A Modification of the Unified Theory of Acceptance Accepted and Signed and Use of Technolog

COLLEGE STUDENT’S ACCEPTANCE OF TABLET PERSONAL COMPUTERS:

A MODIFICATION OF THE UNIFIED THEORY OF ACCEPTANCE

AND USE OF TECHNOLOGY MODEL

by

Mark, J. Moran

A Dissertation Presented in Partial Fulfillment

Of the Requirements for the Degree

Doctor of Philosophy

Capella University

August 2006

© Mark Moran, 2006

COLLEGE STUDENT’S ACCEPTANCE OF TABLET PERSONAL COMPUTERS:

A MODIFICATION OF THE UNIFIED THEORY OF ACCEPTANCE

AND USE OF TECHNOLOGY MODEL

by

Mark, J. Moran

has been approved

August 2006

APPROVED:

CLIFF BUTLER, Ph.D., Faculty Mentor and Chair

VALERIE COXON, Ph.D., Committee Member

MARK HAWKES, Ph.D., Committee Member

ACCEPTED AND SIGNED:

__________________________________________ CLIFF BUTLER, Ph.D.

__________________________________________

Kurt Linberg, Ph.D. Dean, School of Business & Technology

Abstract

Information technology can enhance the learning process for post secondary students.

Many universities have implemented ubiquitous or required notebook or tablet personal

computing for their students but have not studied the acceptance of the technology by

their student populations. This research examines student acceptance of mobile

computing devices using a modification of the “Unified Theory of Acceptance and Use

of Technology (UTAUT)” recently developed by leading researchers in the technology

acceptance field. The objective of the study is to evaluate students’ acceptance of Tablet

PC (TPC) as a mean to forecast, explain, and improve usage patterns of UTAUT in

alternate contexts. The research contributes to UTAUT’s theoretical validity and

empirical applicability and to the management of information technology (IT) based

initiatives in education.

Dedication

This dissertation is dedicated to my family who has been giving me their unwavering

support throughout this life-changing experience. To my wife, Mary, who has been so

strong and supportive all through this educational journey. To my children, Matthew and

Marin, you are my inspiration to reach beyond my potential and reach goals I thought

unattainable. I love you all so much.

iii

Acknowledgments

First, I would like to extend a sincere thank you to Dr. Cliff Butler, who has been

a compassionate and diligent mentor during my voyage at Capella University. I thank you

for the guidance and support as my dissertation chair, professor, and mentor. Your work

ethic and leadership inspire me and I look forward to continuing our relationship for

many years to come.

I want to thank Dr. Valerie Coxon for agreeing to join my committee in a time of

need, and for your many insightful comments during the comprehensive and dissertation

process. Dr. Coxon, your supportive remarks often were inspirational and kept me

moving forward. Dr. Mark Hawkes, thank you for joining me on this trek your

knowledge of instructional technology in education was critical to this task. Because of

all of you I look forward to joining my academic peer in the pursuit of knowledge. I wish

you all the best in life.

Finally, I would like to thank Dr. Omar El-Gayar for being the catalyst that

spurred my dissertation process. Your tireless efforts at research and academia are

something I will strive to attain. Thanks for all your contribution to this dissertation

especially in the area of technology acceptance and the analysis of the data.

iv

Table of Contents

Acknowledgments iii

List of Tables viii

List of Figures x

CHAPTER 1: INTRODUCTION 2

Wireless Data Communication 3

The Device 5

Pen Based Computing Background 7

The Adoption of Technological Innovations 9

Objective of the Study 10

Research Questions 11

Significance of the Study 13

Scope of the Study 14

Study Context 14

CHAPTER 2: LITERATURE REVIEW 16

1

v

History of Technology Acceptance Models 17

The Technology Acceptance Model 20

The Unified Theory of Acceptance and Use of Technology Model 24

UTAUT Supporting Research 27

CHAPTER 3: RESEARCH DESIGN 28

Performance Expectancy 31

Effort Expectancy 32

Social Influence 32

Facilitating Conditions 32

Supplementary Variables 33

Pilot Study 33

The Survey 35

Sample Size 40

Survey Sample Population 40

Human Subjects Protection 41

Survey Questions 42

vi

Data Analysis Methodology 50

Measures 52

CHAPTER 4: DATA ANALYSIS 54

Data Sample 54

Data Validation 59

Model Validity 73

Reliability 74

Construct Validity 77

Partial Least Squares 82

Model Analysis 82

Structural Model Analysis 86

Freshman vs. Upper Classmen 88

CHAPTER 5: SUMMARY & CONCLUSIONS 93

Discussion 98

Moderating Conditions 100

Limitations 101

vii

Future research 102

REFERENCES 105

APPENDIX A: SELECTION OF COURSES TO SURVEY 113

APPENDIX B: SURVEY QUESTIONS 116

APPENDIX C: CROSS TAB REPORTS 120

viii

List of Tables

Table 1. Theory of Planned Behavior Constructs...........................................................19

Table 2. TAM Research..................................................................................................23

Table 3. UTAUT Components........................................................................................26

Table 4. UTAUT Moderators .........................................................................................26

Table 5. Performance Expectancy Questions .................................................................44

Table 6. Effort Expectancy Questions ............................................................................45

Table 7. Attitude Toward Using Technology Questions ................................................45

Table 8. Social Influence Questions ...............................................................................46

Table 9. Facilitating Conditions Questions.....................................................................46

Table 10. Behavior Intent Questions ..............................................................................47

Table 11. Self Efficacy Questions ..................................................................................48

Table 12. Anxiety Questions...........................................................................................48

Table 13. Usage and Other Information Questions ........................................................49

Table 14. Scale Reliabilities ...........................................................................................52

Table15a. Statistical Analysis of the Variables ..............................................................60

Table15b. Statistical Analysis (continued) .....................................................................61

Table 16. Cross tab report for PE1 by Gender................................................................63

Table 17. Cross tab report for PE1 by Class Status. .......................................................64

Table 18. T test and p Values for Participating Groups and PE1 ...................................66

Table 19. Cross tab report for PE1 by College Major ....................................................67

Table 20. Mean, Standard Deviation, Skewness, and Kurtosis for College and PE1.....68

ix

Table 21. Cross tab report for PE1 by First Computer Use. ............................................69

Table 22. Basic statistics for Computer Experience and PE1.........................................70

Table 23. T test and p values for Participant Groups and PE1 .......................................70

Table 24. Cross tab report for PE1 by First Tablet PC Use............................................71

Table 25. Basic Statistics for Tablet PC Use and PE1....................................................72

Table 26. T-test and p-Values for Tablet PC Use and PE1 ............................................72

Table 27a. Reliability of Performance Expectancy Construct Variables .......................75

Table 27b. Reliability of EE, ATUT, SI, FC, SE Variables ..........................................76

Table 28. Reliability of ANX, BI, & USE Variables......................................................77

Table 29. Internal Consistency Factor Analysis for PE1 – PE10 ...................................79

Table 30. Correlation Coefficients for Performance Expectancy Indicators..................80

Table 31. Internal Consistency & Factor Loading Analysis for other Indicators...........81

Table 32. Individual Loadings, Weights, and Internal Consistencies.............................84

Table 33. AVE Scores and Correlation of Latent Variables...........................................86

Table 34. Comparison of Freshman and Upper Classman Model Contributions ...........91

Table 35. Comparison of All, Freshman, & Upper Class Models..................................92

Table A-1. Course on which Survey Tool was Administered ......................................114

Table A-2. Individual Course Section Participation .....................................................115

x

List of Figures

Figure 1. Theory of Planned Behavior............................................................................18

Figure 2. Technology Acceptance Model.......................................................................20

Figure 3. Extended Technology Acceptance Model.......................................................22

Figure 4. Unified Theory of Acceptance and Use of Technology (UTAUT).................25

Figure 5. Dissertation Research Model...........................................................................30

Figure 6. Research Page Link .........................................................................................36

Figure 7. Web Survey Link.............................................................................................37

Figure 8. Questions and Concerns Information ..............................................................38

Figure 9. Web Survey Instrument...................................................................................39

Figure 10. Research Relationship Equations ..................................................................53

Figure 11. Survey Participants Class Status ...................................................................55

Figure 12. Survey Participants College Affiliation.........................................................56

Figure 13. Survey Participants First Computer Use .......................................................57

Figure 14. Survey Participants Length of Use of Table PC............................................58

Figure 15. Histogram of PE1 ..........................................................................................62

Figure 16. Tablet PC Structural Model...........................................................................88

Figure 17. Freshman PLS-Graph Model.........................................................................89

Figure 18. Upper Classman PLS-Graph Model..............................................................90

CHAPTER 1: INTRODUCTION

The application of computer technology in collegiate classrooms has been

demonstrated to improve teaching when used appropriately (Surry & Land, 2000). In the

past few years many universities have introduced mobile computing to their campus as a

way to improve the productivity of and communication between students and faculty. But

some faculty have raised concerns about the distractions caused by mobile computer

hardware (Groves & Zemel, 2000). However, even with philosophical differences among

faculty, many universities including Bentley College (Lowe, 2004), Notre Dame (Abbott,

2004), University of Texas (Mock, 2004), and the University of Washington (Willis &

Miertschin, 2004) have implemented, or are in the process of starting, mobile computing

initiatives.

Some colleges and universities have adopted computing initiatives that require

every student to acquire their own portable computing device or in some cases participate

in experiments with university supplied computers, more than fifty colleges and

universities have, or are in the process of, implementing various mobile computing

initiatives. A website administered by Dr. Ray Brown, of Valley City State University,

contains a list of over seventy institutions that are involved in various levels of mobile

computing implementation (Brown, 2000). Many of these implementations included re-

engineering of the institutions’ data communication facilities. Several of these have

decided to adopt a specialized development of notebook computers that allow pen based

data entry and screen manipulation.

Acceptance of Tablet PC’s 2

Tablet Personal Computer (TPC) based mobile computing initiatives have been

documented in the literature with programs ranging from a preliminary pilot study at the

University of Houston (Willis & Miertschin, 2004) investigating TPCs in a mobile

learning laboratory used by faculty. A university that integrates the TPC into student

teacher interaction is the University of Washington where a Classroom Feedback System

(CFS) is being used to give students the ability to provide feedback and ask real time

questions during an instructor mediated lecture (Steel, et al., 2003). Every student

enrolled in Bentley College gets a TPC (Lowe, 2004). Other universities with TPC

programs include Purdue, MIT, Temple, Seton Hall, Chatham, and many others (Brown,

2000) (Wachsmuth, 2003).

One of these institutions to make significant commitment to ubiquitous, mobile

computing is a small Midwest public university located in South Dakota. This institution

started investigating pen based mobile computing in 2002 when thirteen wireless access

points were installed on their campus (DSU goes wireless, 2002). Approximately twenty

Center of Excellence (CEX) students were given tablet PC devices and given the

assignment to investigate the device as a learning instrument. The initial project was

found to be beneficial resulting in expansion of both wireless network availability and

students with table PCs. A task force was then organized to examine strategies for taking

mobile computing to scale at the university, and to investigate computing device options.

In the fall of 2004, this university moved wireless mobile computing from a small

number of Center of Excellence students and scaled it to include all first and second year

enrolled students (Knowlton, 2004). This initiative required each full time student to


lease a wireless, mobile tablet/laptop personal computing device. The program has been

entitled the wireless mobile computing initiative (WMCI). By the spring of 2006 all

students at this university should have their own TPC (Zolnowsky, 2006).

Wireless Data Communication

Wireless networks, by themselves, can not support the typical communication

traffic on a modern campus. The wireless aspect supports a continuous communication

environment but a high speed wired network backbone must be in place to support the

broadband applications that present applications require. Modern college student’s use

many bandwidth hungry applications such as instant messaging, music downloads and

peer-to-peer programs (Henderson, Kotz, & Abyzov, 2004). These data intensive

applications make it difficult to a consistent quality of service (QOS) for all of the

developing applications.

Over the past decade the nation’s commercial, academic, and telecommunications

sectors have made considerable improvements in their telecommunications infrastructure

(Decusatis, 2002). According to Dr. Kenneth Green (Green, 2004), who has been

conducting surveys of college campus computer use since 1990, roughly 4 out of 5

private and public four year colleges claim they have functional wireless LANs that can

be used by their students, faculty and staff in parts or all of their campus facilities. This

expansion of wireless connectivity capacity has been augmented with many universities

adopting pen based computing initiatives.

Considering the mobility of laptops and TPCs in a wireless LAN environment

steps need to be taken to allow the mobile computing device to keep one Internet


Protocol (IP) as the user moves across campus. Each access point covers about an area

with a radius of about 100 feet so computing devices moving from one area of campus to

another will move through several different AP zones. Most campuses need to segment

their networks to reduce data traffic congestion in any one zone or building. Mobile users

will move from their original zone to other zones as they move across campus requiring

them to obtain a new IP address for each subnet. Obviously this would not be an

acceptable requirement of a WLAN. The solution is to place all the wireless access points

on their own virtual LAN (VLAN) (Ciampa, 2001). Using a VLAN all wireless devices

exist in the same LAN segment so they are not required to change their IP address until

they leave the campus area.

