (1987). A con-

trol system model of organizational motiva-

tion: Theoretical development and applied

implications. Behavioral Science, 32, 161--178.

Lord, R. G., & Hohenfeld, J. A. (1979).

Longitudinal field assessment of equity

aspects on the performance of major

league baseball players. Journal of Applied


Louis, M. R. (1983). Organizations as cul-

ture-bearing milieux. In L. R. Pondy, P. J.

Frost, G. Morgan, & T. Dandridge (Eds.),

556 References

Organizational symbolism (pp. 39--54).


Louis, M. R. (1990). Acculturation in the

workplace: Newcomers as lay ethnogra-

phers. In B. Schneider (Ed.), Organizational

climate and culture (pp. 85--129). San Fran-

cisco: Jossey-Bass.

Lowin, A. (1968). Participative decision

making: A model, literature critique, and

prescriptions for research. Organizational


106.

Luthans, F., & Kreitner, R. (1985). Organi-

zational behavior modification and beyond:

An operant and social learning approach

(2nd ed.). Glenview, IL: Scott, Foresman.

Maidani, E. A. (1991). Comparative study

of Herzberg’s two-factor theory of job sat-

isfaction among public and private sec-

tors. Public Personnel Management, 20,

441--448.

Majchrzak, A. (1987). Effects of manage-

ment policies on unauthorized absence

behavior. Journal of Applied Behavioral Sci-

ence, 23, 501--523.

Mann, R. D. (1959). A review of the rela-

tionships between personality and per-

formance in small groups. Psychological

Bulletin, 56, 241--270.

Margerison, C., & Glube, R. (1979). Lead-

ership decision-making: An empirical test

of the Vroom-Yetton model. Journal of

Management Studies, 16, 45--55.

Marks, M. A., DeChurch, L. A., Mathieu,

J. E., Panzer, F. J., & Alonso, A. (2005).

Teamwork in multi-team systems. Journal

of Applied Psychology, 90(5), 964--971.

Marks, M. A., Mathieu, J. E., & Zaccaro, S.

J. (2001). A temporally based framework

and taxonomy of team processes. Academy


Marrow, A. J. (1969). The practical theorist:

The life and work of Kurt Lewin. New York:

Basic Books.

Martin, J. (2002). Organizational culture: Map-

ping the terrain. Thousand Oaks, CA: Sage.

Martocchio, J. J. (1994). The effects of

absence culture on individual absence.

Human Relations, 47, 243--262.

Martocchio, J. J., & Harrison, D. A.

(1993). To be there or not to be there:

Questions, theories, and methods in

absenteeism research. Research in Personnel

and Human Resources Research, 11, 259--

329.

Maslow, A. H. (1943). A theory of human

motivation. Psychological Review, 50, 370--

396.

Mason, R. O. (2004). Lessons in organiza-

tional ethics from the Columbia disaster:

Can culture be lethal? Organizational

Dynamics, 33, 128--142.

Mathieu, J. E., & Kohler, S. S. (1990). A

cross-level examination of group absence

influences on individual absence. Journal of


References 557

Mathieu, J. E., & Zajac, D. M. (1990). A

review and meta-analysis of the antecedents,

correlates, and consequences of organiza-

tional commitment. Psychological Bulletin,

108, 171--194.

Matteson, M. T., & Ivancevich, J. M.

(1987). Controlling work stress. San Fran-

cisco: Jossey-Bass.

May, D. R., Reed, K., Schwoerer, C. E., &

Potter, P. (2004). Ergonomic office design

and aging: A quasi-experimental field

study of employee reactions to an ergo-

nomics intervention program. Journal of


135.

May, D. R., & Schwoerer, C. E. (1994).

Employee health by design: Using employee

involvement teams in ergonomic job design.


Mayo, E. (1933). The human problems of an

industrial civilization. New York: Macmillan.

McCabe, D. M. (1988). Corporate nonunion

complaint procedures and systems: A strategic

human resources management analysis. West-

port, CT: Praeger.

McCall, M. W., Jr., & Bobko, P. (1990).

Research methods in the service of discov-

ery. In M. D. Dunnette & L. M. Hough

(Eds.), Handbook of industrial and organiza-

tional psychology (2nd ed., Vol 1, pp. 381--

418). Palo Alto, CA: Consulting Psycholo-

gists Press.

McClelland, D. (1961). The achieving soci-

ety. Princeton, NJ: Van Nostrand.

McClelland, D. (1965). Toward a theory of

motive acquisition. American Psychologist,

20, 321--333.

McCloy, R. A., Campbell, J. P., & Cudeck,

R. (1994). A confirmatory test of a model

of performance determinants. Journal of


McDaniel, M. A., Schmidt, F. L., &

Hunter, J. E. (1988). Job experience corre-

lates of job performance. Journal of Applied


McEvoy, G. M., & Cascio, W. F. (1985).

Strategies for reducing employee turnover:

A meta-analysis. Journal of Applied Psychol-

ogy, 70, 342--353.

McEvoy, G. M., & Cascio, W. F. (1987).

Do good or poor performers leave? A

meta-analysis of the relationship between

performance and turnover. Academy of


McGrath, J. E. (1964). Social psychology: A

brief introduction. New York: Holt.

McGrath, J. E. (1990). Time matters in

groups. In J. Galegher, R. E. Kraut, & C.

Egido (Eds.), Intellectual teamwork: Social

and technological foundations of cooperative

work (pp. 23--61). Hillsdale, NJ: Erlbaum.

McGrath, J. E., & Beehr, T. A. (1990).

Time and the stress process: Some tempo-

ral issues in the conceptualization and

measurement of stress. Stress Medicine, 6,

93--104.

McGregor, D. (1960). The human side of

enterprise. New York: McGraw-Hill.

558 References

McKee-Ryan, F., Song, Z., Wanberg, C. R.,

& Kinicki, A. J. (2005). Psychological and

physical well-being during unemployment:

A meta-analytic study. Journal of Applied

Psychology, 90(1), 53--76.

McKenna, D. D., & Wright, P. M. (1992).

Alternative metaphors for organizational

design. In M. D. Dunnette & L. M. Hough

(Eds.), Handbook of industrial and organiza-

tional psychology (2nd ed., Vol. 3, pp.

901--960). Palo Alto, CA: Consulting Psy-

chologists Press.

McMahon, J. (1972). The contingency

theory: Logic and method revisited. Person-


McMillan-Capehart, A. (2005). A configu-

rational framework for diversity: Socializa-

tion and culture. Personnel Review, 34(4),

488--503.

McMullen, T. J. (1991). Personality corre-

lates of on-the-job substance use: Exploring

an alternative to urinalysis. Unpublished

master’s thesis, Central Michigan Univer-

sity, Mount Pleasant.

McNamara, R. S., Blight, J., Brigham, S.,

Biersteker, T., & Schandler, H. (1999).

Argument without end: In search of answers

to the Vietnam tragedy. New York: Public

Affairs.

McNeely, B. L., & Meglino, B. M. (1994).

The role of dispositional and situational

antecedents in prosocial organizational

behavior: An examination of the intended

beneficiaries of prosocial behavior. Journal


Meichenbaum, D. (1977). Cognitive-behavior

modification: An integrated approach. New York:

Plenum Press.

Merton, R. K. (1968). Social theory and

social structure. New York: Free Press.

Meyer, H. H., Kay, E., & French, J. R. P.,

Jr. (1965). Split roles in performance

appraisal. Harvard Business Review, 43,

123--129.

Meyer, J. P., & Allen, N. J. (1991). A three-

component conceptualization of organiza-

tional commitment. Human Resource Man-


Meyer, J. P., & Allen, N. J. (1997). Com-

mitment in the workplace: Theory, research,

and application. Thousand Oaks, CA:

Sage.

Michaels, C. E., & Spector, P. E. (1982).

Causes of employee turnover: A test of the

Mobley, Griffeth, Hand, and Meglino

model. Journal of Applied Psychology, 67,

53--59.

Miles, R. H. (1980). Macro organizational

behavior. Glenview, IL: Scott, Foresman.

Milgram, S. (1974). Obedience to authority.

New York: Harper & Row.

Milkovich, G. T., & Newman, J. (1990).

Compensation. Homewood, IL: Irwin.

Miller, V. D., & Jablin, F. M. (1991).

Information seeking during organizational

entry: Influences, tactics, and a model of

the process. Academy of Management

Review, 16, 92--120.

References 559

Milliken, F. J., Martins, L. L., & Morgan,

H. (1998). Explaining organizational res-

ponsiveness to work-family issues: The

role of human resource executives as inter-

preters. Academy of Management Journal,

41, 580--596.

Mirvis, P., & Seashore, S. (1979). Being

ethical in organizational research. American

Psychologist, 34, 766--780.

Mirvis, P. H. (2005). Large group interven-

tions: Change as theater. Journal of Applied

Behavioral Science, 41(1), 122--138.

Mitchell, T. R. (1974). Expectancy models

of job satisfaction, occupational preferen-

ces and effort: A theoretical, methodologi-

cal, and empirical appraisal. Psychological

Bulletin, 81, 1053--1077.

Mitchell, T. R., Holtom, B. C., Lee, T. W.,

Sablynski, C. J., & Erez, M. (2001). Why

people stay: Using job embeddedness to

predict voluntary turnover. Academy of

Management Journal, 44(6), 1102--1121.

Mitchell, T. R., & Kalb, L. S. (1982).

Effects of job experience on supervisor

attributions for a subordinate’s poor per-

formance. Journal of Applied Psychology, 67,

181--188.

Mitchell, T. R., & Silver, W. R. (1990).

Individual and group goals when workers

are interdependent: Effects on task strategy

and performance. Journal of Applied Psy-

chology, 75, 185--193.

Mobley, W. H. (1977). Intermediate link-

ages in the relationship between job satis-

faction and employee turnover. Journal of


Mobley, W. H., Griffeth, R. W., Hand, H.

H., & Meglino, B. M. (1979). Review and

conceptual analysis of the employee turn-

over process. Psychological Bulletin, 86,

493--522.