The development of mature wired campus networks has enabled many

universities to provide both faculty and students with a ubiquitous wireless mobile

computing environment (Barkhuus, 2005). Initially mobile notebook computers filled this

need but gradually tablet PC devices have become more prevalent on college campuses.

Tablet PC technology is the newest development of pen based computing.

Security is an issue with wireless networks. Most wireless standards require some

configuration to provide a secure communication tool. Many university networks require

user authentication before a user is allowed access to the information on a network. A

common frame work to evaluate security is referred to as the CIA (Confidentiality,

Integrity, and Availability) of security (Maconachy, Schou, Ragsdale, & Welch, 2001).

Confidentiality addresses issues associated with the unauthorized disclosure of

information. Integrity refers to the actions taken to prevent and detect the unauthorized


alteration of the information being transmitted. Availability refers to the issues associated

with ensuring that information is reliably provided to authorize users.

The provision of adequate security in wireless networks requires that the network

be configured to require user identity. That can be accomplished by using provisions built

into existing wireless networking standards. 802.11b and g can be set to support security

equivalent to desktop computer workstation by using windows equivalent privacy (WEP)

and wireless application protocol (WAP). User authentication can be accomplished by

using proprietary applications or 802.1 which is built into Microsoft windows XP. Any

organization planning on adding wireless access to a network must evaluate their desired

level of security and implement solutions to ensure a secure computing environment.

The Device

The tablet PC is the current state of pen based computing. Since their initial

release in 2002 TPCs have gradually gained acceptance as a useful tool for educators,

professionals, and casual users. The devices are currently in their third generation and

now possess sufficient computing power to put them on par with the average desktop

computer (Mock, 2004). The TPC device is essentially an x86 microprocessor based

notebook with an active screen digitizer running Windows XP Tablet Edition.

This is a new version of Windows XP with add-ons that support pen-based input.

Microsoft became the leading provider of software for TPC devices when it introduced

the “Tablet PC edition” of Windows XP in 2002 (Microsoft, 2002). The pen can be used

as a navigation tool and an input device that allows users to write on the screen using

digital ink. This digital ink can be stored directly as a graphic or it can be converted to


text. A TPC can execute any program written for the Windows XP operating system

without a translation as needed on most personal data assistant (PDA) devices.

Models available now are either slate or convertible devices. A slate form factor is

similar to a PDA with a larger screen and, usually includes, a detachable keyboard. The

convertible TPC form factor is very similar to a traditional notebook computer with the

screen display attached with a single swivel hinge to the main portable PC body that can

rotate 360 degrees allowing the screen to lie flat on the keyboard, with the screen up, to

emulate a slate arrangement. Several manufacturers have slate and convertible notebook

tablet models available (Tablet PC 2004 Quick Comparison, 2004).

The TPC is usually used by the instructor in a classroom setting as a presentation

device that eliminates the need for a blackboard or whiteboard. The presentation is

typically projected onto a screen using an overhead projection device. With products such

as Microsoft OneNote, PowerPoint, or Journal, the instructor has the ability to use

prepared lecture notes or slides and annotate them on the fly. Teaching faculty at this

university are able to connect to the projection device wirelessly which eliminates the

need of providing a TPC docking station in each presentation classroom. All students also

have a TPC and, at the discretion of the instructor, can assume control of the projector

allowing them to share what is on their TPC desktop with other class participants. This

capability fosters an “active learning” environment where students are actively

participating in the presentation which has been shown to increase learning (Barkhuus,

2005).


Pen Based Computing Background

Many information systems professionals view the TPC to be the next release of an

existing technology. Pen based computing has been around since the 1950’s. The Semi-

Automated Ground Environment (SAGE) air defense system used a “light gun” to

interact with the computer screen (Ray, 2002). Computerized automated drafting

programs in the 1970’s used light pens or pen based drawing tablets to manipulate objects

on the computer screen (Fisher, 1999).

The first pen based desktop computer system was introduced by Wang in 1988.

This computer allowed users to annotate screen captures with “ink” from an electronic

pen on an opaque tablet attached to a serial port of the computer (Francik 1991). This

Wang Freestyle later allowed voice recordings to be attached to a pen event that could

then be emailed to another Wang user.

The first tablet computer introduced was the GRiDPad in 1989 by Jeff Hawkins,

the founder of Palm and Handspring. It was a PC (Intel based) computing device that

used a pen attached by a cord. GRiDPad could recognize hand printed characters and was

used for data collection, such as filling out forms. GRiDPad devices were used by several

state police agencies and some specialized form filler applications (Jones, 2002).

In 1993 Apple jumped onto the pen based computing bandwagon with the

Newton. This device was the first palm based computing device and was marketed by

Apple from 1993 to 1998. The official name of the device was the “MessagePad” but the

devices operating system called Newton (Apple Newton, 2006) became the devices


name. The Apple iPod is the latest version of this device since it operates by using the

original Newton operating system.

The most successful pen based device, prior to the TPC, was the Palm Pilot. Jeff

Hawkins, originally of Grid computing, founded the company in 1992. Later the

company was acquired by US Robotics and soon after that by 3com, the communications

company (Dillion, 1998). The Pilot was essentially a personal organizer device capable of

storing thousands of phone numbers and addresses. The device suffers from a number of

limitations including; small viewing screen, low processor power, insufficient memory,

and crippled applications. But the devices became popular with frequent travelers due to

their convenience features. Computer Business Review reports that makers shipped

nearly one million units in 2005 and projects that nearly two million units will be sold in

2006 (Fujitsu claims top tablet spot in EMEA, 2006).

In the eighteen years since the original GRiD computer introduction computer

hardware has significantly come down in price while micro processors have become

much more powerful. For these reason, and other developments of computer hardware

components, TPC devices are more affordable and powerful. The entry of Microsoft into

the market in 2000 (Gates, 2000) is the most recent event in TPC development. The

combination of a larger screen, increased computing capacity, and the way people use

computers today contribute the success of TPCs. With the ubiquitous availability of the

Internet more users are reading information on their mobile computing devices (Taylor &

Todd, 2001). Students use their devices to read email, news, web pages and other

information. The present college environment is an ideal environment for TPC adoption


(Abbott, 2004). As the TPC models have matured they have become very price

competitive with their notebook counterparts. Jeff Van West, of Microsoft, estimates that

a TPC will cost about $200-700 more than a comparably equipped notebook computer

(West, 2005). When the performance, features, and convenience of use are considered the

TPC is a viable alternative to notebook computing devices.

The Adoption of Technological Innovations

Many information systems (IS) researchers have published on various theories

that could be used to explain the adoption of information technology innovations. These

theories include; the technology acceptance model (TAM) (Davis, Bagozzi, & Warshaw,

1989); the theory of reasoned action (TRA) (Fishbein & Ajzen, 1975); the theory of

planned behavior (TPB) (Ajzen, 1991) among others which are modifications or

developments of these models. The various lines of research are more extensively

discussed in chapters two and three in this dissertation.

The models were developed to help estimate and measure IS innovation success.

An estimate of the success rate of new IS technology implementation projects since the

1980s is about 50% (Westland & Clark, 2000). Explaining the adoption of new

information technologies has been described as the most mature research area in

contemporary information systems research literature (Hu, Chau, Sheng, & Tam, 1999).

Research in this area has generated adoption metrics that can be used to determine the

probability of successful implementation of information system initiatives. The

combination of these metrics into a single model entitled the “Unified Theory of

Acceptance and Use of Technology” (UTAUT) was proposed by several of the fields


leading researchers (Venkatesh, Morris, Davis, & Davis, 2003). The models used for

technology adoption were able to successfully predict the acceptance of an innovation in

only 40%of the cases (Davis et al., 1989) (Taylor & Todd, 2001) (Venkatesh & Davis,

2000). The new proposed model UTAUT has been demonstrated to be up to 70%

accurate at predicting user acceptance of information technology innovations (Venkatesh

et al., 2003). By generating a significantly higher percentage of technology innovation

success the UTUAT is deemed a superior metric than the prior metrics.

Objective of the Study

The objective of this study is to measure the acceptance of TPCs by students at

this university. This setting provides a unique context for the study of adoption of

technological innovations because of the ubiquity of TPCs, and wireless computing in

general on the campus. The primary instrument used to gather adoption data is a web

survey based on the variables defined in the UTAUT and other TAM studies. The

UTAUT constructs are performance expectancy, effort expectancy, social influence, and

facilitating conditions. Other technology research variables included are self efficacy,

attitude towards using technology, and anxiety. These were added to ensure a thorough

investigation of the acceptance of technology in this environment. The survey instrument

was constructed to enable the researcher to determine differences among the students

acceptance of TPC based on their major area of study, their year of attendance at the

university, and their experience with computers including tablet PCs. The acceptance of

TPC was measured using the UTAUT model proposed by Venkatesh and Davis

(Venkatesh et al., 2003). Additional discussion of the appropriateness of this model is


presented in chapter two and three of this dissertation. The primary research question

guiding this research is to understanding the level and rate of technological buy-in

(adoption) at this campus in order to identify the aspects of the environment that most

contribute to the adoption process, and identify the support structures (social,

environmental, etc.) that facilitate this process.

Research Questions

The primary reason for the study is to measure if college students, at this campus,

accept the wireless mobile computing initiative. The second question examined is the

efficiency of technology adoption as determined by the UTAUT model. The third

question addressed by this study is the impact of the various UTAUT variables, and other

variables not included in the UTAUT model, on user acceptance of the TPC.

From a null hypothesis and alternative hypothesis approach, the research

questions are expressed below.

H01: University students, in the study environment, reject the Tablet PC.

H02: The Unified Theory of Acceptance and Use of Technology (UTAUT) does

not predict the successful acceptance of the Tablet PC

H03: The constructs of the UTAUT will not demonstrate an effect on users

acceptance of the tablet PC.

H04: Computer self efficacy does not impact students acceptance of the Tablet

PC.


H05: Anxiety about computer use does not impact students’ acceptance of the

Tablet PC

H06: Students use pattern of the Tablet PC does not impact their acceptance of the

device.

The alternate hypotheses are listed below.

Ha1: University students, in the study environment, accept the Tablet PC.

Ha2: The Unified Theory of Acceptance and Use of Technology (UTAUT) does

predict the successful acceptance of the Tablet PC

Ha3: The constructs of the UTAUT will demonstrate an effect on user acceptance

of the tablet PC.

Ha4: Computer self efficacy does have an impact on students acceptance of the

Tablet PC.

Ha5: Anxiety about computer use does have an impact on students’ acceptance of

the Tablet PC.

Ha6 Students use of the Tablet PC does not impact student’s acceptance of the

device.

The survey questions were mapped to constructs of the UTAUT model with some

constructs included from the TAM to measure the acceptance or rejection of the


individual hypothesis. The mapping of the questions is described in Chapter 3 of this

dissertation.

Significance of the Study

There are several studies that focus on the adoption of information technology

beginning in 1975 with the theory of planned behavior by Ajzen (Fishbein & Ajzen,

1975). The technology acceptance model was proposed by Davis in 1989 (Davis, 1989)

followed by several studies that offer support and suggested modification of the TAM

model (Mathieson, 1991) (Legris, Ingham, & Collerette, 2003). This is a significant

project because it will study a relatively new model, UTAUT, to determine the

acceptance of an information technology initiative by college students. This dissertation

will test the validity of the UTAUT model in a context that is unique to other study

settings.

Many corporate and educational institutions have the existing infrastructure to

support a ubiquitous wireless computing environment. This study will support the

migration to a more mobile computing environment by identifying the structural and

contextual factors that facilitate the adoption of wireless technology and mobile

computing devices. There are many benefits, and some disadvantages, of ubiquitous

computing in a wireless environment. Ubiquitous wireless access to electronic

classrooms and overhead projectors allow teachers to use “all the things teachers can use

to enrich their presentations with multimedia” (Burton, 2004, p. 55).

The study could aid academic institutions with their decisions whether or not to

implement a new information system technology. The findings of this research can


indicate areas that could improve the acceptance of technology implementations.

Although this study does not investigate the financial impact of TPC initiatives it may aid

academic institutions with the decision of adopting this new technology.

Scope of the Study

This study examines the acceptance of TPC by students newly enrolled in the fall

of 2005. The results of the study should not be applied to all university students as other

educational environments may lead to different acceptance decisions. The population size

was chosen to attempt to provide some statistical significance to the study but the best

predictive efficiency of the UTAUT model is only 70%. Technology adoption scenarios

are not correctly predicted in nearly 30% of the cases. An examination of the studies

conducted using technology adoption models reveals that many have been conducted on

undergraduate and graduate students. A fair number of research has been conducted using

IS adoption models on business services such as mobile internet (Pederson & Ling,

2002), text messaging, contact services, mobile payment (Pederson, Nysveen, &

Thorbjornsen, 2003), mobile gaming, and mobile parking services (Pederson & Nysveen,

2003).