Mohamed, M. A. K. (2002). Assessing

determinants of departmental innovation.

Personnel Review, 31, 620--641.

Mohrman, S. A., & Quam, K. (2000).

Consulting to team-based organizations:

An organizational design and learning

approach. Consulting Psychology Journal:

Practice and Research, 52, 20--35.

Mone, M. A., Mueller, G. C., & Mauland,

W. (1996). The perceptions and usage of

statistical power in applied psychology and

management research. Personnel Psychology,

49, 103--120.

Mone, M. A., & Shalley, C. E. (1995).

Effects of task complexity and goal specif-

icity on change in strategy and perfor-

mance over time. Human Performance, 8,

243--252.

Monroy, J., Jonas, H., Mathey, J., & Mur-

phy, L. (1998). Holistic stress management

at Corning, Incorporated. In M. K. Gow-

ing, J. D. Kraft, & J. C. Quick (Eds.), The

new organizational reality: Downsizing,

restructuring, and revitalization. Washington,

DC: American Psychological Association.

Moorhead, G., & Griffin, R. W. (1998).

Organizational behavior: Managing people and

560 References

organizations (5th ed.). Boston: Houghton

Mifflin.

Moorman, R. H. (1991). Relationship

between organizational justice and organi-

zational citizenship behaviors: Do fairness

perceptions influence employee citizen-

ship? Journal of Applied Psychology, 76,

845--855.

Moreland, R. L., & Levine, J. M. (1982).

Socialization in small groups: Temporal

changes in individual-group relations.

Advances in Experimental Social Psychology,

15, 137--192.

Moreno, J. L. (1953). Who shall survive? A

new approach to the problem of human inter-

relations (revised ed). Washington, DC:

Nervous and Mental Disease Publishing Co.

Morgan, G. (1986). Images of organization.

Beverly Hills, CA: Sage.

Morgeson, F. P., & Campion, M. A.

(2002). Minimizing tradeoffs when rede-

signing work: Evidence from a longitudinal

quasi-experiment. Personnel Psychology, 55,

589--612.

Morgeson, F. P., & Humphrey, S. E.

(2006). The work design questionnaire

(WDQ): Developing and validating a com-

prehensive measure for assessing job

design and the nature of work. Journal of


Morrison, E. W. (1993). Longitudinal

study of the effects of information seeking

on newcomer socialization. Journal of


Morrison, E. W. (1994). Role definitions

and organizational citizenship behavior:

The importance of the employee’s perspec-

tive. Academy of Management Journal, 37,

1543--1567.

Morrison, E. W. (2002). Newcomers’ rela-

tionships: The role of social network ties

during socialization. Academy of Management

Journal, 45, 1149--1160.

Morrison, E. W., & Robinson, S. L.

(1997). When employees feel betrayed: A

model of how psychological contract viola-

tion develops. Academy of Management

Review, 22, 226--256.

Moscovici, S. (1994). Three concepts:

Minority, conflict, and behavioral styles. In

S. Moscovici, A. Mucchi-Faina, & A. Maas

(Eds.), Minority influence (pp. 233--251).

Chicago: Nelson-Hall.

Motowidlo, S. J., Packard, J. S., & Manning,

M. R. (1986). Occupational stress: Its causes

and consequences for job performance. Jour-


Mowday, R. T., Porter, L. W., & Steers,

R. M. (1982). Organizational linkages: The

psychology of commitment, absenteeism, and

turnover. San Diego, CA: Academic Press.

Mowday, R. T., Steers, R. M., & Porter,

L. W. (1979). The measurement of orga-

nizational commitment. Journal of Voca-

tional Behavior, 14, 224--247.

Muchinsky, P. M. (1977). A comparison of

within- and across-subjects analyses of the

expectancy-value model for predicting

References 561

effort. Academy of Management Journal, 20,

154--158.

Muchinsky, P. M., & Morrow, P. C.

(1980). A multidisciplinary model of

employee turnover. Journal of Vocational

Behavior, 17, 263--290.

Mudrack, P. E. (1989). Defining group

cohesiveness: A legacy of confusion. Small

Group Behavior, 20, 37--49.

Mulcahy, C. (1991). Workplace stress

reaches ‘‘epidemic’’ proportion. National

Underwriter, 4, 20.

Mullen, B., & Cooper, C. (1994). The rela-

tion between group cohesion and perfor-

mance: An integration. Psychological Bulletin,

115, 210--227.

Mumford, M. D., & Licuanan, B. (2004).

Leading for innovation: Conclusions,

issues, and directions. The Leadership Quar-

terly, 15, 163--171.

Murphy, K. R. (1989a). Dimensions of job

performance. In R. Dillion & J. W. Pelligrino

(Eds.), Testing: Theoretical and applied perspec-

tives (pp. 218--247). New York: Praeger.

Murphy, K. R. (1989b). Is the relationship

between cognitive ability and performance

stable over time? Human Performance, 2,

183--200.

Murphy, K. R. (1994). Toward a broad

conceptualization of jobs and job perfor-

mance: Impact of changes in the military

environment on the structure, assessment,

and prediction of job performance. In

M. G. Rumsey, C. B. Walker, & J. H. Har-

ris (Eds.), Personnel selection and classifica-

tion (pp. 85--102). Hillsdale, NJ: Erlbaum.

Murphy, K. R. (2004). Assessment in work

settings. In S. N. Haynes & E. M. Heiby

(Eds.), Comprehensive handbook of psycho-

logical assessment, Vol. 3: Behavioral Assess-

ment (pp. 346--364). Hoboken, NJ: Wiley

Publishers.

Murphy, K. R., & Cleveland, J. N. (1990).

Performance appraisal: An organizational

perspective. Boston: Allyn & Bacon.

Murphy, K. R., Thornton, G. C., III., &

Reynolds, D. H. (1990). College students’

attitudes toward employee drug testing

programs. Personnel Psychology, 43, 615--

631.

Murphy, L. R. (1984). Occupational stress

management: A review and appraisal. Jour-

nal of Occupational Psychology, 57, 1--15.

Murphy, L. R. (1996). Future directions for

job stress research and practice: Expanding

the focus from worker health to organiza-

tional health. Opening keynote speech at

the 2nd National Occupational Stress,

Conference 1996, Brisbane, Queensland,

Australia.

Murray, H. A. (1938). Explorations in per-

sonality. New York: Oxford University

Press.

Nadler, D. A., & Tushman, M. L. (1992).

Designing organizations that have good

fit: A framework for understanding new

architectures. In D. A. Nadler, M. Gerstein,

& R. B. Shaw (Eds.), Organizational

562 References

architecture (pp. 39--56). San Francisco:

Jossey-Bass.

Nadler, D. A., & Tushman, M. L. (1997).

Competing by design: The power of organiza-

tional architecture. New York: Oxford Uni-

versity Press.

Nagel, C. M. (1999). The perception of fit

within the interview process. Unpublished

master’s thesis, University of Wisconsin,

Oshkosh.

National Council on Compensation Insur-

ance. (1988). Emotional stress in the work-

place: New legal rights in the eighties. New

York: Author.

National Council on Compensation Insur-

ance. (1991). Issues report, 1991. Boca

Raton, FL: Author.

Neal, A., & Griffin, M. A. (2006). A study

of the lagged relationships among safety

climate, safety motivation, safety behavior,

and accidents at the individual and group

levels. Journal of Applied Psychology, 91(4),

946--953.

Nelson, D. L., & Simmons, B. L. (2003).

Deriving benefits from stressful events:

The role of engagement in meaningful

work and hardiness. In J. C. Quick &

L. E. Tetrick (Eds.), Handbook of occupa-

tional health psychology (pp. 97--119).

Washington, DC: American Psychological

Association.

Nelson, T. D. (1993). The hierarchical

organization of behavior: A useful feed-

back model of self-regulation. Current

Directions in Psychological Science, 2, 121--

126.

Neuman, G. A., Edwards, J. E., & Raju,

N. S. (1989). Organizational development

interventions: A meta-analysis of their

effects on satisfaction and other attitudes.


Ng, E. S. W., & Burke, R. J. (2005).

Person-organization fit and the war for

talent: Does diversity management make

a difference. International Journal of

Human Resources Management, 16(7),

1195--1210.

Nielsen, I. K., Jex, S. M., & Adams, G. A.

(2000). Development and validation of

scores on a two-dimensional workplace

friendship scale. Educational and Psychologi-

cal Measurement, 60, 628--643.

Nunnally, J. C., & Bernstein, I. H. (1994).

Psychometric theory (3rd ed.). New York:

McGraw-Hill.

O’Donnell, M. P. (1986). Definition of

health promotion. Part II: Levels of pro-

grams. American Journal of Health Promo-

tion, 1, 6--9.

Offermann, L. R., & Spiros, R. K. (2001).

The science and practice of team develop-

ment: Improving the link. Academy of


Oh, H., Chung, M. H., & Labianca, G.

(2004). Group social capital and group

effectiveness: The role of informal socializing

ties. Academy of Management Journal, 47(6),

860--875.

References 563

O’Leary-Kelly, A. M., Griffin, R. W., Glew,

D. J. (1996). Organization-motivated aggres-

sion: A research framework. Academy of


O’Leary-Kelly, A. M., Martocchio, J. J., &

Frink, D. D. (1994). A review of the influ-

ence of group goals on group performance.


1301.

Ones, D. S., & Viswesvaran, C. (2000).

Most published authors in Journal of

Applied Psychology and Personnel Psy-

chology during the 1990s. Industrial-

Organizational Psychologist, 37, 26--35.

Ones, D. S., Viswesvaran, C., & Schmidt,

F. L. (1993). Comprehensive meta-analysis

of integrity test validities: Findings and

implications for personnel selection and

theories of job performance. Journal of

Applied Psychology Monograph, 78, 679--703.

O’Reilly, C. A., & Caldwell, D. (1979).

Information influences as a determinant of

task characteristics and job satisfaction.