Study Context

This study was conducted at a small Midwestern university where freshman and

sophomore students are required to lease a TPC since the fall semester of 2004. In

subsequent years new freshman, and transfer students have been required to lease a TPC.

The result of this initiative is that all freshman, sophomore, and junior students at this


university have TPCs in 2006. This college has a long tradition of supporting data

communication and networking innovations appearing in the top ten most wired

campuses in the year 2000 (Schmidt, 2000).


CHAPTER 2: LITERATURE REVIEW

Information systems research literature is rich with articles about organizational

and individual acceptance of IS innovations. Explaining how end user chooses to accept

technical innovations require psychology based theories. Social Cognitive Theory (SCT)

is a broad psychometric research area that studies the factors involved in individual

decision making. SCT distinguishes itself from traditional social learning theory by

incorporating mental processing (cognition) into the interpretation of observational

learning. Albert Bandura, of Stanford University, has led development of SCT since the

1960’s. His research is important to technology acceptance models because he laid the

foundation allowing us to understand human behavior. Bandura postulates that human

behavior is a triadic, dynamic, and reciprocal interaction of three factors: personal

factors, behavior, and the environment (Bandura, 1977, 1986). While some social

scientists propose that behavior is a result of consequences, SCT postulates that goal-

directed and self-regulation processes play a large part in how we react to different

situations. Furthermore, SCT suggests that there are both direct and indirect effects of

reinforcement t hat learners conscientiously choose. Bandura’s research stimulated

researchers to study techniques for promoting organizational change and measuring the

success of change.

Recently Bandura has moved his primary research to health psychology (Bandura,

2002). A new branch of research has developed that use the principles of social

cognitivism proposed by Bandura, and others, to measure the acceptance of technical

innovations. These theories can be collectively referred to as technology acceptance


models. This chapter discusses the history and progression of technology acceptance

models in depth. The unified theory of acceptance and use of technology (UTAUT)

model including its underlying construction, previous applications, and its similarity to

other implementation models is also discussed.

History of Technology Acceptance Models

Technology Acceptance Models (TAM) have been developed to measure system

use, acceptance, and user satisfaction of those systems (Davis, Bagozzi, & Warshaw,

1989). The Davis model specifically focuses on information systems use and is based on

the theory of reasoned action (TRA) originally introduced by Ajzen and Fishbein in the

early 80’s (Ajzen & Fishbein, 1980) and further refined by Ajzen as the extended TRA in

1991 (Ajzen, 1991).

TRA is a technology acceptance model that can be used to predict behavior in a

wide variety of situations, not just the adoption of information systems technology. Ajzen

states that an individual’s beliefs influence his/her attitude towards various situations.

The users’ attitude joins with subjective norms to shape the behavior intentions of each

individual. This theory was further refined and called the theory of planned behavior

(TPB) which is also titled the extended theory of reasoned action. The TPB is a general

behavior model which can be used to study broader acceptance situations than the TAM

but it has been applied to information systems studies (Mathieson, 1991) & (Taylor &

Todd, 2001).


Figure 1 illustrates the theory of planned behavior. The model helps to explain

how to affect the behavior of people. Ajzen proposed the model to predict deliberate

behavior, because behavior can be deliberate and planned.

Figure 1. Theory of Planned Behavior Note. From Ajzen, I. (1991). “The Theory of Planned Behavior.” by Izak Ajzen, 1991, Organizational Behavior and Human Decision Processes, p. 50, 179-211.

TPB includes many factors, or constructs, used to determine users’ acceptance of

innovations. The three considerations are behavioral beliefs, normative beliefs, and

control beliefs. These are the users core beliefs about the consequences of the action, the

expectations of others, and beliefs about how the user controls, or does not control, the

end result of the behavior. Table 1 further describes the model parameters.


Table 1. TPB Constructs (Ajzen, 1991)

Construct Description Attitude Toward the Behavior The user’s evaluation of the desirability of his or her

using the system

Subjective Norm The individual’s perception of social pressure to

perform the behavior.

Perceived Behavioral Control The individual’s perception of his or her control over

performance of the behavior.

Intention The impact of the first three constructs, attitudes, on

the strength of an individual’s intent to perform the

behavior.

Behavior Belief the subjective probability that the behavior will lead

to a particular outcome

Outcome Evaluation A rating of the desirability of the outcome

Normative Belief The individual’s perception of a referent other’s

opinion about the individual’s performance

Control Belief The perception of the availability of skills, resources

and opportunities

Behavior The outcome of the process.

The complexity of TPB model limits its use in information systems research.

TPBs include more variables than may be important in most information systems


technology implementations (Taylor & Todd, 2001). Some of the variables that have

been removed from the TPB have shown up in more modern models such as the

influence of people considered significant by the participants. These factors are important

to modern acceptance models.

The Technology Acceptance Model

The TAM model, and its derivations, gradually became the accepted model for

research in information systems adoption cases. Debate and refinement of technology

adoption models has continued in IS research literature. The advantage of a TAM is that

it is specifically designed to address the acceptance of IS technology. The TAM model

replaced the first three attitudinal constructs from the TPB with two technology

acceptance measures perceived usefulness and perceived ease of use. This was done in an

attempt to simplify the model making prediction of acceptance easier to predict.

Figure 2. The Technology Acceptance Model. Note. From “Extending the Technology Acceptance Model: The Influence of Perceived User Resources,” by Mathieson, K., Peacock, E., & Chin, W. W. (2001). The Data Base for Advances in Information Systems, 32(2), p. 90.


The two theoretical components that are the foundation of Davis’s TAM model

(Davis et al., 1989, p. 985). The first is perceived usefulness (PU) and perceived ease of

use (PEOU). Davis defines usefulness as “the degree to which a person believes that

using a particular system would enhance his or her job performance.” Davis goes on to

define perceived ease of use as “the degree to which a person believes that using a

particular system would be free of effort” (Davis et al., 1989, p. 985).

A recognized limitation of TAM is that it does not take into consideration any

barriers that would prevent an individual from adopting a particular information systems

technology (Taylor & Todd, 2001). These variables that are not included in TAM are

system design characteristics, training, support, and decision maker characteristics.

The research studying modifying factors to Davis’s technology acceptance model

attempt to improve upon it by adding user resources and restrictions to the model.

Mathieson termed these factors as external control factors. These external factors include

subjective norm, voluntariness, job relevance, output quality, and result demonstrability

(Mathieson, 1991, p. 87). By adding robustness to the model the researchers hope to

improve the predictive value of the tool. Recently Venkatesh et al. published

improvement in the prediction of acceptance from 17% to 42% (Venkatesh et al., 2003, p.

439).

Venkatesh and Davis attempted to incorporate these discussions of external

factors into a TAM model when they proposed a model in 2000 extending the TAM.

Their new model incorporates additional theoretical variables that include social


influence processes and cognitive processes. Figure 3 illustrates the result of the extended

model.

Figure 3. Extension of the Technology Acceptance Model. Note. From. “A Theoretical Extension of the Technology Acceptance Model: Four Longitudinal Field Studies,” by Venkatesh, V., & Davis, F. D. (2000) Management Science, 46(2), p. 190.

Examination of current literature on technology acceptance does indicate that

additional factors need to be included that were not in original TAM models. Examples

of the types of variables are demographics, managerial knowledge, social factors,

environmental characteristics, and task-related characteristics (Pijpers, 2001). Some

examples of possible additional factors are the motivational factors introduced by

Vallerand (Vallerand, 1997). The perception that users will want to perform an activity

“because it is perceived to be instrumental in achieving valued outcomes that are distinct

from the activity itself, such as improved job performance, pay, or promotions are


extrinsic while the perception that users will want to perform the activity “for no apparent

reinforcement other that the process of performing the activity per se” (Davis et al., 1992,

p. 1112).

Information Systems research has validated the TAM, and its derivations. Table 2

presents an abbreviated list of researchers and dates that have used the TAM model to

study their technology acceptance research problems.

Table 2

TAM Research (Legris, Ingham, & Collerette, 2003) Year Author Model Used ________________________________________________________________________

1989 Davis TAM

1992 Davis & Bagozzi Extended TAM

1991 Mathieson TAM & TPB

1995 Taylor& Todd TAM + subjective norm

1997 Jackson TAM + validation tools

1998 Bajaj et. al. TAM + loop back adjust

1999 Hu et al. TAM

2000 Venkatesh & Davis TAM + subjective norms

2002 Hwang & M. Yi TAM + goal orientation, CSE

This body of research validates and extends the application of TAM. But TAM is

only capable of predicting technology adoption success between 30% (Meister &

Compeau, 2002) and 40%of the cases (Venkatesh & Davis, 2000). As a result of this,

researchers have searched for better technology acceptance models that can deliver a


higher prediction of success (Legris et al., 2003) (Plouffe, Hulland, & Vandenbosch,

2001). The call for a modified model that incorporates both human and social variables

led to the development of an extended TAM and eventually the Unified Theory of

Acceptance and Use of Technology model.

The Unified Theory of Acceptance and Use of Technology Model

The result of the discussion and debate over the best technology acceptance tool

resulted in several tools being available to IS researchers. As many as eight models have

received support in recent literature. The September 2003 issue of the MIS Quarterly has

addressed this issue (Venkatesh, Morris, Davis, & Davis, 2003). This research article

examines the current state of knowledge in this area comparing similarities and

differences in the current models. The Unified Theory of Acceptance and Use of

Technology (UTAUT) model resulted from this study. Dr. Venkatesh et al. attempted to

validate the tool by testing UTAUT on historical data from previous TAM researchers.


Figure 4. UTAUT Model (Venkatesh et al., 2003, p. 447) Note. From ”Theories Used in IS Research” Website, www.istheory.yourku.ca.

Figure four illustrates the UTAUT model that compiles all the variables found in

the eight existing models and a selected subset of additional constructs. Venkatesh et al.

then validated the model using both existing data, from the previous TAM studies, and

data obtained in two new surveys. The UTAUT model postulates that three direct

variables (performance expectancy, effect expectancy, and social influence) determine

the behavioral intent of technology use and a direct determinant of usage behavior in

facilitating conditions. The model integrates four moderating factors (gender, age,

experience, and voluntariness) having varying influence the primary constructs. In

summary, the UTAUT model has condensed the 32 variables found in eight existing

models into four main effects and four moderating factors.


Table 3. UTAUT Components Construct Description Performance Expectancy (PE) Degree to which an individual believes that using the

system will help attain gains in job performance

Effort Expectancy (EE) The degree of ease associated with the use of the

system

Social Influence (SI) The degree to which an individual perceives that

important others believe he or she should use the new

system

Facilitating Conditions (FC) The degree to which an individual believes that an

organizational and technical infrastructure exists to

support use of the system.

Table 4. UTAUT Moderators Construct Description Gender Gender roles have a strong psychological basis and

are enduring.

Age Age has an effect on attitudes.

Experience Effort is expected to be more important in the early

stages of new behavior.

Voluntariness of Use Is usage voluntary or mandated


The moderating factors have influence on the four model constructs. Gender and

Age influence performance expectance, effort expectance, and social influence. Age and

experience moderate the facilitating conditions. Experience moderates effort expectancy,

social influence, and facilitating conditions. Voluntariness of use moderates the effect of

social influence in UTAUT. The combinations of the constructs and moderating factors

have increased the predictive efficiency to 70%, a major improvement over previous

TAM model rates (Venkatesh et al., 2003).

UTAUT Supporting Research

Research is currently being conducted to test the UTAUT model. Dr. John

Anderson and Dr. Paul Schwager, of East Carolina University are examining an

application of the UTAUT model that is being presented at AMCIS conference this

coming summer in Acapulco, Mexico (Anderson & Schwager, 2006). Dr. Christer

Calrsson et al (Carlsson, Carlsson, Hyvönen, Puhakainen, & Walden, 2006) have studied

the adoption of wireless mobile communication in Europe with UTAUT. Dr. Monica

Garfield used the UTAUT tool to analyze the acceptance of table PCs at Bentley College

(Garfield, 2005).


CHAPTER 3: RESEARCH DESIGN

The objective of this study was to measure the acceptance of TPCs by college

students at a small Midwestern University. The expectations are that the survey will

provide evidence of the acceptance of the devices by students at this research site. The

research questions guiding this study are:

H01: University students, in the study environment, reject the Tablet PC.

H02: The Unified Theory of Acceptance and Use of Technology (UTAUT) does

not predict the successful acceptance of the Tablet PC

H03: The constructs of the UTAUT will not demonstrate an effect on users’

acceptance of the tablet PC.

H04: Computer self efficacy does not impact students acceptance of the Tablet

PC.

H05: Anxiety does not impact student’s acceptance of the Tablet PC

H06: Students use pattern of the Tablet PC does not impact their acceptance of the

device

The alternate hypotheses are listed below.

HA1: University students, in the study environment, accept the Tablet PC.


HA2: The Unified Theory of Acceptance and Use of Technology (UTAUT) does

predict the successful acceptance of the Tablet PC

HA3: The constructs of the UTAUT will demonstrate an effect on user acceptance

of the tablet PC.