O’Reilly, C. A., Chatman, J., & Caldwell,

D. (1991). People and organizational cul-

ture: A profile comparison approach to

assessing person-organization fit. Academy

of Management Journal, 34(3), 487--516.

Organ, D. W. (1977). A reappraisal and

reinterpretation of the satisfaction-causes-

performance hypothesis. Academy of Man-


Organ, D. W. (1994). Organizational citi-

zenship behavior and the good soldier. In

M. G. Rumsey, C. B. Walker, & J. H. Har-

ris (Eds.), Personnel selection and classifica-

tion (pp. 53--67). Hillsdale, NJ: Erlbaum.

Organ, D. W., & Konovsky, M. (1989).

Cognitive versus affective determinants of

organizational citizenship behavior. Journal


Organ, D. W., & Ryan, K. (1995). A meta-

analytic review of attitudinal and disposi-

tional predictors of organizational citizen-

ship behavior. Personnel Psychology, 48,

775--802.

Ostroff, C. (1992). The relationship bet-

ween satisfaction, attitudes, and perform-

ance: An organizational level analysis.


Ostroff, C., & Kozlowski, S. W. J. (1992).

Organizational socialization as a learning

process: The role of information acquisi-

tion. Personnel Psychology, 45, 849--874.

O’Toole, L. J., & Meier, K. J. (2003).

Bureaucracy and uncertainty. In B. C. Bur-

den (Ed.), Uncertainty in American politics

(pp. 98--117). Cambridge, UK: Cambridge

University Press.

Pacanowsky, M. E., & O’Donnell-Trujillo,

N. (1983). Organizational communication

as cultural performance. Communication

Monographs, 50, 126--147.

Page, R. C., Davis, K. C., Berkow, D. N.,

O’Leary, E. (1989). Analysis of group

process in marathon group therapy with

users of illicit drugs. Small Group Behavior,

20, 220--227.

564 References

Parker, C. P. (1999). The impact of lead-

ers’ implicit theories of employee participa-

tion on tests of the Vroom-Yetton model.

Journal of Social Behavior and Personality, 4,

45--61.

Parker, S., & Wall, T. (1998). Job and work

redesign: Organizing work to promote well-

being and effectiveness. Thousand Oaks, CA:

Sage.

Parsons, T. (1951). The social system. New

York: Free Press.

Paul, R. J., & Ebadi, Y. M. (1989). Leader-

ship decision-making in a service organization:

A field test of the Vroom-Yetton model.

Journal of Occupational Psychology, 62, 201--

211.

Pearlman, J. (2006). Love me, hate me:

Barry Bonds and the making of an antihero.

HarperCollins.

Peeters, M. A. G., Van Tuijl, H. F. J. M.,

Rutte, C. G., & Reymen, I. M. M. J.

(2006). Personality and team performance:

A meta-analysis. European Journal of Per-

sonality, 20, 377--396.

Pelletier, K. R. (1991). A review and anal-

ysis of the health and cost effective out-

come studies of comprehensive health

promotion and disease preventive pro-

grams. American Journal of Health Promo-

tion, 5, 311--315.

Perkins, L. A., Thomas, K. M., & Taylor,

G. A. (2000). Advertising and recruitment:

Marketing to minorities. Psychology & Mar-

keting, 17(3), 235--255.

Perlaki, I. (1994). Organizational develop-

ment in Eastern Europe: Learning to build

culture-specific OD theories. Journal of


Perlman, B., McCann, L. I., & McFadden,

S. H. (1999). How to land that first teach-

ing job. In B. Perlman, L. I. McCann, and

S. H. McFadden (Eds.), Lessons learned:

Practical advice for the teaching of psychology

(Vol 1. pp. 3--6). Washington, DC: The

American Psychological Association.

Perrewe, P. L., & Ganster, D. C. (1989).

The impact of job demands and behavioral

control on experienced job stress. Journal

of Organizational Behavior, 10, 213--229.

Peter, R., Geissler, H., & Siegrist, J. (1998).

Associations of effort-reward imbalance at

work and reported symptoms in difference

groups of male and female public transport

workers. Stress Medicine, 14, 175--182.

Peters, L. H., & O’Connor, E. J. (1980).

Situational constraints and work outcomes:

The influences of a frequently overlooked

construct. Academy of Management Review,

5, 391--397.

Peters, L. H., & O’Connor, E. J. (1988).

Measuring work obstacles: Procedures,

issues, and implications. In F. D. Schoor-

man & B. Schneider (Eds.), Facilitating

work group effectiveness (pp. 105--123).

Lexington, MA: Lexington Books.

Peters, T. J., & Waterman, R. H. (1982).

In search of excellence: Lessons from Ameri-

ca’s best-run companies. New York: Harper

& Row.

References 565

Peterson, M., & Wilson, J. (1998). A

culture-work-health model. American Jour-

nal of Health Behavior, 22, 378--390.

Peterson, M. F., Smith, P. B., Akande, A.,

Ayestaran, S., Bocher, S., Callan, V., et al.

(1995). Role conflict, ambiguity, and over-

load: A 21-nation study. Academy Manage-


Peterson, S. J., & Luthans, F. (2006).

The impact of financial and nonfinancial

incentives on business-unit outcomes

over time. Journal of Applied Psychology,

91, 156--165.

Pettigrew, A. M. (1979). On studying

organizational cultures. Administrative Sci-


Petty, M. M., Singleton, B., & Connell,

D. W. (1992). An experimental evaluation

of an organizational incentive plan in the

electrical utility industry. Journal of Applied


Petty, R. E., & Cacioppo, J. T. (1981).

Attitudes and persuasion: Classic and contem-

porary approaches. Dubuque, IA: Brown.

Piccolo, R. F., & Colquitt, J. A. (2006).

Transformational leadership and job

behaviors: The mediating role of core job

characteristics. Academy of Management

Journal, 49, 327--340.

Piderit, S. K. (2005). Rethinking resistance

and recognizing ambivalence: A multidi-

mensional view of attitudes toward an

organizational change. Academy of Manage-

ment Review, 25(4), 783--794.

Pierce, J. L., & Dunham, R. B. (1992). The

12-hour work day: A 48-hour, eight day

week. Academy of Management Journal, 35,

1086--1098.

Pierce, J. L., & Newstrom, J. W. (1982).

Employee responses to flexible work

schedules: An inter-organization, inter-

system comparison. Journal of Management,

8, 9--25.

Pierce, J. L., Newstrom, J. W., Dunham, R.

B., & Barber, A. E. (1989). Alternative work

schedules. Needham Heights, MA: Allyn &

Bacon.

Pinder, C. C. (1998). Work motivation in

organizational behavior. Upper Saddle River,

NJ: Prentice-Hall.

Ployhart, R. E., & Hakel, M. D. (1998). The

substantive nature of performance variabil-

ity: Predicting interindividual differences in

intraindividual performance. Personnel Psy-

chology, 51, 859--901.

Ployhart, R. E., Weekly, J. A., & Baughman,

K. (2006). The structure and function of hu-

man capital emergence: A multilevel exami-

nation of the attraction-selection-attrition

model. Academy of Management, 49, 661--

677.

Podsakoff, P. M. (1982). Determinants of

a supervisor’s use of rewards and punish-

ments: A literature review and suggestions

for further research. Organizational Behavior

and Human Performance, 29, 58--83.

Podsakoff, P. M., Ahearne, M., & MacKenzie,

S. B. (1997). Organizational citizenship

behavior and the quantity and quality of

566 References

work group performance. Journal of Applied


Podsakoff, P. M., MacKenzie, S. B., Lee,

J. Y., & Podsakoff, N. M. (2003). Com-

mon method biases in behavioral research:

A critical review of the literature and rec-

ommended remedies. Journal of Applied

Psychology, 88(5), 879--903.

Podsakoff, P. M., & Williams, L. J.

(1986). The relationship between job per-

formance and job satisfaction. In E. A.

Locke (Ed.), Generalizing from laboratory to

field settings (pp. 207--253). Lexington,

MA: Heath.

Porras, J. I., & Robertson, P. J. (1992).

Organizational development: Theory, prac-

tice, and research. In M. D. Dunnette &

L. M. Hough (Eds.), Handbook of industrial

and organizational psychology (2nd ed.,

Vol. 3, pp. 719--822). Palo Alto, CA: Con-

sulting Psychologists Press.

Porras J. I., & Silvers, R. C. (1991). Orga-

nization development and transformation.

Annual Review of Psychology, 42, 51--78.

Powell, C. L. (1995). My American journey.

New York: Ballantine Books.

Powers, W. T. (1973a). Feedback: Beyond

behaviorism. Science, 179, 351--356.

Powers, W. T. (1973b). Behavior: The con-

trol of perception. Chicago: Aldine.

Powers, W. T. (1978). Quantitative analy-

sis of purposive systems: Some spadework

at the foundations of scientific psychology.

Psychological Review, 85, 417--435.

Pritchard, R. D. (1969). Equity theory as a

predictor of productivity and work quality.

Psychological Bulletin, 70, 596--610.

Pritchard, R. D. (1992). Organizational

productivity. In M. D. Dunnette & L. M.





Puffer, S. M. (1999). Global statesman:

Mikhail Gorbachev on globalization. Acad-

emy of Management Executive, 13, 8--14.

Pulakos, E. D. (1984). A comparison of

rater training programs: Error training and

accuracy training. Journal of Applied Psy-

chology, 69, 581--588.

Puranam, P., Singh, H., & Zollo, M.

(2006). Organizing for innovation: Manag-

ing the coordination-autonomy dilemma in

technology acquisitions. Academy of Man-


Purser, R. E., & Petranker, J. (2005).

Unfreezing the future: Exploring the dynamic

of time in organizational change. Journal of


Purvanova, R. K., Bono, J. E., & Dziewec-

zynski, A. (2006). Transformational leader-

ship, job characteristics, and organizational

citizenship performance. Human Perfor-

mance, 19, 1--22.

Quinones, M. A., Ford, J. K., & Teachout,

M. S. (1995). The relationship between

work experience and job performance: A

conceptual and meta-analytic review. Per-


References 567

Rao, T. V., & Vijayalakshmi, M. (2000).

Organization development in India. Orga-

nization Development Journal, 18, 51--63.

Rau, R. (2004). Job strain or healthy work:

A question of task design. Journal of Occu-

pational Health Psychology, 9(4), 322--338.

Raven, B. H. (1993). The bases of power:

Origins and recent developments. Human

Relations, 49, 227--251.

Ravisi, D., & Schultz, M. (2006). Responding

to organizational identity threats: Exploring

the role of organizational culture. Academy of


Reid, G. L. (1972). Job search and the effec-

tiveness of job-finding methods. Industrial

and Labor Relations Review, 25, 479--495.

Reilly, R. R., Brown, B., Blood, M. R., &

Malatesta, C. (1981). The effect of realistic

previews: A study and discussion of the liter-

ature. Personnel Psychology, 34, 823--834.

Renn, R. W., & Vandenberg, R. J. (1995).

The critical psychological states: An under-

represented component in job characteris-

tics model research. Journal of Management,

21, 279--303.

Rentsch, J. R., & Schneider, B. (1991).

Expectations for postcombination organiza-

tional life: A study of responses to merger

and acquisition scenarios. Journal of

Applied Social Psychology, 21, 233--252.

Riggs, M. L., & Knight, P. A. (1994). The

impact of perceived group success-failure

on motivational beliefs and attitudes: A

causal model. Journal of Applied Psychology,

79, 755--766.

Riordan, C. M., & Griffeth, R. W. (1995).

The opportunity for friendship in the work-

place: An underexplored construct. Journal

of Business and Psychology, 10, 141--154.

Roberts, C. K. (1995). The role of person-

ality in perceived free riding. Unpublished

doctoral dissertation, Central Michigan

University, Mount Pleasant.

Roccas, S., & Schwartz, S. H. (1993).

Effects of intergroup similarity on inter-

group relations. European Journal of Social


Rodgers, R., & Hunter, J. E. (1991). Impact

of management by objectives on organiza-

tional productivity. Journal of Applied Psy-

chology, 76, 322--336.

Rokeach, M. (1973). The nature of human

values. New York: Free Press.

Roman, P. M., & Blum, T. C. (1995).

Employers. In R. H. Coombs & D. M. Ziedonis

(Eds.), Handbook on drug abuse prevention

(pp. 139--158). Boston: Allyn & Bacon.

Ronen, S. (2001). Self-actualization versus

collectualization: Implications for motiva-

tion theories. In M. Erez, U. Kleinbeck, &

H. Thierry (Eds.), Work motivation in the

context of a globalized economy. Mahwah,

NJ: Erlbaum.

Rosen, C., Klein, K. J., & Young, K. M.

(1986). Employee ownership in America: The

equity solution. Lexington, MA: Lexington.

Rosen, T. H. (1987). Identification of sub-

stance abusers in the workplace. Public

Personnel Management, 16, 197--205.

568 References

Rosenthal, R. (1991). Meta-analytic techni-

ques for social research (Rev. ed.). Newbury

Park, CA: Sage.

Ross, L. (1977). The intuitive psychologist

and his shortcomings: Distortions in the

attribution process. In L. Berkowitz (Ed.),

Advances in experimental social psychology.


Rotter, J. B. (1966). Generalized expectan-

cies for internal versus external control of

reinforcement. Psychological Monographs

(Entire issue, No. 609).

Rousseau, D. (1985). Issues of level in

organizational research: Multi-level and

cross-level perspectives. In L. L. Cum-

mings & B. M. Staw (Eds.), Research in

organizational behavior (Vol. 7, pp. 1--38).


Runkel, P. J., & McGrath, J. E. (1972).

Research on human behavior. New York:

Holt, Reinhart and Winston.

Russo, M. V., & Harrison, N. S. (2005).

Organizational design and environmental

performance: Clues from the electronics

industry. Academy of Management Journal,

48, 582--583.

Ryan, A. M., Chan, D., Ployhart, R. E., &

Slade, L. A. (1999). Employee attitude sur-

veys in a multinational organization: Con-

sidering language and culture in assessing

measurement equivalence. Personnel Psy-

chology, 52, 37--58.

Ryan, R. M., & Deci, E. L. (2000). Self-

determination theory and the facilitation of

intrinsic motivation. American Psychologist,

55, 68--78.

Ryan, R. M., & Deci, E. L. (2001). On

happiness and human potentials: A review

of research on hedonic and eudaimonic

well-being. Annual Review of Psychology, 52,

141--166.

Rynes, S. L. (1991). Recruitment, job

choice, and post hire consequences: A call

for new research directions. In M. D. Dun-



ed., Vol. 2, pp. 399--444). Palo Alto, CA:

Consulting Psychologists Press.

Rynes S. L., Bretz, R. D., Jr. & Gerhart, B.

(1991). The importance of recruitment in

job choice: A different way of looking. Per-


Sackett, P. R., & Larsen, J. R., Jr. (1990).

Research strategies and tactics in industrial

and organizational psychology. In M. D.



(2nd ed., Vol.1, pp. 419--490). Palo Alto,


Salancik, G. R., & Pfeffer, J. (1977). An

examination of need satisfaction models of

job attitudes. Administrative Science Quar-

terly, 22, 427--456.

Salancik, G. R., & Pfeffer, J. (1978). A

social information processing approach to

job attitudes and task design. Administra-

tive Science Quarterly, 23, 224--253.

Salas, E., Sims, D. E., & Burke, C. S.

(2005). Is there a ‘‘big five’’ in teamwork?

Small Group Research, 36(5), 555--599.

Sales, S. M., & House, J. (1971). Job dis-

satisfaction as a possible risk factor in

References 569

coronary heart disease. Journal of Chronic

Diseases, 23, 189--194.

Salvendy, G. (1978). An industrial engineer-

ing dilemma: Simplified versus enlarged

jobs. In R. Muramatsu & N. A. Dudley

(Eds.),. Production and industrial systems

(pp. 965--975). London: Taylor & Francis.

Scandura T. A., & Williams, E. A. (2000).

Research methodology in management:

Current practices, trends, and implications

for future research. Academy of Manage-

ment Journal, 43, 1248--1264.

Scarpello, V., & Jones, F. F. (1996). Why

justice matters in compensation decision

making. Journal of Organizational Behavior,

17, 285--299.

Schacter, S. (1959). The psychology of affili-

ation. Stanford, CA: Stanford University

Press.

Schaffer, B. S., & Riordan, C. M. (2003).

A review of cross-cultural methodologies

for organizational research: A best practice

approach. Organizational Research Methods,

6(2), 169--215.

Schat, A. C. H., & Kelloway, K. E. (2005).

Reducing the adverse consequences of

workplace aggression and violence: The

buffering effects of organizational support.

Journal of Occupational Health Psychology,

8(2), 110--122.

Schaubroeck, J., Ganster, D. C., & Jones,

J. R. (1998). Organization and occupation

influences in the attraction-selection-attri-

tion process. Journal of Applied Psychology,

83, 869--891.

Schaubroeck, J., Ganster, D. C., & Kem-

merer, B. E. (1994). Job complexity, Type

A behavior, and cardiovascular disorder: A

prospective study. Academy of Management

Journal, 34, 966--975.

Schaubroeck, J., Ganster, D. C., Sime, W.

E., & Ditman, D. (1993). A field experi-

ment testing supervisory role clarification.


Schaubroeck, J. & Keuhn, K. (1992).

Research design in industrial and organiza-

tional psychology. In C. L. Cooper and

I. T. Robertson (Eds.), International review


1991 (pp. 99--121). Chichester, England:

Wiley.

Schaubroeck, J., & Merritt, D. E. (1997).

Divergent effects of job control on coping

with work stressors: The key role of self-

efficacy. Academy of Management Journal,

40, 738--754.

Schein, E. H. (1969). Process consultation:

Its role in organization development. Read-

ing, MA: Addison-Wesley.

Schein, E. H. (1983). The role of the

founder in creating organizational cul-

ture. Organizational Dynamics, Summer,

13--28.

Schein, E. H. (1985). Organizational culture

and leadership: A dynamic view. San Fran-

cisco: Jossey-Bass.

Schein, E. H. (1987). Process consultation:

Lessons for managers and consultants (Vol.

2). Reading, MA: Addison-Wesley.

570 References

Schein, E. H. (1990). Organizational cul-

ture. American Psychologist, 45, 109--119.

Schein, E. H. (1992). Organizational culture

and leadership: A dynamic view (2nd ed.).


Schleicher, D. J., Watt, J. D., & Greguras,

G. J. (2004). Reexamining the job

satisfaction-performance relationship. Jour-


Schmidt, F. L., & Hunter, J. E. (1998).

The validity and utility of selection meth-

ods in personnel psychology: Practical and

theoretical implications of 85 years of

research findings. Psychological Bulletin,

124, 262--274.

Schmidt, F. L., Hunter, J. E., & Outer-

bridge, A. N. (1986). The impact of job

experience and ability on job knowledge,

work sample performance, and supervisory

ratings of performance. Journal of Applied


Schminke, M., Ambrose, M. L., & Cropan-

zano, R. S. (2000). The effect of organiza-

tional structure on perceptions of

procedural fairness. Journal of Applied Psy-

chology, 85, 294--304.

Schmitt, N. S. (2004). Beyond the Big Five:

In creases in understanding and practical

utility. Human Performance, 17, 347--357.

Schneider, B. (1987). The people make the

place. Personnel Psychology, 40, 437--454.

Schneider, B., Smith, D. B., Taylor, S., &

Fleenor, J. (1998). Personality and organi-

zations: A test of the homogeneity of per-

sonality hypothesis. Journal of Applied


Schneider, B., Hanges, P. J., Smith, D. B.,

& Salvaggio, A. N. (2003). Which comes

first: Employee attitudes or organizational

financial and market performance? Journal


Schneider, J., & Cook, K. (1995). Status

inconsistency and gender: Combining

revisited. Special issue: Extending interaction

theory. Small Group Research, 26, 372--

399.