HA4: Computer self efficacy does have an impact on students acceptance of the

Tablet PC.

HA5: Anxiety does impact student’s acceptance of the Tablet PC

HA6: Students use of the Tablet PC does not impact student’s acceptance of the

device.

The participants of this study were college students who had been using the

device since the fall of 2005. The sample of student participants is over three hundred

students. The population of students who enrolled at the university for the fall semester of

2005 was 356. The number of subjects available to be surveyed was lower due to

dropouts and transfers to other educational institutions and the timing of the sampling

which was at the beginning of the summer semester. The researcher approached the

instructors of the various courses that would be a normal progression for this student

population and obtained their permission to conduct the survey during a scheduled class

period.

The instrument used to gather adoption data was a web survey based on the

variables defined in this study. The survey tool presented questions based on the Unified


Theory of Acceptance and Use of Technology (UTAUT) Model, and since this is a

relatively new model, the survey included questions addressing constructs that were

present in the original Davis Technology Acceptance model (TAM) dealing with

computer self efficacy, anxiety, and usage. These questions were added as suggested by a

group of technology acceptance

researchers.

Figure 5: Dissertation Research Model (Moran, 2006) Note. Figure created with PLS-Graph from the hypothesized research model.

The UTAUT model theorizes that four constructs have a significant determination

on user acceptance of IT innovations (Venkatesh et al., 2003). These variables are


performance expectancy, effort expectancy, social influence, and facilitating conditions.

These constructs are moderated, in varying degrees, by gender, age, experience, and

voluntariness of use. Figure 5 graphically displays the relationship between interacting

variables used for this study.

Performance Expectancy

Performance expectancy is defined as the degree to which the student believes

that using the TPC will help him or her to accomplish the various academic assignment at

a typical university. Venkatesh (Venkatesh et al., 2003) postulates that performance

expectancy is the strongest of the four constructs in his model. This theory is support by

other researchers publishing papers on acceptance models (Agarwal & Prasad, 1998,

Compeau & Higgins, 1995; Taylor & Todd, 2001) [APA style references indicate that

multiple references given at the end of a sentence be enclosed within the same

parentheses]. Performance expectancy will be measured using ten questions that focus on

task accomplishment. These questions are only slightly modified versions of questions

used consistently in most prominent research publications dealing with TAM. Recent

studies have determined that this construct may have a gender bias (Lynott &

McCandless, 2000, Venkatesh et al., 2003) determined that the effect performance

expectancy was moderated by age and age such that it was more important to younger

male workers in particular.


Effort Expectancy

Effort expectancy is defined as the degree of ease associated with the use of the

TPC. This construct will be measure by asking eight questions based on the common

current literature set. Some researchers suggest that there are gender differences related to

roles in life (Lynott et al. 2000). Researchers in technology acceptance have addressed

these criteria (Venkatesh & Morris, 2000) thus a gender effect is expected in this study

and will be measured by determining the gender of the survey participants.

Social Influence

Social influence is the degree to which an individual perceives that important

others believe he/she should use the new systems. This construct deals with the notion

that the individuals behavior is influenced by the way in which they believe others will

view them as a result of using the technology. The importance of social factors becomes

more significant in mandated environments is postulated by Venkatesh and Davis in their

2000 publication (Venkatesh & Davis, 2000). In mandatory adoption settings social

influence appears to be significant only in the early stages with its effect eroding over

time.

Facilitating Conditions

Facilitating conditions are defined as the degree to which an individual believes

that an organizational and technical infrastructure exists to support use of the TPC. Five

common TAM research questions will be used to measure this construct. Venkatesh et al.


published in 2003 that this variable was not significant as a determinant of intention.

However this variable was retained because of discussion pertaining to its importance in

other publications (Taylor & Todd, 1995).

Supplementary Variables

The TAM model is a simpler model than the UTAUT but includes constructs that

address other personal and situational variables that may be pertinent in this study. The

additional variables added to the study from TAM are; attitude toward using technology

(ATUT), self-efficacy (SE), and Anxiety (ANX). These variables will be measured by

asking five to six questions dealing with each area that have been modified to address

tablet PC use. In addition, the researcher is interested in the usage of the TPC as a pen

based mobile computing platform so questions are asked to determine the usage of the

device. These questions were constructed with the help of a group of researchers

interested in tablet PCs in an educational environment.

Pilot Study

A pilot study was conducted to test the survey instrument with a small group of

upper class students enrolled in a one credit FrontPage application class. The survey used

in the pilot study contained approximately sixty statements pertaining to the various

constructs included in the dissertation model. The researcher asked the participants to

complete the sample survey during normal class meeting time. Following the group

survey completion, which was approximately fifteen minutes, the group of participants

and the researcher discussed the survey instrument for areas that could be improved.


Areas of discussion included; confusion about what the survey was; confusion about the

wording of some statements, and the length of the survey.

Following the first test of the survey the researcher made the recommended

changes to the instrument. The preliminary material, before participants entered the

survey, was reworded to clarify the reasons for the research and the confidentiality of the

participants. Six survey statements that were thought to be confusing or redundant were

removed and many statements were reworded to clarify the statements. Another result of

the test was reorganization of the questions into construct groups allowing a full set of

statements to be answered on one screen. The final addition was the elimination of all the

submit buttons and replacing them with a submit command button at the end of the

survey.

The modified survey tool was re-evaluated by the pilot group in a subsequent

class session with discussion following the second trial run. The pilot group was pleased

with the changes and suggested a desire that participants may wish to know the final

result. This resulted in a research link being established on the researcher’s website that

students can visit to view the survey results and publications based on data obtained by

the survey. Survey modifications were based on information found at web survey design

hosted by San Diego State University (Hoffman, 2006). Specifically the layout of the

final survey design is based on information found in an excellent book by Don Dillman

of Washington State University (Dillman, 1999).


The Survey

After the survey instrument and the scope of the research had been approved by

the university Human Subjects Committee, then all subsequent survey participants were

given identical survey forms. Participants were assured response anonymity by not being

required to provide identifying information on the survey. The survey instrument was

made available to the participants via the World Wide Web. An increasing number of

research studies are being conducted in this fashion (Gall, Gall, & Borg, 2003). There is a

tremendous benefit of using a web survey over paper survey because the responses are

transferred automatically to a database eliminating clerical errors. The researcher

prepared both the survey instrument and the response database in preparation for survey

administration.

The survey instrument was constructed using Hypertext Markup Language

(HTML) and Active Server Pages (ASP), and the database utilized Access, from the

Microsoft Office Suite, to capture and store the data. The survey tool was demonstrated

to faculty and staff both at research university and another tablet university, where it was

critiqued resulting in a more streamlined tool. Initially the survey was deemed to be too

long and several questions were removed in an attempt to eliminate fatigue from

adversely affecting survey results.

Based on the expert committee input, the survey layout was changed to include

multiple related questions into a single screen and the length of the survey was limited to

five screens. The survey instrument was beta-tested by administration to sixteen students

who were enrolled in an intermediate Microsoft Access course. These upperclassmen also


considered the design and format of the survey and made suggestions that resulted in

limiting the surveys length to five screens.

The questions or statements used in survey instrument can be examined in

appendix B. Figure 6 illustrates the webpage link utilized in the study.

Figure 6. Research Page Link (http://www.homepages.dsu.edu/moranm/Research/research_page.htm)

Survey participants accessed the tool by clicking on the “TabletPC Survey” link

on the lower left hand corner of the figure. When this link was selected the screen

displayed in Figure 7 appeared explaining the purpose of the survey. When the

participants clicked that link they are indicating informed consent. Participants were

allowed to exit the process or enter the actual survey instrument. Survey participants were


in a class setting and were guided to the instrument by the survey coordinator who is the

primary researcher.

Figure 7. Web survey link


The screen below (Figure 8) was displayed to provide survey participants with

links to responsible individuals, and Capella University, if the participants had any

questions, or concerns, about the survey or the process. If participants clicked on a link

they were directed to the appropriate email address for the responsible person or a

mailing address and phone number were displayed.

Figure 8. Questions & Concern Information


When survey participants clicked the “TabletPC Survey” button they were taken

to the Website containing the survey instrument. Figure 8 shows the first screen the

participants saw containing the first set of questions.

Figure 9. Web Survey Instrument

After survey completion the participants submitted their selections to the survey

database by selecting a command button. The instrument and database are stored on a

secure web server administered by computing services staff at the participant university.

The survey questions were broken down by construct, survey layout, and human

subjects’ committee approval documentation are available from the researcher.


Sample Size

One method of determining the necessary sample size for a survey is based on the

error the researcher is willing to accept. It is common in the social sciences to try to

achieve a statistical power of at least 95% confidence or an alpha of 0.05. Using the

method and a table provided by Cohen (Cohen, 1988, p. 52) it was determined that a

sample of at least 175 participants would be needed to achieve 95% confidence. One of

the benefits of using PLS-Graph is that it can resample the initial data set enlarging it thus

reducing overall sample requirements. Guidelines provided with PLS-Graph recommend

a sample size equal to the larger of two possibilities: (1) ten times the number of

indicators on the most formative construct, in this study ten times the ten indicators of

performance expectancy or one hundred participants, or (2) ten time the largest number

of antecedent constructs used to determine a dependent variable, in this study ten times

six, the number of constructs used to determine behavior intent. In all cases the 263

survey participants is greater than the calculated sample size.

Survey Sample Population

This survey tool was administered to 361 students at the research site, refer to

appendix A for the exact number of students in each class. The participant pool were

students enrolled in the following courses

Introduction to computers (CSC 105) – about one hundred eighty students

Introduction to visual basic (CIS 130)– about one hundred fifty students

Computer hardware and networks (CIS 351)– about ninety students


Computer science 2 (CSC 150)– about forty students

Management of Information Systems (CIS 325) – about thirty students

These classes were selected for the survey because they contained students who

had been using the TPC since August 2005; it was postulated that most of these students

were introduced to the device during the fall semester of 2005. Students who are enrolled

in more than one of the classes surveyed were instructed to not complete the survey by

the survey administrator. The survey was administered during normal class meeting times

by the researcher who verified that no participant completed the instrument more than

once. The survey was conducted during normal class sessions during the last ten minutes

of class using each student’s TPC. The time required to complete the survey was five to

seven minutes.

Human Subjects Protection

The human subjects research committee (HSC) at the subject university

determined that this research is exempt from the rules governing the protection of human

subjects because of the method used to obtain and record the information. This exemption

is based on information published in the Code of Federal Regulations Title 34

Department of Education PART 97 - Protection of Human Subjects (Code of Federal

Regulations Title 34 Department of Education PART 97 - Protection of Human Subjects,

2004). The rules of exemption state that

“(2) Research involving the use of educational tests (cognitive, diagnostic, aptitude, achievement), survey procedures, interview procedures or observation of public behavior, unless:


(i) Information obtained is recorded in such a manner that human subjects can be identified, directly or through identifiers linked to the subjects; and

(ii) Any disclosure of the human subjects' responses outside the research could reasonably place the subjects at risk of criminal or civil liability or be damaging to the subjects' financial standing, employability, or reputation.” (PART 97.101, paragraph b2)

The exemption was granted because survey is a common educational tool

for this university and no information will be recorded that could be used to

identify the subjects. The survey information will be maintained and protected by

the primary investigator for the duration of its use and retained for five years.

The two class sections taught by the researcher have enrollments of thirty

six of the approximately three hundred survey target audience. These participants

will be compared to the results from the rest of the survey to determine if they can

be included in the statistical calculations based on the similarity of survey

responses. The difference will be evaluated in chapter four by comparing the

mean, standard deviation, and variance, using a t-test, of the two groups. Since

these sections are under the instruction of the primary investigator there may be

an influence from the instructor.

Survey Questions

Venkatesh et al. (2003) in the September issue of the MIS Quarterly used a survey

comprised of thirty one questions to support the UTAUT model. The questions were

derived from the eight models analyzed in the paper. The Venkatesh team used the

survey instrument in two studies to validate the new model. They studied a financial


service company that was primarily involved in research and a retail electronics company

whose primary functional area was customer service.

The survey instrument contained four questions addressing each of the technology

acceptance areas. The eight variables are performance expectancy, effort expectancy,

attitude toward using technology, social influence, facilitating conditions, self-efficacy,

anxiety, and behavior intention to use the system. The validation process used in the

Venkatesh publication eliminated three of these variables which were found to not be

significant determinants of use. The Venkatesh eliminated variables are attitude toward

using technology, self-efficacy, and anxiety.

The survey used at the research site utilizes a model similar to that used in the

Venkatesh publication in that it contains questions pertaining to the constructs of the

UTAUT. This dissertation survey includes questions pertaining to constructs that the

researcher believes are important in this environment. There are also questions measuring

the four moderating UTAUT items, questions pertaining to the TAM model, and question

that address the participants age and computer experience including computer experience

prior to their enrollment in college. Survey participants were asked to indicate their

response to each statement using a seven item likert scale with one representing a strong

disagreement and seven being a strong agreement with the statement.