Schopler, J., Insko, C. A., Wieselquist, J.,

Pemberton, M., Witcher, B., Kozar, R.,

Roddenberry, C., & Wildschut, T. (2001).

When groups are more competitive than

individuals: The domain of the discontinu-

ity effect. Journal of Personality and Social


Schriesheim, C. A., & Kerr, S. (1977).

Theories and measures of leadership: A

critical appraisal of current and future

directions. In J. G. Hunt & L. L. Larson

(Eds.), Leadership: The cutting edge (pp. 9--

45). Carbondale: Southern Illinois Univer-

sity Press.

Schriescheim C. A., Tepper, B. J., & Tet-

rault, L. A. (1994). Least preferred co-

worker score, situational control, and lead-

ership effectiveness: A meta-analysis of

contingency model performance predic-

tions. Journal of Applied Psychology, 79,

561--573.

Schuler, M. (2004). Management of the

organizational image: A method for

References 571

organizational image configuration. Corpo-

rate Reputation Review, 7(1), 37--53.

Schutz, W. C. (1958). FIRO: A three-

dimensional theory of interpersonal behavior.

New York: Rinehart.

Schwab, D. P. (1991). Contextual variables

in employer performance-turnover rela-

tionships. Academy of Management Journal,

34, 966--975.

Schwartz, S. H. (1994). Are there universal

aspects in the structure and content of

human values?, Journal of Social Issues, 50,

19--45.

Schwartz, P., & Ogilvy, J. (1998). Plotting

your scenarios. In L. Fahey & R. Tandall

(Eds.), Learning from the future (pp. 57--

80). New York: John Wiley.

Schweiger, D. M., & DeNisi, A. S. (1991).

Communication with employees following

a merger: A longitudinal field experiment.


135.

Scott, W. R. (1990). Technology and struc-

ture: An organizational-level perspective. In

P. S. Goodman & L. S. Sproull (Eds.),

Technology and organizations (pp. 109--143).


Sears, W. (1986). Winning the productiv-

ity battle: Roles and responsibilities for

HRD practitioners. Leadership & Organiza-

tional Development Journal, 7(1), 17--20.

Seashore, S. E. (1954). Group cohesiveness

in the industrial work group. Ann Arbor:

University of Michigan Press.

Seers, A., Petty, M. M., Cashman, J. F.

(1995). Team-member exchange under

team and traditional management: A natu-

rally occurring quasi-experiment. Group

and Organization Management, 20, 18--38.

Selye, H. (1956). The stress of life. New

York: McGraw-Hill.

Shamir, B., House, R. J., & Arthur, M. B.

(1993). The motivational effect of charis-

matic leadership: A self-concept based

theory. Organization Science, 4, 577--594.

Shaw, J. D., Duffy, M. K., Mitra, A., Lock-

hart, D. E., & Bowler, M. (2003). Reac-

tions to merit pay increases: A longitudinal

test of a signal sensitivity perspective. Jour-


Shaw, M. E. (1964). Communication net-

works. Advances in Experimental Social Psy-

chology, 1, 111--147.

Shaw, M. E. (1978). Communication net-

works fourteen years later. In L. Berkowitz

(Ed.), Group processes. New York: Aca-

demic Press.

Shaw, I. F., Greene, J. C., & Mark, M. M.

(Eds.). (2006). The sage handbook of evalua-

tion. Beverly Hills, CA: Sage.

Shea, G. P., & Guzzo, R. A. (1987).

Groups as human resources. In K. M.

Rowland & G. R. Ferris (Eds.), Research in

personnel and human resources management

(Vol. 5, pp. 323--356). Greenwich, CT: JAI

Press.

Sheldon, K. M., & Elliot, A. J. (1998). Not

all personal goals are personal: Comparing

572 References

autonomous and controlled reasons for

goals as predictors of effort and attain-

ment. Personality and Social Psychology Bul-

letin, 24, 546--557.

Sheldon, K. M., & Houser-Marko, L.

(2001). Self-concordance, goal attainment,

and the pursuit of happiness: Can there be

an upward spiral? Journal of Personality and


Sheldon, K. M., Turban, D. B., Brown, K.

G., Barrick, M. R., & Judge, T. A. (2003).

Apply self-determination theory to organi-

zational research. Research in Personnel and

Human Resources Management, 22, 357--

393.

Sherif, M. (1966). In common predicament:

Social psychology of intergroup conflict and

cooperation. Boston: Houghton Mifflin.

Sherif, M., Harvey, O. J., White, B. J.,

Hood, W. R., & Sherif, C. W. (1961).

Intergroup conflict and cooperation: The Rob-

bers Cave experiment. Norman, OK: Insti-

tute of Group Relations.

Shim, W., & Steers, R. M. (1994). Media-

ting influences on the employee commitment-

job performance relationship. Unpublished

manuscript.

Shperling, Z. & Shirom, A. (2005). A field

experiment assessing the impact of the

focused diagnosis intervention on job

autonomy. Journal of Applied Behavioral Sci-

ence, 41(2), 222--240.

Si, S. X., & Hitt, M. A. (2004). A study of

organizational image resulting from inter-

national joint ventures in transitional

economies. Journal of Business Research, 57,

1370--1377.

Siebold, G. (2006). Military group cohe-

sion. In T. W. Britt, C. A. Castro, & A. B.

Adler (Eds.), Military life: The psychology of

serving in peace and combat. Vol. 1: Military

performance (pp. 185--201). Westport, CT:

Praeger Security International.

Siegrist, J. (1996). Adverse health effects of

high-effort/low-reward conditions. Journal

of Occupational Health Psychology, 1, 27--

41.

Siegrist, J. (2002). Effort-reward imbalance

at work and health. In P. L. Perrewe & D.

C. Ganster (Eds.), Historical and current

perspectives on stress and health (pp. 261--

291). Amsterdam: JAI Elsevier.

Siegrist, J., Starke, D., Chandola, T.,

Godin, I., Marmot, M., Niedhammer, I., &

Richard, P. (2004). The measurement of

effort-reward imbalance at work: European

comparisons. Social Science & Medicine, 58,

1483--1499.

Silvester, J., Anderson, N. R., & Patterson,

F. (1999). Organizational culture change:

An inter-group attributional analysis. Jour-

nal of Occupational and Organizational Psy-

chology, 72, 1--23.

Since, W. D., Mitsuhashi, H., & Kirsch, D.

(2006). Revisiting Burns and Salker: For-

mal structure and new venture perfor-

mance in emerging economic sectors.


132.

References 573

Sinclair, R. R., Tucker, J. S., Cullen, J. C.,

& Wright, C. (2005). Performance differ-

ences among four organizational commit-

ment profiles. Journal of Applied Psychology,

90, 1280--1287.

Skibba, J. S., & Tan, J. A. (2004). Person-

ality predictors of firefighter job perfor-

mance and job satisfaction. Applied H. R.

M. Research, 9, 39--40.

Skinner, B. F. (1971). Beyond freedom and

dignity. New York: Knopf.

Skogstad, A., Einarsen, S., Torsheim, T.,

Aasland, M. S., & Hetland, H. (2007). The

destructiveness of laissez-faire leadership

behavior. Journal of Occupational Health


Slaughter, J. E., & Zickar, M. J. (2006). A

new look at the roles of insiders in the

newcomer socialization process. Group and

Organization Management, 31(2), 264--290.

Small, M. (1999). The presidency of Richard

Nixon. Lawrence: University of Kansas

Press.

Smith, B. (1996). Care and feeding of the

office grapevine. Management Review, 85, 6.

Smith, E. R., Murphy, J., & Coats, S.

(1999). Attachment to groups: Theory and

measurement. Journal of Personality and


Smith, J. C. (1993). Understanding stress

and coping. New York: Macmillan.

Smith, P. C., Kendall, L. M., & Hulin,

C. L. (1969). Measurement of satisfaction in

work and retirement. Chicago: Rand

McNally.

Soupata, L. (2005). Engaging employees in

company success: The UPS approach to a

winning team. Human Resource Manage-

ment, 44, 95--98.

Sparks, K., Cooper, C., Fried, Y., &

Shirom, A. (1997). The effect of hours of

work on health: A meta-analytic review.

Journal of Occupational and Organizational


Spector, P. E. (1982). Behavior in organi-

zations as a function of employees’ locus

of control. Psychological Bulletin, 91, 482--

497.

Spector, P. E. (1985). Measurement of

human service staff satisfaction: Develop-

ment of Job Satisfaction Survey. American

Journal of Community Psychology, 13, 693--

713.

Spector, P. E. (1986). Perceived control by

employees: A meta-analysis of studies con-

cerning autonomy and participation at

work. Human Relations, 11, 1005--1016.

Spector, P. E. (1987a). Interactive effects

of perceived control and job stressors on

affective reactions and health outcomes for

clerical workers. Work and Stress, 1, 155--

162.

Spector, P. E. (1987b). Method variance as

an artifact in self-reported affect and per-

ceptions at work: Myth or significant prob-

lem? Journal of Applied Psychology, 72,

438--443.

574 References

Spector, P. E. (1994). Using self-report

questionnaires in OB research: A comment

on the use of a controversial method. Jour-

nal of Organizational Behavior, 15, 385--

392.

Spector, P. E. (1997a). Job satisfaction:

Application, assessment, causes, and conse-

quences. Thousand Oaks, CA: Sage.

Spector, P. E. (1997b). The role of frustra-

tion in anti-social behavior at work. In

R. A. Giacalone & J. Greenberg (Eds.),

Anti-social behavior in the workplace (pp. 1--

17). Thousand Oaks, CA: Sage.

Spector, P. E. (2006). Industrial and orga-

nizational psychology: Research and practice.

Australia: Wiley.

Spector, P. E., Dwyer, D. J., & Jex, S. M.

(1988). Relations of job stressors to

affective, health, and performance out-

comes: A comparison of multiple data

sources. Journal of Applied Psychology, 73,

11--19.