Table 5 lists the questions that are being used to measure performance

expectancy.

Table 5. Performance Expectancy Questions Question Item Using the Tablet PC in my classes would enable me to accomplish tasks more quickly [font size not

consistent with the other table]

Using the Tablet PC in my classes would hamper my performance (reverse scored)

Using the Tablet PC in my classes would increase my productivity

Using the Tablet PC in my classes would hamper my effectiveness in class (reverse scored)

Using the Tablet PC in my classes would make it easier to do my homework

Using the Tablet PC in my classes would hamper the quality of the work I do (reverse scored)

Using the Tablet PC in my classes would cause my classmates perceive me as competent

Using the Tablet PC in my classes would increase the instructors respect for me

Using the Tablet PC in my classes would decrease my chances of getting a good grade (reverse scored)

Using the Tablet PC in my classes would be useful in my classes


Table 6 lists the statements that are being used to measure effort expectancy with

appropriate responses on a seven item Likert scale.

Table 6. Effort Expectancy Questions Question Item Learning to operate the Tablet PC is easy for me.

I find it easy to get the Tablet PC to do what I want it to do

My interaction with the Tablet PC would be clear and understandable

I find the Tablet PC to be flexible to interact with

It is easy for me to become skillful at using the Tablet PC

I find the Tablet PC easy to use

Using the Tablet PC takes too much time from my normal duties.

Working with the Tablet PC is so complicated, and difficult to understand

Table 7 lists the statements that are being used to measure attitude toward using

technology with appropriate responses on a seven item Likert scale.

Table 7. Attitude toward using technology Questions Question Item Using the Tablet PC is a good idea

I dislike the idea of using the Tablet PC (reverse scored)

Using the Tablet PC is pleasant

The Tablet PC makes schoolwork more interesting

Using the Tablet PC is fun

I like working with the Tablet PC


Table 8 lists the statements that are being used to measure social influence with

appropriate responses on a seven item Likert scale.

Table 8. Social Influence Questions Question Item People who influence my behavior think that I should use the Tablet PC.

People who are important to me think that I should use the Tablet PC.

Professors in this university have been helpful in the use of the Table PCs

My advisor is very supportive of the use of the Tablet PC for my class.

In general, the university has supported the use of the Tablet PC.

Having the Tablet PC is a status symbol in my university.

Table 9 lists the statements that are being used to measure facilitating conditions

with appropriate responses on a seven item Likert scale.

Table 9. Facilitating Conditions Questions Question Item I have the resources necessary to use the Tablet PC.

I have the knowledge necessary to use the Tablet PC.

The Tablet PC is not compatible with other computer systems I use.

The help desk is available for assistance with the Tablet PC difficulties.

Using the Tablet PC fits into my work style.


Table 10 lists the statements that are being used to measure behavioral intention

with appropriate responses on a seven item Likert scale.

Table 10. Behavioral Intention Questions Question Item Whenever possible, I intend to use the Tablet PC in my studies

I perceive using the Tablet PC as Involuntary

I plan to use the Tablet PC in the next three months.

To the extent possible, I would use Tablet PC to do different things (school or not school)

related

To the extent possible, I would use Tablet PC in my studies frequently.

In addition to the statements regarding constructs included in the UTAUT model

this research added statements that address additional variables included in many

technology acceptance models which are self efficacy, anxiety, and usage. The researcher

postulates that including these construct will both strengthen the study and improve the

support for the UTAUT.


Table 11 list the statements used to measure self efficacy, responses are in a seven

item likert scale with one representing a strong disagreement and seven being a strong

agreement with the statement..

Table 11. Self Efficacy Question Item I could complete a task using the Tablet PC if there was no one around to tell me what to

do as I go.

I could complete a task using the Tablet PC if I had seen someone else demonstrate how

it could be used

I could complete a task using the Tablet PC if I could call someone to help if I got stuck

I could complete a task using the Tablet PC if I had a lot of time to complete the job.

I could complete a task using the Tablet PC if J had just the built in help facility for

assistance

Table 12 list the statements used to measure anxiety with appropriate responses

on a seven item Likert scale.

Table 12. Anxiety Questions Question Item I feel apprehensive about using the Tablet PC.

It scares me to think that I could lose a lot of information by using the Tablet PC and

pressing the wrong key.

I hesitate using the Tablet PC for fear of making mistakes I cannot correct

The Tablet PC is somewhat intimidating to me.


Table 13 list the statements and questions used to measure usage and other

information with appropriate responses on a seven item Likert scale. The use questions

were developed from input provided by the faculty experts who helped validate the

survey and are referred to in the next discussion.

Table 13. Usage and Other Information Questions Question Item I use my TPC in Slate mode.

I use my TPC stylus for navigation.

I use my TPC primarily as a notebook computer.

I use Windows Journal

I am a male/female

I am currently at DSU.

A B C D E Freshman Sophomore Junior Senior Grad Student

My major is represented by the ____________ college at Dakota State University.

A B C D E Arts & Sciences Business& Info Sys Education. Graduate. Other

I began using computers regularly in _______________ school.

A B C D E

Elementary Middle High. College Do not use

Approximately how many months have you been using the Tablet PC?

A B C D E One Three – Six Twelve More than Twelve

My age at the time of this survey is __________.

A B C D E

Seventeen Eighteen – Nineteen twenty twentyone


The questions are similar to those used by the researchers cited as leading

authorities in technology acceptance research. The wording has been modified to address

the Tablet PC device. Some questions are constructed to check the validity of the

participants’ submissions by requiring an opposite response in the separate question. The

survey instrument was constructed with the help of Dr. Omar El-Gayar, vita available

from the primary investigator. Other researchers were contacted but did not participate in

the process.

Data Analysis Methodology

This is a modification of the UTAUT model which identifies interactions between

technology acceptance constructs. The most common method used for information

systems research for moderator analysis is regression analysis. Partial least squares is

used frequently in the research literature concerning psychometric evaluations. The

dissertation model consists of nine latent variables, or constructs. The latent variables

performance expectancy, effect expectancy, social influence, self efficacy, attitude

toward using technology, and anxiety, which are used to determine behavioral intent.

Behavioral intent and facilitating conditions are the two determinants of use behavior.

Venkatesh et al., the developers of the underlying UTAUT model, also include four

moderating factors (gender, age, experience, and voluntariness) that have varying

influence on the constructs. In addition to the constructs associated with the UTAUT

model this study will add attitude toward the use of technology, self efficacy, and

anxiety.


The statistical analysis method used for this dissertation was partial least squares

(PLS), a powerful second generation statistical technique of covariance based structural

equation modeling. PLS has been used by many researchers in the technology acceptance

field (Compeau & Higgins 1995, Venkatesh et al. 2003, & others). The tool used for the

analysis was PLS Graph (PLS Graph, Version 2.91.03.04) obtained from Wynne W. Chin

of the University of Houston. The software was used to determine the validity of the

various measurements or questions. The questions contained in the survey instrument

were evaluated for variance (R2) and retained if the variable had a variance greater than

0.7. This is consistent in current literature outlining technology adoption research. PLS

was used to study the assessment of latent variables and can also weight the relationship

between the questions used to determine unobservable model constructs.

The dissertation model was evaluated to measure the inclusion of the various

statement response variables used to contribute to the model constructs. PLS is preferred

in applications where the constructs are measured primarily by formative indicators such

as those used in this study. PLS is supported in a recent publication by Jarvis et al.

(Jarvis, MacKenzie, & Podsakoff, 2003) and Chin (Chin, Marcolin, & Newsted, 2003).

A complete analysis of the model required an examination of both the goodness

of fit criteria and the factor loading indicators. The goodness of fit indices measured how

well the variable parameter estimates were able to reproduce the sample covariance

matrix. The technique did this by taking the presented model as true and modifies the

parameter estimates until the covariance difference between the parameter estimates and

the sample is minimized.


Measures

Survey participants responded to questions regarding the TPC by recording their

agreement to the statement with a seven-point likert scale where a one represents strongly

disagree and a seven represents a strong agreement response. Internal consistencies of the

variables were determined using Cronbach’s alpha after the negative scored items were

converted to the correct scale. Table 14 contains the measures of scale reliabilities for the

various construct variable groups. Generally reliability numbers greater than 0.6 are

considered acceptable in technology acceptance literature (Zhang, Li, & Sun, 2006).

Table 14. Scale Reliabilities Construct Number of Questions Reliability of group Performance Expectancy 10 0.84

Effort Expectancy 8 0.89

Attitude Toward Using Technology 6 0.89

Social Influence 6 0.76

Facilitating Conditions 5 0.70

Behavioral Intention 5 0.80

Self Efficacy 5 0.84

Anxiety 4 0.84

Usage 4 0.67

The data was collected and the reliability of the individual constructs were

evaluated using partial least squares. Figure 5 illustrates the dissertation model and the


nine constructs contained in the model. These constructs are modified, in a varying

degree, by moderating variables which are gender, age, experience, and voluntariness of

use. In context of this study it is hypothesized that six factors; performance expectancy

(PE), effort expectancy (EE), social influence (SI), attitude towards using technology

(ATUT), self efficacy (SE) and anxiety (ANX) together determine behavioral intent (BI).

Behavioral intent with facilitating conditions (FC) determines use behavior (B). Stated in

a formula the relationship can be expressed as displayed in figure 10.

Figure 10. Research Relationship Equations

At the same time the moderating factors, to a varying degree, have an effect on

each of the constructs stated in the two equations above. There are unobservable factors

that have an effect on the model and are measured in the determination of the variables

and are included to arrive at a good estimation of acceptance. Partial least squares has

been documented to be able to account for these factors and calculate the correct weights

allowing researchers to study the strength and direction of the relationships between

model variables and the relationships between moderating variables simultaneously

(Fornell & Bookstein, 1982, Lohmoller, 1989, Igbaria, 1994, & Chin, 2003).

B = w1BI + w2FC

BI= w3PE + w4EE + w5SI + w6ATUT +w7SE +w8ANX


CHAPTER 4: DATA ANALYSIS

Data Sample

The participants for this study were students enrolled at a Midwestern university

that adopted a tablet PC initiative requiring all students to lease a tablet personal

computer (TPC) device when they enrolled at the institution. A result of the initiative is

that all students who participated in this study had already pre-selected tablet PC use as

criteria of enrollment at this university. Because of this situation the results are skewed

toward TPC use and do not follow a normal population distribution.

The participants of the survey who were enrolled in the two sections taught by the

researcher (20 participants) were compared to the individuals in all the other sections

(243 participants) to determine if there were any statistically significant differences.

Analysis with SAS 9.0 did not find any statistically significant difference at the 95%

confidence level. The variance of the smaller group was greater than the larger group

which should be expected due to the low numbers involved and the lowest p-value was

0.076. The 95% confidence threshold used to prove that the groups are significantly

different is 0.05.

Data was collected from students in all the sections of courses thought to be most

likely enrolled in by students in their first year of table PC use. The available participant

pool was about 360 individuals enrolled in the selected courses (see Appendix A for

complete detail of sections). The database recorded responses from 268 participants

resulting in a response rate of 74%. Five survey submissions were disqualified due to


incomplete submissions, meaning that less than 75 %of the survey was completed

(Moore & Benbasat, 1991). A total of 263 responses were included in data analysis and

model construction.

The general demographics of the survey participants for gender showed that 119

survey participants were female (45%) and 142 identified themselves as male (54%),

three participants did not indicate a gender. The mean age of the participants was twenty

two with ninety four percent (94%) being younger than twenty four. One hundred and

twenty seven (48%) reported themselves as freshman students, sixty-six (25%) reported

themselves as sophomore students, forty-nine (18%) reported themselves as junior

students, sixteen (6%) reported themselves as senior students, and five did not indicate

any student classification (see Figure 11).

Class Status

Freshman48%

Sophomore25%

Junior19%

Senior6%

None2%

Figure 11. Survey Participants Class Status.


Figure 12 illustrates survey respondents by college affiliation. Sixty one students

(23%) indicated their major to be affiliated with the College of Arts & Sciences, one

hundred and twenty two students (46%) indicated their major to be affiliated with the

College of Business & Information Systems, sixty two students (24%) indicated their

major to be affiliated with the College of Education, and eighteen students (7%) indicated

that they had not selected a college major yet.

College Affiliation

Arts&Sc23%

BIS46%

Education24%

Other7%

Figure 12: Survey Participants College Affiliation (N=263)


Figure 13 summarizes the participant’s response to the question of when their first

computer use occurred indicated that ninety six participants (36%) started using

computers in elementary grades, ninety seven participants (37%) started using computers

in during the middle school grades, thirty seven participants (14%) started using

computers in during their high school years, and thirty three participants (13%) began

using computers when they entered college. This data suggests few novice computer

users in the sample.

First Computer Use

Elementary36%

Middle37%

High School14%

College13%

Figure 13. Survey Participants First Computer Use (N=263).


Thirty four (13%) survey participants indicated that their first use of a tablet

personal computer (TPC) occurred in the past three month period, one hundred

participants (38%) had been using the TPC for six months, forty seven participants (18%)

had been using TPC for nine months, eight participants (3%) reported using a TPC for

one year, and seventy four participants (28%) reported TPC use of more that one year.