Spector, P. E., & Jex, S. M. (1991). Rela-

tions of job characteristics from multiple

data sources with employee affect, absence,

turnover intentions, and health. Journal of


Spector, P. E., & Jex, S. M. (1998). Devel-

opment of four self-report measures of job

stressors and strain: Interpersonal conflict

at work scale, organizational constraints

scale, quantitative workload inventory, and

physical symptoms inventory. Journal of

Occupational Health Psychology, 3, 356--

367.

Spector, P. E. & O’Connell, B. J. (1994).

The contribution of individual dispositions

to the subsequent perceptions of job stres-

sors and job strains. Journal of Occupational

and Organizational Psychology, 67, 1--11.

Spector, P. E, Sanchez, J. I., Siu, L. O.,

Selgado, J., & Ma, J. (2004). Eastern ver-

sus Western control beliefs at work: An

investigation of secondary control, socioin-

strumental control, and work locus of con-

trol in China and the US. Applied

Psychology, 53(1), 38--50.

Spielberger, C. D. (1979). Preliminary man-

ual for the state-trait personality inventory

(STPI). Unpublished paper, University of

South Florida, Tampa.

Staber, U. (2004). Networking beyond

organizational boundaries: The case of

project organizations. Creativity and Inno-

vation Management, 13, 30--40.

Stajkovic, A. D., & Luthans, F. (1997). A

meta-analysis of the effects of organiza-

tional behavior modification on task per-

formance, 1975--1995. Academy of


Stajkovic, A. D., & Luthans, F. (1998).

Self-efficacy and work-related performance:

A meta-analysis. Psychological Bulletin, 124,

240--261.

Stajkovic, A. D., & Luthans, F. (2003).

Behavioral management and task perfor-

mance in organizations: Conceptual back-

ground, meta-analysis, and test of

alternative models. Personnel Psychology,

56, 155--194.

References 575

Staw, B. M. (1975). Attribution of the

‘‘causes’’ of performance: A general alterna-

tive interpretation of cross-sectional research

on organizations. Organizational Behavior and

Human Performance, 13, 414--432.

Staw, B. M., Bell, N. E., & Clausen, J. A.

(1986). The dispositional approach to job

attitudes: A lifetime longitudinal test.

Administrative Science Quarterly, 31, 56--77.

Staw, B. M., & Boettger, R. D. (1990).

Task revision: A neglected form of work

performance. Academy of Management Jour-

nal, 33, 534--559.

Staw, B. M., & Ross, J. (1985). Stability

in the midst of change: A dispositional

approach to job attitudes. Journal of


Steel, R. P. (1996). Labor market dimen-

sions as predictors of the reenlistment

decisions of military personnel. Journal of


Steel, R. P., & Griffeth, R. W. (1989). The

elusive relationship between perceived

employment opportunity and turnover

behavior: A methodological or conceptual

artifact? Journal of Applied Psychology, 74,

846--854.

Steel, R. P., & Rentsch, J. R. (1995). Influ-

ence of cumulation strategies on the long-

range prediction of absenteeism. Academy

of Management Journal, 6, 1616--1634.

Steers, R. M., Mowday, R. T., & Shapiro,

D. L. (2004). The future of work motiva-

tion theory. Academy of Management

Review, 29, 379--387.

Steers, R. M., & Rhodes, S. R. (1978).

Major influences on employee attendance:

A process model. Journal of Applied Psychol-

ogy, 63, 391--407.

Steil, G., & Gibbons-Carr, M. (2005).

Large group scenario planning: Scenario

planning with the whole system in the

room. Journal of Applied Behavioral Science,

41(1), 15--29.

Steiner, I. D. (1972). Group process and

productivity. New York: Academic Press.

Stevens, M. J., & Campion, M. A. (1999).

Staffing work teams: Development and val-

idation of a selection test for teamwork

settings. Journal of Management, 25, 207--

228.

Stogdill, R. M. (1948). Personal factors

associated with leadership: A survey of the

literature. Journal of Psychology, 25, 35--71.

Stone, D. L., & Kotch, D. A. (1989). Indi-

viduals’ attitudes toward organizational

drug testing policies and practices. Journal


Storms, P. L., & Spector, P. E. (1987).

Relationships of organizational frustration

with reported behavioral reactions: The

moderating effect of locus of control. Jour-

nal of Occupational Psychology, 60, 227--

234.

Straus, S. G., & McGrath, J. E. (1994).

Does the medium matter? The interaction

of task type and technology on group per-

formance and member reactions. Journal of


576 References

Strodtbeck, F. L., & Lipinski, R. M.

(1985). Becoming first among equals:

Moral considerations in jury foreman

selection. Journal of Personality and Social


Stroh, L. K., & Dennis, L. E. (1994). An

interview with Madame Nguyen Minh

Hoa: Vietnam’s move to a market economy

and the impact on women in the work-

place. Industrial-Organizational Psychologist,

31, 37--42.

Sturman, M. C., Cheramie, R. A., &

Cashen, L. H. (2005). The impact of job

complexity and performance measurement

on the temporal consistency, stability, and

test-retest reliability of employee job per-

formance ratings. Journal of Applied Psychol-

ogy, 90, 269--283.

Sullivan, S. E., & Bhagat, R. S. (1992).

Organizational stress, job satisfaction, and

job performance: Where do we go from

here? Journal of Management, 18, 353--

374.

Sulsky, L., & Smith, C. (2005). Work

Stress. Belmont, CA: Thomson Wadsworth.

Sundstrom, E., DeMeuse, K. P., & Futrell,

D. (1990). Work teams: Applications and

effectiveness. American Psychologist, 45,

120--133.

Sverke, M., Hellgren, J., & Naswall, K.

(2002). No security: A meta-analysis and

review of job insecurity and its consequen-

ces. Journal of Occupational Health Psychol-

ogy, 7(3), 242--264.

Swanson, N. G., & Murphy, L. R. (1991).

Mental health counseling in industry. In

C. L. Cooper & I. T. Robertson (Eds.),

International review of industrial and organi-

zational psychology (pp. 265--282). New

York & London: Wiley.

Sweeney, P. D., & McFarlin, D. B. (1997).

Process and outcome: Gender differences

in the assessment of justice. Journal of


Tabachnick, B. G., & Fidell, L. S. (1983).

Using multivariate statistics. New York:

Harper & Row.

Tabachnick, B. G., & Fidell, L. S. (1996).

Using multivariate statistics (3rd ed.). New

York: Harper & Row.

Tajfel, H. (1981). Human groups and social

categories. New York: Cambridge Univer-

sity Press.

Taylor, F. W. (1911). Principles of scientific

management. New York: Harper.

Taylor, S. E. (1991). Asymmetrical effects

of positive and negative events: The

mobilization-minimization hypothesis. Psy-

chological Bulletin, 110, 67--85.

Taylor, S. E., Klein, L. C., Lewis, B. P.,

Gruenewald, T. L., Gurung, R. A. R., &

Updegraff, J. A. (2000). Biobehavioral

responses to stress in females: Tend-and-

befriend, not fight-or-flight. Psychological

Review, 107(3), 411--429.

Teerlink, R., & Ozley, L. (2000). More

than a motorcycle: The leadership journey at

References 577

Harley-Davidson. Boston: Harvard Business

School Press.

Terpstra, D. E. (1981). Relationship

between methodological rigor and reported

outcomes in organization development

evaluation research. Journal of Applied Psy-

chology, 66, 541--543.

Tesluk, P. E., Farr, J. L., Klein, S. R.

(1997). Influences of organizational culture

and climate on individual creativity. Jour-

nal of Creative Behavior, 31, 27--41.

Tesluk, P. E., & Jacobs, R. R. (1998).

Toward an integrated model of work expe-

rience. Personnel Psychology, 51, 321--355.

Tesluk, P. E., & Mathieu, J. E. (1999). Over-

coming roadblocks to effectiveness: Incorpo-

rating management of performance barriers

into models of work group effectiveness.


Tetrick, L. E., & Quick, J. C. (2003). Pre-

vention at work: Public health in occupa-

tional settings. In J. C. Quick & L. E.

Tetrick (Eds.). Handbook of occupational

health psychology (pp. 3--18). Washington,

DC: American Psychological Association

Tett, R. P., Jackson, D. N., Rothstein, M.

(1991). Personality measures as predictors

of job performance: A meta-analytic

review. Personnel Psychology, 44, 703--742.

Tett, R. P., & Meyer, J. P. (1993). Job

satisfaction, organizational commitment,

turnover intention, and turnover: Path

analyses based on meta-analytic findings.


Theodory, G. C. (1982). The validity of

Fiedler’s contingency logic. Journal of Psy-

chology, 110, 115--120.

Thomas, J. L., & Castro, C. A. (2003).

Organizational behavior and the U.S.

peacekeeper. In T. W. Britt & A. B. Adler

(Eds.), The psychology of the peacekeeper:

Lessons from the field (pp. 127--146). West-

port, CT: Praeger Publishers.

Thomas, K. (1976). Conflict and conflict

management. In M. D. Dunnette (Ed.),


psychology (pp. 889--935). Chicago: Rand

McNally.

Thomas, L. T., & Ganster, D. C. (1995).

Impact of family-supportive work variables

on work-family conflict and strain: A con-

trol theory perspective. Journal of Applied


Thomas, P., McDonnell, J., McCulloch, J.,

While, A., Bosanquet, N., & Ferlie, E.

(2005). Increasing capacity for innovation

in bureaucratic primary care organizations:

A whole system participatory action

research project. Annals of Family Medicine,

3, 312--317.

Thoreson, C. J., Bradley, J. C., Bliese, P.

D., & Thoreson, J. D. (2004). The big five

personality traits and individual job per-

formance growth strategies in maintenance

and transitional job stages. Journal of


Tiegs, R. B., Tetrick, L. E., & Fried, Y.

(1992). Growth need strength and context

satisfaction as moderators of the relations

578 References

of the job characteristics model. Journal of


Tierney, P., Farmer, S. M., & Graen, G. B.

(1999). An examination of leadership and

employee creativity: The relevance of traits

and relationships. Personnel Psychology, 52,

591--620.

Tolson, J. (2001, February 26). Founding

rivalries: More like squabbling brothers

than ‘‘fathers,’’ how did they succeed. U.S.

News and World Report, 50--55.

Tombaugh, J. R., & White, L. P. (1990).

Downsizing: An empirical assessment of

survivors’ perceptions of a postlayoff envi-

ronment. Organization Development Journal,

8, 32--43.

Toquam, J. L., Macaulay, J. L., Westra, C.

D., Fujita, Y., & Murphy, S. E. (1997).

Assessment of nuclear power plant crew

performance variability. In M. T. Brannick,

E. Salas, & C. Prince (Eds.), Team perfor-

mance assessment and measurement (pp.

253--287). Mahwah, NJ: Lawrence Erl-

baum Associates.

Trevino, L. K. (1992). The social effects

of punishment in organizations: A justice

perspective. Academy of Management Review,

17, 647--676.

Trevor, C. O., Gerhart, B., Boudreau, J.

W. (1997). Voluntary turnover and job

performance: Curvilinearity and the mod-

erating influences of salary growth and

promotions. Journal of Applied Psychology,

82, 44--61.

Trice, H. M., & Beyer, J. M. (1984).

Studying organizational culture through

rites and ceremonials. Academy of Manage-

ment Review, 9, 653--669.

Trip, T. M., Bies, R. J., & Aquino, K.

(2002). Poetic justice or petty jealousy?

The aesthetics of revenge. Organizational

Behavior and Human Decision Processes, 89,

966--984.

Trist, E. L., & Bamforth, K. W. (1951).

Some social and psychological consequen-

ces of the long-wall method of coal-getting.

Human Relations, 4, 3--38.

Trubisky, P., Ting-Toomey, S., & Lin, S.

(1991). The influence of individualism col-

lectivism and self-monitoring on conflict

styles. International Journal of Intercultural

Relations, 15(1), 65--84.

Tubre, T. C., Sifferman, J. J., & Collins, J.

M. (1996, April). Jackson and Schuler

(1985) revisited: A meta-analytic review of

the relationship between role stress and job

performance. Paper presented at the

annual meeting of the Society for Indus-

trial and, Organizational Psychology, San

Diego, CA.

Tucker, L. A., Aldana, S., & Friedman, F.

M. (1990). Cardiovascular fitness and

absenteeism in 8,301 employed adults.

American Journal of Health Promotion, 5,

140--145.

Tuckman, B. W. (1965). Developmental

sequences in small groups. Psychological

Bulletin, 63, 384--399.

References 579

Tuckman, B. W., & Jensen, M. A. C.

(1977). Stages of small group development

revisited. Group and Organization Studies, 2,

419--427.

Turner, A. N., & Lawrence, P. R. (1965).

Industrial jobs and the worker. Cambridge,

MA: Harvard University, Graduate School

of Business.

Tziner, A., & Eden, D. (1985). Effects of

crew composition on crew performance:

does the whole equal the sum of its parts?

Journal of Applied Psychology, 70(1), 85--93.

U.S. Bureau of Labor Statistics. (1999).

National census of fatal occupational injuries

1998. Washington, DC: Author.

U.S. Bureau of Labor Statistics. (2004).

[Online] Available at http://www.bls.gov/

iif/home.html/ Accessed January 25,

2007.

U.S. Department of Health and Human

Services. (2002). National occupational

research agenda: Update 2001. Washington,

DC: Author

VandenHeuvel, A., & Wooden, M. (1995).

Do explanations of absenteeism differ for

men and women? Human, Relations, 48,

1309--1329.

Van De Vliert, E., & Van Yperen, N. W.

(1996). Why cross-national differences in

role overload? Don’t overlook ambient,

temperature! Academy of Management Jour-

nal, 39, 986--1004.

Van Dyne, L., & LePine, J. A. (1998).

Helping and voice extra-role behaviors:

Evidence of construct and predictive valid-

ity. Academy of Management Journal, 41,

108--119.

Van Eerde, W., & Thierry, H. (1996).

Vroom’s expectancy models and work-

related criteria: A meta-analysis. Journal of


Van Maanen, J. (1975). Police socializa-

tion: A longitudinal examination of job

attitudes in an urban police department.

Administrative Science Quarterly, 20, 207--

228.

Van Maanen, J. (1991). The smile factory:

Work at Disneyland. In P. J. Frost, L. F.

Moore, M. R. Louis, C. C. Lundberg, & J.

Martin (Eds.), Reframing organizational cul-

ture (pp. 58--76). Newbury Park, CA: Sage.

Van Maanen, J., & Schein, E. H. (1979).

Toward a theory of organizational social-

ization. In B. M. Staw (Ed.), Research in

organizational behavior (Vol. 1, pp. 209--

264). Greenwich, CT: JAI Press.

Van Rooy, D. L., Dilchert, S., & Viswes-

varan, C. (2006). Multiplying intelligences:

Are general, emotional, and practical intel-

ligences equal? In K. R. Murphy (Ed.), A

critique of emotional intelligence: What are

the problems and how can they be fixed (pp.

235--262). Mahwah, NJ: Lawrence Erl-

baum Associates.

Van Rooy, D. L., & Viswesvaran, C.

(2004). Emotional intelligence: A meta-

analytic investigation of predictive validity

and nomological net. Journal of Vocational

Behavior, 65, 71--95.

580 References

Van Scotter, J. R., & Motowidlo, S. J.

(1996). Interpersonal facilitation and job

dedication as separate facets of contextual

performance. Journal of Applied Psychology,

81, 525--531.

Van Vianen, A. E. M. (2000). Person-

organization fit: The match between new-

comers’ and recruiters’ preferences for

organizational cultures. Personnel Psychol-

ogy, 53, 113--150.

Van Vianen, A. E. M., & De Dreu, C. K.

W. (2001). Personality in teams: Its rela-

tionship to social cohesion, task cohesion,

and team performance. European Journal of

Work and Organizational Psychology, 10(2),

97--120.

Vecchio, R. P. (1977). An empirical exami-

nation of the validity of Fiedler’s model of

leadership effectiveness. Organizational


206.

Vigoda-Gadot, E., & Ben-Zion, E. (2004).

Bright shining stars: The mediating effect

of organizational image on the relationship

between work variables and army officers’

intention to leave the service for a job in

high-tech industry. Public Personnel Man-

agement, 33(2), 201--224.

Vigoda, E., & Cohen, A. (2002). Influence

tactics and perceptions of organizational

politics: A longitudinal study. Journal of

Business Research, 55, 311--324.

Vigoda-Gadot, E., Vinarski-Peretz, H., Ben-

Zion, E. (2003). Politics and image in the

organizational landscape: An empirical

examination among public sector employ-

ees. Journal of Managerial Psychology, 18(8),

764--787.

Villanova, P., & Roman, M. A. (1993). A

meta-analytic review of situational con-

straints and work-related outcomes: Alter-

native approaches to conceptualization.

Human Resource Management Review, 3,

147--175.

Vinokur, A. D., van Ryn, M., Gramlich,

E. M., Price, R. H. (1991). Long-term

follow-up and benefit-cost analysis of the

Jobs program: A preventive intervention

for the unemployed. Journal of Applied Psy-

chology, 76, 213--219.

Viswesvaran, C. (2002). Assessment of

individual performance: A review of the

past century and a look ahead. In N.

Anderson, D. S. Ones, H. K. Sinangil, &

C. Viswesvaran (Eds.), Handbook of indus-

trial, work, and organizational psychology

(pp. 110--126). Thousand Oaks, CA: Sage.

Viteles, M. S. (1932). Industrial psychology.

New York: Norton.

Viteles, M. S. (1953). Motivation and morale

in industry. New York: Norton.

Vizquel, O.(with Bob Dyer). (2002). Omar!

My life on and off the field. Cleveland, OH;

Gray & Company Publishers.

von Bertalanffy, L. (1956). General systems

theory. General systems Yearbook of the Soci-

ety for General Systems Theory, 1, 1--10.

Vroom, V. H. (1964). Work and motivation.

New York: Wiley.

References 581

Vroom, V. H. (1995). Work and motivation

(2nd ed.). New York: Wiley.

Vroom, V. H., & Jago, A. C. (1988). The

new leadership: Managing participation in org-

anizations. Englewood Cliffs, NJ: Prentice-

Hall.

Vroom, V. H., & Jago, A. G. (2007). The

role of the situation in leadership. Ameri-

can Psychologist, 62, 17--24.

Vroom, V. H., & Yetton, P. W. (1973).

Leadership and decision making. Pittsburgh,

PA: University of Pittsburgh Press.

Wageman, R. (1996). Interdependence and

group effectiveness. Administrative Science

Quarterly, 40, 145--180.

Wageman, R., & Baker, G. (1997). Incen-

tives and cooperation: The joint effects of

task and reward interdependence on group

performance. Journal of Organizational

Behavior, 18, 139--158.

Wagman, R., Hackman, R. J., & Lehman,

E. (2005). Team Diagnostic Survey:

Development of an Instrument. Journal of

Applied Behavioral Science, 41(4), 373--

398.

Wagner, J. A. (1994). Participation’s effect

on performance and satisfaction: A recon-

sideration of research evidence. Academy of


Wagner, J. A. (1995). Studies of individu-

alism-collectivism: Effects on cooperation

in groups. Academy of Management Journal,

38, 152--172.

Wagner, J. A., & Gooding, R. Z. (1987).

Shared influence and organizational behav-

ior: A meta-analysis of situational variables

expected to moderate participation-outcome

relationships. Academy of Management Jour-

nal, 30, 524--541.

Waldman, D. A., & Spangler, W. D.

(1989). Putting together the pieces: A

closer look at the determinants of job per-

formance. Human Performance, 2, 29--59.

Wanberg, C. R. (1997). Antecedents and

outcomes of coping behaviors among

unemployed and reemployed individuals.