Figure 14 graphically summarizes this information.

Length of Tablet PC Use

Three mo13%

Six mo38%

Nine mo18%

12 mo3%

> 12 mo28%

Figure 14. Survey Participants Length of Use of Table PC (N=263).

The fact that over a quarter of the participants reported tablet PC use of over a

year was not expected by the researcher but will allow discussion of the differences

between new TPC users and this more experienced group. The individuals in this group,

who indicated use of the TPC for more than one year, voluntarily joined the TPC

initiative by obtaining their own devices. When the university started the TPC initiative

only first and second year students were required to participate.


Data Validation

The data was examined with SAS 9.0 by David Zolnowsky, vita available from

the primary researcher, for basic statistics. The determination of skewness, and kurtosis

was included because previous literature in psychometric studies is often identified as

non-normal (Hair, Tatham, Anderson, & Black, 1998). Table 15 contains the variable

names for all the construct indicators, and the mean, standard deviation, skewness, and

kurtosis of the survey responses. PLS-Graph documentation refers to the construct

variables as indicators so throughout this dissertation variables and indicators indicate the

same information. Skewness is a measure of the symmetry of a distribution. A skewness

value of zero indicates that a distribution looks the same on the right and left of the center

point. A negative skewness value indicates that the left histogram tail is long compared to

the right histogram tail. A positive value indicates that the right tail is longer than the left

tail (Newbold, Carlson, & Thorene, 2007). Kurtosis measures the relative peak of the

mean in a distribution. Data distributions with high kurtosis have a high peak near the

mean with a heavy tail in one direction. Low kurtosis would be a flat top near the mean.

The tablet PC survey data is significantly peaked compared to a normal distribution

which would have a value of three. This type of a peaked distribution is called a Weibull

distribution. Unpaired T tests are valid for evaluating Weibull populations (Johnson,

Kotz, & Balakrishnan, 1994).


Table 15a. Statistical Analysis of the Variables. Standard Variables Mean Deviation Skewness Kurtosis Performance Expectancy 1 5.84 1.09 -1.96 5.84Performance Expectancy 2 4.75 1.55 -0.50 4.75Performance Expectancy 3 5.39 1.21 -1.05 5.39Performance Expectancy 4 4.38 1.64 -0.21 4.38Performance Expectancy 5 5.92 1.07 -1.78 5.92Performance Expectancy 6 5.23 1.47 -0.94 5.23Performance Expectancy 7 4.24 1.32 -0.36 4.24Performance Expectancy 8 4.06 1.27 -0.22 4.06Performance Expectancy 9 4.97 1.52 -0.63 4.97Performance Expectancy 10 5.82 1.04 -1.51 5.82 Effort Expectancy 1 5.96 1.13 -1.80 5.96Effort Expectancy 2 5.77 1.07 -1.26 5.77Effort Expectancy 3 5.82 1.03 -1.12 5.82Effort Expectancy 4 5.79 1.01 -1.30 5.79Effort Expectancy 5 5.93 0.98 -1.24 5.93Effort Expectancy 6 5.99 1.00 -1.64 5.99Effort Expectancy 7 5.08 1.66 -1.00 5.08Effort Expectancy 8 5.79 1.54 -1.69 5.79 Attitude Toward Using Technology 1 5.80 1.05 -1.67 5.80Attitude Toward Using Technology 2 5.67 1.23 -1.60 5.67Attitude Toward Using Technology 3 5.69 0.94 -1.26 5.69Attitude Toward Using Technology 4 5.39 1.24 -1.05 5.39Attitude Toward Using Technology 5 5.78 1.14 -1.65 5.78Attitude Toward Using Technology 6 5.87 1.10 -1.85 5.87 Social Influence 1 4.75 1.25 -0.11 4.75Social Influence 2 4.76 1.27 -0.17 4.76Social Influence 3 5.61 1.07 -1.22 5.61Social Influence 4 5.20 1.19 -0.79 5.20Social Influence 5 6.20 0.83 -1.44 6.20Social Influence 6 4.91 1.82 -0.65 4.91 Facilitating Conditions 1 5.98 0.92 -1.91 5.98Facilitating Conditions 2 6.04 0.80 -1.17 6.04Facilitating Conditions 3 4.87 1.64 -0.78 4.87Facilitating Conditions 4 5.66 1.18 -1.47 5.66Facilitating Conditions 5 5.71 1.18 -1.69 5.71 Self Efficacy 1 5.45 1.31 -1.09 5.45Self Efficacy 2 5.51 1.29 -1.25 5.51Self Efficacy 3 5.44 1.16 -1.06 5.44Self Efficacy 4 5.23 1.35 -0.84 5.23Self Efficacy 5 4.97 1.40 -0.89 4.97


Table 15b. Statistical Analysis (continued). Standard Variables Mean Deviation Skewness Kurtosis Anxiety 1 3.14 1.69 0.63 3.14 Anxiety 2 3.50 1.89 0.32 3.50 Anxiety 3 2.52 1.41 1.28 2.52 Anxiety 4 2.43 1.45 1.24 2.43 Behavioral Intent 1 5.68 1.16 -1.53 5.68 Behavioral Intent 2 4.62 1.59 -0.32 4.62 Behavioral Intent 3 6.08 1.04 -2.17 6.08 Behavioral Intent 4 5.94 1.07 -1.82 5.94 Behavioral Intent 5 5.87 1.13 -1.60 5.87 Use 1 4.95 1.57 -0.94 4.95 Use 2 4.92 1.61 -0.96 4.92 Use 3 5.31 1.53 -0.89 5.31 Use 4 5.11 1.57 -1.04 5.11

Based on this information and tests of univariate normality (Anderson-Darling

test) none of the variables in this study were normally distributed. This phenomenon is

similar to other studies of technology acceptance (Compeau et al., 1999) (Chin et al.,

2003) (Heijden, 2004). The use of the partial least squares (PLS) for data analysis is

appropriate for this study because of its ability to model latent constructs under non-

normal conditions (Haenlein & Kaplan, 2004). A Histogram plot of the first performance

expectancy variable will be used to illustrate the typical variable distributions for this

study. PE1 has been selected to illustrate the distribution characteristics of the data for

this study. The type of data distribution shown in PE1 is consistent with that seen in the

overall data used for this dissertation.


PE1

Freq

uenc

y

87654321

140

120

100

80

60

40

20

0

Histogram of PE1

Figure 15. Histogram of PE1, N=263, mean=5.8, StDev=1.09 Note. Histogram prepared from experimental data with MiniTab and presented here to illustrate data distribution trends.

Figure 15 illustrates the skewness and kurtosis of the data set for PE1 in graphical

form. This plot has a skewness of -1.96 and kurtosis of 5.84. Examination of Table 15

indicates considerable variation from normal distributions for all variables used to

determine the constructs in this study. In studies where the data distribution follows a

non-normal curve an Anderson-Darling (AD) test is better indicator of distribution than a

Chi-Square evaluation (Heiat, 2005). AD values less than 1.5 generally indicate that the

data fits a normal distribution, the AD value for performance expectancy variable 1 is

18.2 which is indicates non-normal data. Examination of the subgroups should be done to

determine if any of the different participant groups are causing the non-normality of the

distribution. Evaluation of cross tab reports for performance expectancy variable 1 will


illustrate the various population subgroups. The population subgroups are based on the

moderating variables contained in the model. Those variables are; gender, class status,

college major, period of computer use, and tablet PC use. Tables 16 through 26 illustrate

the various cross tabulated reports for the variable PE1.

Table 16, Cross tab report for PE1 by Tablet PC Use. The FREQ Procedure Table of gender by PE1 gender(Gender) PE1(PE1-Accomplish Tasks More Quickly) Frequency‚ Percent ‚ Row Pct ‚ Col Pct ‚ 1‚ 2‚ 3‚ 4‚ 5‚ 6‚ 7‚ Total ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ a ‚ 2 ‚ 1 ‚ 1 ‚ 1 ‚ 33 ‚ 72 ‚ 32 ‚ 142 Male ‚ 0.77 ‚ 0.38 ‚ 0.38 ‚ 0.38 ‚ 12.64 ‚ 27.59 ‚ 12.26 ‚ 54.41 ‚ 1.41 ‚ 0.70 ‚ 0.70 ‚ 0.70 ‚ 23.24 ‚ 50.70 ‚ 22.54 ‚ ‚ 50.00 ‚ 33.33 ‚ 33.33 ‚ 20.00 ‚ 57.89 ‚ 58.06 ‚ 49.23 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ b ‚ 2 ‚ 2 ‚ 2 ‚ 4 ‚ 24 ‚ 52 ‚ 33 ‚ 119 Female ‚ 0.77 ‚ 0.77 ‚ 0.77 ‚ 1.53 ‚ 9.20 ‚ 19.92 ‚ 12.64 ‚ 45.59 ‚ 1.68 ‚ 1.68 ‚ 1.68 ‚ 3.36 ‚ 20.17 ‚ 43.70 ‚ 27.73 ‚ ‚ 50.00 ‚ 66.67 ‚ 66.67 ‚ 80.00 ‚ 42.11 ‚ 41.94 ‚ 50.77 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ Total 4 3 3 5 57 124 65 261 1.53 1.15 1.15 1.92 21.84 47.51 24.90 100.00 Frequency Missing = 2

N = 261, two participants did not indicate gender. Note. All crosstab reports were prepared using SAS 9.0 and are presented in SAS format using the SAS monospace font.

The basic statistics for the gender subgroups indicate the male mean is 5.86 (N =

142) and the female mean is 5.82 (N = 119). The male group standard deviation is 1.00,

the skewness is -2.08, and the kurtosis is 5.86. The female group standard deviation is

1.19, the skewness is -1.86, and the kurtosis is 5.82. The t-test result from this paired

comparison is 0.69 with a p-value of 0.41 which is higher than the 95% confidence level


(0.05). The T test indicates that there is no significant difference between the male and

females in this study.

Table 17, Cross tab report for PE1 by Class Status. The FREQ Procedure Table of grade by PE1 grade(Class Status) PE1(PE1-Accomplish Tasks More Quickly) Frequency‚ Percent ‚ Row Pct ‚ Col Pct ‚ 1‚ 2‚ 3‚ 4‚ 5‚ 6‚ 7‚ Total ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ a ‚ 2 ‚ 1 ‚ 0 ‚ 1 ‚ 18 ‚ 67 ‚ 38 ‚ 127 Freshman ‚ 0.77 ‚ 0.39 ‚ 0.00 ‚ 0.39 ‚ 6.95 ‚ 25.87 ‚ 14.67 ‚ 49.03 ‚ 1.57 ‚ 0.79 ‚ 0.00 ‚ 0.79 ‚ 14.17 ‚ 52.76 ‚ 29.92 ‚ ‚ 50.00 ‚ 33.33 ‚ 0.00 ‚ 25.00 ‚ 32.14 ‚ 54.47 ‚ 57.58 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ b ‚ 1 ‚ 2 ‚ 1 ‚ 1 ‚ 17 ‚ 29 ‚ 15 ‚ 66 Sophomore‚ 0.39 ‚ 0.77 ‚ 0.39 ‚ 0.39 ‚ 6.56 ‚ 11.20 ‚ 5.79 ‚ 25.48 ‚ 1.52 ‚ 3.03 ‚ 1.52 ‚ 1.52 ‚ 25.76 ‚ 43.94 ‚ 22.73 ‚ ‚ 25.00 ‚ 66.67 ‚ 33.33 ‚ 25.00 ‚ 30.36 ‚ 23.58 ‚ 22.73 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ c ‚ 1 ‚ 0 ‚ 2 ‚ 1 ‚ 12 ‚ 23 ‚ 10 ‚ 49 Junior ‚ 0.39 ‚ 0.00 ‚ 0.77 ‚ 0.39 ‚ 4.63 ‚ 8.88 ‚ 3.86 ‚ 18.92 ‚ 2.04 ‚ 0.00 ‚ 4.08 ‚ 2.04 ‚ 24.49 ‚ 46.94 ‚ 20.41 ‚ ‚ 25.00 ‚ 0.00 ‚ 66.67 ‚ 25.00 ‚ 21.43 ‚ 18.70 ‚ 15.15 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ d ‚ 0 ‚ 0 ‚ 0 ‚ 1 ‚ 8 ‚ 4 ‚ 3 ‚ 16 Senior ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 0.39 ‚ 3.09 ‚ 1.54 ‚ 1.16 ‚ 6.18 ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 6.25 ‚ 50.00 ‚ 25.00 ‚ 18.75 ‚ ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 25.00 ‚ 14.29 ‚ 3.25 ‚ 4.55 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ e ‚ 0 ‚ 0 ‚ 0 ‚ 0 ‚ 1 ‚ 0 ‚ 0 ‚ 1 Graduate ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 0.39 ‚ 0.00 ‚ 0.00 ‚ 0.39 ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 100.00 ‚ 0.00 ‚ 0.00 ‚ ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 1.79 ‚ 0.00 ‚ 0.00 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ Total 4 3 3 4 56 123 66 259 1.54 1.16 1.16 1.54 21.62 47.49 25.48 100.00 Frequency Missing = 4

N = 259, four did not indicate class placement.


The basic statistics for the study grouped by grade classification are the freshman

mean is 6.03 for a group of 127 participants; the sophomore mean is 5.70 for a group of

66 participants; the junior mean is 5.70 for a group of 49 participants; and the senior

mean is 5.56 for a group of 16 participants. The corresponding standard deviation,

skewness, and kurtosis for each group is as follows; freshman 1.00, -2.54, 10.41;

sophomore 1.22, -1.78, 4.36; junior 1.16, -1.80, 5.78; and senior 0.89, 0.43, -0.061. The t-

test results for each paired comparison indicate that there is a difference in the groups

between freshman level students and the junior and senior groups. This is important as it

points to the moderating effect of experience in this study.

Table 18 summarizes the results of the t-test analysis. The hypothesis of the

paired groups is that the distributions are the same. The hypothesis is not true if the p-

value is less than 0.05 (95% confidence). There are statistically significant differences

between freshman students and non-freshman students in this study indicating very

significant differences between lower and upper classman opinions to the moderating

effects of facilitating conditions and social influences on this study.


Table 18. T test and p values for Participant Groups and PE1 Comparison Groups T stat p value Freshman vs. sophomore 1.67 0.04

Freshman vs. junior 1.45 0.07

Freshman vs. senior 1.59 0.06

Sophomore vs. junior -0.10 0.46

Sophomore vs. senior 0.39 0.35

Junior vs. senior 0.45 0.33


Table 19. Cross tab report for PE1 by College Major. The FREQ Procedure Table of major by PE1 major(College) PE1(PE1-Accomplish Tasks More Quickly) Frequency‚ Percent ‚ Row Pct ‚ Col Pct ‚ 1‚ 2‚ 3‚ 4‚ 5‚ 6‚ 7‚ Total ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ a ‚ 1 ‚ 1 ‚ 0 ‚ 0 ‚ 14 ‚ 27 ‚ 18 ‚ 61 Arts&Sc ‚ 0.38 ‚ 0.38 ‚ 0.00 ‚ 0.00 ‚ 5.32 ‚ 10.27 ‚ 6.84 ‚ 23.19 ‚ 1.64 ‚ 1.64 ‚ 0.00 ‚ 0.00 ‚ 22.95 ‚ 44.26 ‚ 29.51 ‚ ‚ 25.00 ‚ 33.33 ‚ 0.00 ‚ 0.00 ‚ 24.14 ‚ 21.77 ‚ 27.27 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ b ‚ 3 ‚ 1 ‚ 2 ‚ 3 ‚ 22 ‚ 57 ‚ 34 ‚ 122 BIS ‚ 1.14 ‚ 0.38 ‚ 0.76 ‚ 1.14 ‚ 8.37 ‚ 21.67 ‚ 12.93 ‚ 46.39 ‚ 2.46 ‚ 0.82 ‚ 1.64 ‚ 2.46 ‚ 18.03 ‚ 46.72 ‚ 27.87 ‚ ‚ 75.00 ‚ 33.33 ‚ 66.67 ‚ 60.00 ‚ 37.93 ‚ 45.97 ‚ 51.52 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ c ‚ 0 ‚ 0 ‚ 1 ‚ 1 ‚ 20 ‚ 30 ‚ 10 ‚ 62 Education‚ 0.00 ‚ 0.00 ‚ 0.38 ‚ 0.38 ‚ 7.60 ‚ 11.41 ‚ 3.80 ‚ 23.57 ‚ 0.00 ‚ 0.00 ‚ 1.61 ‚ 1.61 ‚ 32.26 ‚ 48.39 ‚ 16.13 ‚ ‚ 0.00 ‚ 0.00 ‚ 33.33 ‚ 20.00 ‚ 34.48 ‚ 24.19 ‚ 15.15 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ e ‚ 0 ‚ 1 ‚ 0 ‚ 1 ‚ 2 ‚ 10 ‚ 4 ‚ 18 Not ‚ 0.00 ‚ 0.38 ‚ 0.00 ‚ 0.38 ‚ 0.76 ‚ 3.80 ‚ 1.52 ‚ 6.84 Indicated‚ 0.00 ‚ 5.56 ‚ 0.00 ‚ 5.56 ‚ 11.11 ‚ 55.56 ‚ 22.22 ‚ ‚ 0.00 ‚ 33.33 ‚ 0.00 ‚ 20.00 ‚ 3.45 ‚ 8.06 ‚ 6.06 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ Total 4 3 3 5 58 124 66 263 1.52 1.14 1.14 1.90 22.05 47.15 25.10 100.00

N = 263


Table 20 summarizes the basic statistics for the various college affiliations

indicated by the survey participants for your examination.

Table 20. Mean, Standard Deviation, Skewness, and Kurtosis for College and PE1 College Groups N Mean Stdev Skewness Kurtosis Arts & Sciences 61 5.92 1.01 -2.17 7.60

Business Info Systems 122 5.84 1.20 -2.06 5.77

Education 62 5.76 0.80 -0.51 1.10

No College Selected 16 5.78 1.21 -1.96 4.97

N = 263

The T test results for each paired group indicate that there is no difference in the

groups between any college designated groups.


Table 21. Cross tab report for PE1 by First Computer Use.

The FREQ Procedure Table of compuse by PE1 compuse(First Computer Use) PE1(PE1-Accomplish Tasks More Quickly) Frequency‚ Percent ‚ Row Pct ‚ Col Pct ‚ 1‚ 2‚ 3‚ 4‚ 5‚ 6‚ 7‚ Total ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ a ‚ 1 ‚ 2 ‚ 1 ‚ 2 ‚ 19 ‚ 36 ‚ 34 ‚ 95 Elementary‚ 0.38 ‚ 0.76 ‚ 0.38 ‚ 0.76 ‚ 7.22 ‚ 13.69 ‚ 12.93 ‚ 36.12 ‚ 1.05 ‚ 2.11 ‚ 1.05 ‚ 2.11 ‚ 20.00 ‚ 37.89 ‚ 35.79 ‚ ‚ 25.00 ‚ 66.67 ‚ 33.33 ‚ 40.00 ‚ 32.76 ‚ 29.03 ‚ 51.52 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ b ‚ 1 ‚ 0 ‚ 1 ‚ 1 ‚ 17 ‚ 55 ‚ 22 ‚ 97 Middle ‚ 0.38 ‚ 0.00 ‚ 0.38 ‚ 0.38 ‚ 6.46 ‚ 20.91 ‚ 8.37 ‚ 36.88 School ‚ 1.03 ‚ 0.00 ‚ 1.03 ‚ 1.03 ‚ 17.53 ‚ 56.70 ‚ 22.68 ‚ ‚ 25.00 ‚ 0.00 ‚ 33.33 ‚ 20.00 ‚ 29.31 ‚ 44.35 ‚ 33.33 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ c ‚ 0 ‚ 0 ‚ 0 ‚ 2 ‚ 11 ‚ 20 ‚ 4 ‚ 37 High ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 0.76 ‚ 4.18 ‚ 7.60 ‚ 1.52 ‚ 14.07 School ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 5.41 ‚ 29.73 ‚ 54.05 ‚ 10.81 ‚ ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 40.00 ‚ 18.97 ‚ 16.13 ‚ 6.06 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ d ‚ 2 ‚ 1 ‚ 1 ‚ 0 ‚ 10 ‚ 13 ‚ 6 ‚ 33 College ‚ 0.76 ‚ 0.38 ‚ 0.38 ‚ 0.00 ‚ 3.80 ‚ 4.94 ‚ 2.28 ‚ 12.55 ‚ 6.06 ‚ 3.03 ‚ 3.03 ‚ 0.00 ‚ 30.30 ‚ 39.39 ‚ 18.18 ‚ ‚ 50.00 ‚ 33.33 ‚ 33.33 ‚ 0.00 ‚ 17.24 ‚ 10.48 ‚ 9.09 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ e ‚ 0 ‚ 0 ‚ 0 ‚ 0 ‚ 1 ‚ 0 ‚ 0 ‚ 1 Don’t ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 0.38 ‚ 0.00 ‚ 0.00 ‚ 0.38 Use ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 100.00 ‚ 0.00 ‚ 0.00 ‚ ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 1.72 ‚ 0.00 ‚ 0.00 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ Total 4 3 3 5 58 124 66 263 1.52 1.14 1.14 1.90 22.05 47.15 25.10 100.00 N = 263


Table 22 summarizes the basic statistics for first computer use and PE1. This is an

indication of experience, a moderating variable for the model.

Table 22. Basic statistics for Computer Experience and PE1 College Groups N Mean Stdev Skewness Kurtosis Elementary 95 5.95 1.16 -1.81 4.67

Middle School 97 5.95 0.89 -2.13 9.58

High School 37 5.70 0.74 -0.32 0.12

College 33 5.35 1.54 -1.65 2.74

The T test results for each paired group summarized in Table 23 indicates that

there is a differences in the between first computer use in middle school and college with

a p value of 0.04.

Table 23. T test and p values for Participant Groups and PE1 Comparison Groups T stat p value Elementary vs. middle school -0.08 0.50

Elementary vs. high school 1.24 0.11

Elementary vs. college 1.79 0.05

Middle School vs. high school 1.47 0.07

Middle School vs. college 1.84 0.04

High School vs. college 0.89 0.19

The group difference between first computer use in elementary and middle school

versus first computer use in college is significant at the 95%confidence. This result again

reinforces that there are statistically significant differences in this population for the


moderating variable of experience. This population’s computer experience, when

comparing first computer use in elementary and middle school versus college is

significant at the 95% confidence level.

Table 24. Cross tab report for PE1 by First Tablet PC Use.

The FREQ Procedure Table of TPC_use by PE1 TPC_use(Months Experience with Tablet) PE1(PE1-Accomplish Tasks More Quickly) Frequency‚ Percent ‚ Row Pct ‚ Col Pct ‚ 1‚ 2‚ 3‚ 4‚ 5‚ 6‚ 7‚ Total ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ >twelve ‚ 1 ‚ 0 ‚ 1 ‚ 2 ‚ 22 ‚ 32 ‚ 16 ‚ 74 ‚ 0.38 ‚ 0.00 ‚ 0.38 ‚ 0.76 ‚ 8.37 ‚ 12.17 ‚ 6.08 ‚ 28.14 ‚ 1.35 ‚ 0.00 ‚ 1.35 ‚ 2.70 ‚ 29.73 ‚ 43.24 ‚ 21.62 ‚ ‚ 25.00 ‚ 0.00 ‚ 33.33 ‚ 40.00 ‚ 37.93 ‚ 25.81 ‚ 24.24 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ nine ‚ 0 ‚ 1 ‚ 0 ‚ 1 ‚ 6 ‚ 27 ‚ 12 ‚ 47 ‚ 0.00 ‚ 0.38 ‚ 0.00 ‚ 0.38 ‚ 2.28 ‚ 10.27 ‚ 4.56 ‚ 17.87 ‚ 0.00 ‚ 2.13 ‚ 0.00 ‚ 2.13 ‚ 12.77 ‚ 57.45 ‚ 25.53 ‚ ‚ 0.00 ‚ 33.33 ‚ 0.00 ‚ 20.00 ‚ 10.34 ‚ 21.77 ‚ 18.18 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ six ‚ 2 ‚ 2 ‚ 1 ‚ 1 ‚ 20 ‚ 45 ‚ 29 ‚ 100 ‚ 0.76 ‚ 0.76 ‚ 0.38 ‚ 0.38 ‚ 7.60 ‚ 17.11 ‚ 11.03 ‚ 38.02 ‚ 2.00 ‚ 2.00 ‚ 1.00 ‚ 1.00 ‚ 20.00 ‚ 45.00 ‚ 29.00 ‚ ‚ 50.00 ‚ 66.67 ‚ 33.33 ‚ 20.00 ‚ 34.48 ‚ 36.29 ‚ 43.94 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ three ‚ 0 ‚ 0 ‚ 1 ‚ 1 ‚ 8 ‚ 18 ‚ 6 ‚ 34 ‚ 0.00 ‚ 0.00 ‚ 0.38 ‚ 0.38 ‚ 3.04 ‚ 6.84 ‚ 2.28 ‚ 12.93 ‚ 0.00 ‚ 0.00 ‚ 2.94 ‚ 2.94 ‚ 23.53 ‚ 52.94 ‚ 17.65 ‚ ‚ 0.00 ‚ 0.00 ‚ 33.33 ‚ 20.00 ‚ 13.79 ‚ 14.52 ‚ 9.09 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ twelve ‚ 1 ‚ 0 ‚ 0 ‚ 0 ‚ 2 ‚ 2 ‚ 3 ‚ 8 ‚ 0.38 ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 0.76 ‚ 0.76 ‚ 1.14 ‚ 3.04 ‚ 12.50 ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 25.00 ‚ 25.00 ‚ 37.50 ‚ ‚ 25.00 ‚ 0.00 ‚ 0.00 ‚ 0.00 ‚ 3.45 ‚ 1.61 ‚ 4.55 ‚ ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ Total 4 3 3 5 58 124 66 263 1.52 1.14 1.14 1.90 22.05 47.15 25.10 100.00

N = 263.