744.

Wang, H., Law, K. S., Hackett, R. D.,

Wang, D., & Chen, Z. X (2005). Leader-

member exchange as a mediator of the

relationship between transformational lead-

ership and followers’ performance and

organizational citizenship behavior. Acad-

emy of Management Journal, 48, 420--432.

Wanous, J. P. (1989). Installing realistic

job previews: Ten tough choices. Personnel


Wanous, J. P., Poland, T. D., Premack, S.

L., Davis, K. S. (1992). The effects of met

expectations on newcomer attitudes and

behaviors: A review and meta-analysis.


Wanous, J. P., & Zwany, A. (1977). A

cross-sectional test of the need hierarchy.


formance, 18, 78--97.

582 References

Wasserman, N. (2006). Stewards, agents,

and the founder discount: Executive com-

pensation in new ventures. Academy of


Watson, D., & Clark, L. (1984). Negative

affectivity: The disposition to experience

aversive emotional states. Psychological Bul-

letin, 96, 465--490.

Wayne-andy, S. J., & Green-hawn, S. A.

(Dec 1993). The effects of leader-member

exchange on employee citizenship and

impression management behavior. Human

Relations, Vol 46(12), pp. 1431--1440.

Webb, E. J., Campbell, D. T., Schwartz, R.

D., Sechrest, L., & Grove, J. B. (1981).

Nonreactive measures in the social sciences

(2nd ed.). Boston: Houghton Mifflin.

Weber, M. (1947). The theory of social and

economic organization (A. M. Henderson &

T. Parsons. Trans.). New York: Free Press.

Webster’s New World Dictionary (1984).

(2nd College ed.). New York: Simon &

Schuster.

Weed, E. D. (1971). Job enrichment

‘‘Cleans up’’ at Texas Instruments. In J. R.

Maher (Ed.), Perspectives in job enrichment.

New York: Van, Nostrand.

Weiner, Y., & Vardi, Y. (1990). Relation-

ships between organizational culture and

individual motivation---A conceptual inte-

gration. Psychological Reports, 67, 295--306.

Weiss, D. J., Dawis, R. V., England, G.

W., & Lofquist, L. H. (1967). Manual for

the Minnesota Satisfaction Questionnaire

(Minnesota Studies in Vocational Rehabili-

tation. No. 22). Minneapolis: University of

Minnesota.

Weiss, H. M. (1990). Learning theory and

industrial and organizational psychology.



psychology (2nd ed., Vol.1, pp. 171--222).

Palo Alto, CA: Consulting Psychologists

Press.

Weiss, H. M., & Shaw, J. (1979). Social

influences on judgments about tasks.


formance, 24, 126--140.

Weitz, J. (1952). A neglected concept in

the study of job satisfaction. Personnel Psy-

chology, 5, 201--205.

White, S., & Mitchell, T. (1979). Job

enrichment versus social cues: A compari-

son and competitive test. Journal of Applied


Whitehead, T. N. (1935). Social relation-

ships in the factory: A study of an indus-

trial group. Human Factor, 9, 381--394.

Whitehead, T. N. (1938). The industrial

worker. Cambridge, MA: Harvard Univer-

sity Press.

Whitney, K. (1994). Improving group task

performance: The role of group goals and

group efficacy. Human Performance, 7, 55--

78.

Whyte, G. (1998). Recasting Janis’s Group-

think Model: The key role of collective

efficacy in decision making fiascoes.

References 583


Processes, 73, 185--209.

Wickens, C. D., & Hollands, J. G. (2000).

Engineering psychology and human perform-

ance, (3rd ed.). Upper Saddle River, NJ:

Prentice-Hall.

Wildschut, T., Insko, C. A., & Gaertner,

L. (2002). Intragroup social influence and

intergroup competition. Journal of Personal-

ity and Social Psychology, 82, 975--992.

Wilkins, A. L., & Ouchi, W. G. (1983).

Efficient cultures: Exploring the relation-

ship between culture and organizational

performance. Administrative Science Quar-

terly, 28, 468--481.

Williams, C. R., & Livingstone, L. P.

(1994). Another look at the relationship

between performance and voluntary turn-

over. Academy of Management Journal, 37,

269--298.

Williams, L. J., & Anderson, S. E. (1991).

Job satisfaction and organizational commit-

ment as predictors of organizational citi-

zenship and in-role behavior. Journal of


Williams, L. J., & Anderson, S. E. (1994).

An alternate approach to method effects

using latent variable models: Applications

in organizational behavior research. Journal


Williams, L. J., Cote, J. A., & Buckley, M.

R. (1989). Lack of method variance in

self-reported affect and perceptions at

work: Reality of artifact? Journal of Applied


Williams, M. L., McDaniel, M. A., &

Nguyen, N. T. (2006). A meta-analysis of

the antecedents and consequences of pay

level satisfaction. Journal of Applied Psychol-

ogy, 91, 392--413.

Wilson, M., Northcraft, G. B., & Neale, M.

A. (1985). The perceived value of fringe

benefits. Personnel Psychology, 38, 309--320.

Wilson, T. B. (1995). Innovative reward

systems for the changing workplace. New

York: McGraw-Hill.

Wofford, J. C., & Liska, L. Z. (1993).

Path-goal theories of leadership: A meta-

analysis. Journal of Management, 19, 857--

876.

Wood, R. E. (1986). Task complexity: Def-

inition of the construct. Organizational

Behavior and Human Decision Processes, 37,

60--82.

Woodman, R. W., & Wayne, S. J. (1985).

An investigation of positive findings bias

in evaluation of organization development

interventions. Academy of Management Jour-

nal, 28, 889--913.

Woodward, J. (1965). Industrial organiza-

tion: Theory and practice. London: Oxford

University Press.

Worchel, S., Cooper, J., Goethals, G. R., &

Olson, J. M. (2000). Social psychology. Bel-

mont, CA: Wadsworth.

Wright, K. B. (2005). Researching Internet-

based populations: Advantages and dis-

advantages of online survey research, online

questionnaire authoring software packages,

584 References

and web survey services. Journal of Com-

puter-Mediated Communication, 10(3), http://

jcmc.indiana.edu/vol10/issue3/wright.html.

Wright, P. M. (1992). An examination of the

relationships among monetary incentives,

goal level, goal commitment, and perform-

ance. Journal of Management, 18, 677--693.

Wright, P. M., George, J. M., Farnsworth,

S. R., & McMahan, G. (1993). Productivity

and extra-role behavior: The effects of

goals on spontaneous helping. Journal of


Xie, J. L. (1996). Karasek’s model in the

People’s Republic of China: Effects of job

demands, control, and individual differ-

ences. Academy of Management Journal, 39,

1594--1618.

Yen, H. R., & Niehoff, B. P. (2004). Orga-

nizational citizenship behaviors and organiza-

tional effectiveness: Examining relationships

in Taiwanese banks. Journal of Applied Social

Psychology, 34, 1617--1637.

Yoffie, D., & Bergenstein, S. (1985). Creat-

ing political advantage: The rise of corpo-

rate enterpreneurs. California Management

Review, Fall, 124--139.

Yukl, G. (1989). Leadership in organiza-

tions. Englewood Cliffs, NJ: Prentice-Hall.

Yukl, G. A. (2006). Leadership in organiza-

tions (6th Ed.). Upper Saddle River. NJ:

Pearson Prentice Hall.

Yukl, G., Kim, H., Falbe, C. M. (1996).

Antecedents of influence outcomes. Journal


Yukl, G., & Tracey, J. B. (1992). Conse-

quences of influence tactics used with sub-

ordinates, peers, and the boss. Journal of


Yukl, G., & Van Fleet, D. D. (1992). Theory

and research on leadership in organizations.



chology (2nd ed., Vol.2, pp. 147--197). Palo


Yun, S., Faraj, S., & Sims, H. P. (2005).

Contingent leadership and effectiveness of

trauma resuscitation teams. Journal of


Zaccaro, S. J. (2007). Trait-based perspec-

tives on leadership. American Psychologist,

62, 6--16.

Zaccaro, S. J., Blair, V., Peterson, C., &

Zazanis, M. (1995). Collective efficacy. In

J. E. Maddux (Ed.), Self-efficacy, adaptation,

and adjustment: Theory, research, and applica-

tion (pp. 305--328). New York: Plenum Press.

Zaccaro, S. J., Foti, R. J., & Kenney, D. A.

(1991). Self-monitoring and trait-based

variance in leadership: An investigation of

leader flexibility across multiple group

situations. Journal of Applied Psychology, 76,

308--315.

Zammuto, R. F. & O’Connor, E. J. (1992).

Gaining advanced manufacturing technolo-

gies benefits: The role of organization

design and culture. Academy of Manage-

ment Review, 17, 701--728.

Zapf, D., Dormann, C., Frese, M. (1996).

Longitudinal studies in organizational stress

References 585

research: A review of the literature with

reference to methodological issues. Journal of

Occupational Health Psychology, 2, 145--169.

Zickar, M. J. (2001). Using personality

inventories to identify thugs and agitators:

Applied psychology’s contribution to the

war against labor. Journal of Vocational

Behavior, 58, 149--164.

Zickar, M. J. (2003). Remembering Arthur

Kornhauser: Industrial psychology’s advo-

cate for worker well-being. Journal of


Zohar, D. (2000). A group-level model of

safety climate: Testing the effect of group

climate on microaccidents in manufactur-

ing jobs. Journal of Applied Psychology,

85(4), 587--596.

Zohar, D. (2003). Safety climate: Concep-

tual and measurement issues. In J. C.

Quick & L. E. Tetrick (Eds.), Handbook of

occupational health psychology (pp. 123--

142). Washington, DC: American Psycho-

logical, Association.

Zweig, D., & Webster, J. (2002). Where is

the line between benign and invasive?: An

examination of psychological barriers to

the acceptance of awareness monitoring

systems. Journal of Organizational Behavior,

23(5), 605--633.

586 References

Documents

ORGANIZATIONAL PSYCHOLOGY - UPB