Table 25 summarizes the basic statistics for tablet personal computer use and

PE1. This is also an indication of experience, a moderating variable for the model.

Table 25. Basic Statistics for Tablet PC Use and PE1 College Groups N Mean Stdev Skewness Kurtosis Three months 34 5.79 0.88 -0.99 1.99

Six months 100 5.86 1.20 -2.06 5.76

Nine months 47 6.00 0.91 -2.00 7.32

Twelve months 8 5.50 2.00 -1.93 4.22

More than 12 mo. 74 5.76 1.02 -1.57 5.66

The T test results for each paired group indicate that there is no statistically

significant difference between the groups for tablet PC usages as illustrated in Table 26.

Table 26. T test and p values for Tablet PC Use and PE1 Comparison Groups T stat p value Three months vs. six months 0.31 0.38

Three months vs. nine months 0.05 0.48

Three months vs. 12 months -0.11 0.46

Three months vs. > 12 months 0.04 0.49

Six month vs. nine months -0.84 0.20

Six month vs. 12 months -0.03 0.49

Six month vs. > 12 months 0.40 0.34

Nine months vs. 12 months 0.16 0.44

Nine months vs. > 12 months 1.22 0.11

12 months vs. > 12 months 0.12 0.45


These analyses of the moderating effects of the various groups on variable PE1

reinforce the variance of the participant population. Other findings of significant

variations of the groups will be discussed in chapter 5. The finalized tablet PC adoption

model used in this dissertation ended up with 36 variables used in its construction. There

are five different participant groups which result in 180 different cross tabulated reports.

The remainder of the crosstab reports are available from the researcher.

Model Validity

The dissertation model used for this research measures the interaction effects

between the constructs of the model. The most common methods used for information

systems research for moderator analysis is regression analysis. The Tablet PC acceptance

dissertation model under study is comprised of nine latent variables, or constructs which

cannot be directly measured. These latent variables are performance expectancy, effect

expectancy, social influence, attitude toward using technology, self efficacy, anxiety, self

efficacy, and attitude toward using technology which are formative variables that, in

combination, determine behavioral intent and two direct determinants of usage behavior

(behavioral intent and facilitating conditions). Venkatesh et al., the developers of

UTAUT, include four moderating factors (gender, age, experience, and voluntariness)

that have varying influence the primary constructs.

PLS-Graph was be used to determine the validity of the various measurements or

questions used in this study. The statements contained in the survey instrument were

evaluated for internal consistency (IC) and retained if the question had an IC greater than

0.7. This is consistent with current articles outlining technology adoption research. PLS


weighed the relationship between the questions used to determine unobservable model

constructs.

A complete analysis of the dissertation model requires an examination of both the

goodness of fit criteria and the factor loadings. The goodness of fit indices measures how

well the model parameter estimates were able to reproduce the sample covariance matrix.

The technique does this by evaluating the dissertation model as true and modifies the

parameter estimates until the covariance difference between the parameter estimates and

the sample is minimized.

Reliability

Reliability in technology acceptance models refers to the degree which the

variables, or indicators, are stable and consistent with what they are suppose to be

measuring (Singleton & Straits, 2004). Cronbach alpha is commonly used by researchers

in this area and is recommended by Straub (Straub, 1989) and many others including

Venkatesh & Davis who originated the UTAUT model. Table 27a & b, and 28 list the

reliability scores for all the variables. Variables with an alpha value greater that 0.70 are

within a common threshold for psychometric research and will be used to evaluate the

constructs in this model (Nunnally & Bernstein, 1994; Straub, Boudreau, & Gefen,

2004).


Table 27a. Reliability of Performance Expectancy Construct Variables

Raw to total Raw Standard Standard All PE Variable Correlation Alpha Correlation Alpha Alpha PE1 0.59 0.82 0.60 0.82

PE2 0.63 0.81 0.60 0.82

PE3 0.55 0.82 0.57 0.82

PE4 0.59 0.81 0.56 0.82

PE5 0.54 0.82 0.57 0.82

PE6 0.61 0.81 0.58 0.82

PE7 0.27 0.84 0.30 0.85

PE8 0.37 0.83 0.38 0.84

PE9 0.53 0.82 0.52 0.83 0.84 (raw)

PE10 0.67 0.81 0.68 0.81 0.84 (std)

Table 27 is continued on the next page.


Table 27b. Reliability of *EE, ATUT, SI, FC, & SE Construct Variables (Continued) Raw to total Raw Standard Standard Group Variable Correlation Alpha Correlation Alpha Alpha Raw/Std EE1 0.69 0.83 0.73 0.86 EE2 0.75 0.83 0.80 0.86 EE3 0.75 0.83 0.80 0.86 EE4 0.70 0.83 0.74 0.86 EE5 0.73 0.83 0.77 0.86 EE6 0.70 0.83 0.73 0.86 EE7 0.41 0.88 0.37 0.90 0.86 EE8 0.40 0.87 0.35 0.90 0.89 ATUT1 0.71 0.88 0.71 0.87 ATUT2 0.51 0.90 0.52 0.90 ATUT3 0.75 0.86 0.75 0.87 ATUT4 0.65 0.88 0.65 0.88 ATUT5 0.82 0.84 0.82 0.86 0.89 ATUT6 0.83 0.84 0.84 0.85 0.89 SI1 0.60 0.65 0.60 0.70 SI2 0.66 0.63 0.67 0.68 SI3 0.59 0.66 0.63 0.69 SI4 0.45 0.69 0.47 0.73 0.73 SI5 0.37 0.72 0.40 0.75 0.76 FC1 0.61 0.50 0.65 0.57 FC2 0.52 0.56 0.55 0.61 FC3 0.22 0.72 0.23 0.74 FC4 0.37 0.59 0.40 0.68 0.64 FC5 0.45 0.56 0.48 0.64 0.70 SE1 0.50 0.83 0.51 0.84 SE2 0.70 0.78 0.71 0.78 SE3 0.70 0.78 0.71 0.78 SE4 0.67 0.79 0.67 0.80 0.83 SE5 0.61 0.81 0.60 0.81 0.84 SE6 0.27 0.79 0.26 0.79 * Effort Expectancy, Attitude Toward Using Technology, Social Influence, Facilitating Conditions, and Self Efficacy.


All the variables used to determine the various constructs except facilitating

conditions met this level of reliability. The items used to measure facilitating conditions

were left in the model with alpha values (raw and standard) of 0.64 and 0.70 because of

the importance of facilitating conditions to the dissertation model. The inclusion of

facilitating conditions in this model is supported by the variables internal consistencies

which are displayed latter in this dissertation in table 31.

Table 28. Reliability of *ANX, BI, & USE Construct Variables Raw to total Raw Standard Standard Group Variable Correlation Alpha Correlation Alpha Alpha Raw/Std ANX1 0.53 0.83 0.53 0.85 ANX2 0.62 0.80 0.63 0.80 ANX3 0.75 0.74 0.76 0.75 0.82 ANX4 0.74 0.74 0.75 0.75 0.84 BI1 0.65 0.66 0.68 0.73 BI2 0.12 0.89 0.12 0.89 BI3 0.67 0.67 0.69 0.73 BI4 0.69 0.66 0.74 0.71 0.75 BI5 0.72 0.64 0.77 0.70 0.80 USE1 0.56 0.53 0.55 0.53 USE2 0.61 0.50 0.60 0.49 USE3 0.16 0.78 0.16 0.78 0.67 USE4 0.53 0.55 0.53 0.55 0.67 *Anxiety, Behavioral Intent, and Usage

Construct Validity

Construct validity refers to the degree which a variable measures what it was

intended to measure (Cronbach, 1951; Straub et al, 2004). To prove construct validity


both convergent and discriminate validity must be proven. Convergent validity is degree

which similar constructs are related; while discriminate validity is the degree that

different constructs are different from each other. The survey instrument was evaluated

by performing principal component factor analysis as described by Straub in 1989 and

2004 (Straub et al, 1989; 2004). To determine convergent validity the survey items were

reviewed by professionals in the field of IS research and the instrument was reviewed by

a prominent researcher, Dr Omar El-Gayar, who publishes in this area. In addition, the

instrument was pre-tested as recommended by Straub (Straub et al., 2004). The

statements included in the instrument were based on statements contained in seminal

works of technology acceptance (Davis, 1992) (Venkatesh et al., 2003) (Compeau et al.,

1995). PLS-Graph v. 3.0 (Chin, 1998), was used to determine convergent validity. PLS

does this by examining the loading factors and the standard error estimate (Segars, 1997).

The original fifty three variables initially included in the survey instrument were

analyzed using principle component analysis in PLS-Graph, resulting in the removal of

seventeen items. Table 25 presents the initial component analysis. In PLS-Graph the

variables, or indicators, for each construct are evaluated for their internal consistency (IC)

within the model, if any ICs are less than 0.7 they are removed from the model and the

ICs are recalculated. If any indicator in the model has an IC value that is less than 0.7

then the variable with the lowest internal consistency (IC) factor is removed and the

model is recalculated again. The next lowest variable IC, if it is lower than 0.7, is

removed and the recalculation process is repeated. This process is continued until no IC

factor for any one construct is less than 0.7 (Gefen & Straub, 2005). The iterative process


is displayed for the performance expectancy construct to provide a thorough illustration

of the process in Table 29; this detail will not be provided for the other constructs.

Table 29. Internal Consistency Factor Analysis for PE1 – PE10 Indicators*

Scale Initial Item 2nd Third Forth Fifth Sixth 7th Items Calc Xed Calc Calc Calc Calc Calc Calc PE1 0.73 0.73 0.74 0.77 0.78 0.80 0.81

PE2 0.61 6th 0.63 0.65 0.64 0.61 0.57 Gone6

PE3 0.68 0.67 0.68 0.70 0.71 0.73 0.73

PE4 0.60 4th 0.62 0.64 0.623 Gone4 Gone4 Gone4

PE5 0.68 0.69 0.68 0.70 0.73 0.73 0.76

PE6 0.61 5th 0.62 0.65 0.625 0.60 Gone5 Gone5

PE7 0.41 1st Gone1 Gone1 Gone1 Gone1 Gone1 Gone1

PE8 0.53 2nd 0.51 Gone2 Gone2 Gone2 Gone2 Gone2

PE9 0.59 3rd 0.60 0.62 Gone3 Gone3 Gone3 Gone3

PE10 0.78 0.79 0.79 0.79 0.81 0.82 0.82

*The subscript of the Gonex indicates the order in which the variables were removed.

Examination of Table 29 illustrates that the internal consistencies of some items

improve when other items are removed. Other items loading coefficients get lower

indicating the convergence and divergence of the different variables used to measure the

latent constructs. It is interesting to note that the raw correlation coefficients of the ten


performance expectancy variables calculated with SAS indicates pe1, pe3, pe5, and pe10

have the highest inter relational coefficients of variation as illustrated in table 26 below.

Table 30. Correlation Coefficients for Performance Expectancy Indicators. PE1 – PE10 Variable Correlation with all PE’s Selected by PLS-Graph

PE1 0.48 YES

PE2 0.37 NO

PE3 0.55 YES

PE4 0.33 NO

PE5 0.49 YES

PE6 0.31 NO

PE7 0.29 NO

PE8 0.47 NO

PE9 0.36 NO

PE10 0.60 YES

The final weight applied to the variables by PLS-Graph are; PE1 (0.32), PE3

(0.33), PE5 (0.28), and PE10 (0.36).

Table 31 indicates the variables retained for each of the remaining constructs with

their final internal consistencies and loading factors. These include the independent

constructs effort expectancy, attitude toward the use of technology (ATUT), social

influence (SI), facilitating conditions (FC), self efficacy (SE), anxiety (ANX), and the

dependent constructs of behavioral intent (BI), and usage (USE).


Table 31. Internal Consistency Factor Loading Analysis for the Other Indicators. Variable model IC factor Weight

EE1 0.84 0.15 EE2 0.93 0.20 EE3 0.92 0.17 EE4 0.90 0.22 EE5 0.90 0.17 EE6 0.88 0.20 ATUT1 0.83 0.35 ATUT3 0.85 0.35 ATUT4 0.75 0.21 ATUT5 0.83 0.31 SI1 0.81 0.27 SI2 0.86 0.31 SI3 0.83 0.36 SI5 0.72 0.30 FC1 0.82 0.32 FC2 0.77 0.29 FC4 0.71 0.25 FC5 0.83 0.41 SE2 0.85 0.37 SE3 0.79 0.26 SE4 0.85 0.33 SE5 0.76 0.26 ANX2 0.80 0.28 ANX3 0.92 0.47 ANX4 0.89 0.38 BI1 0.85 0.35 BI3 0.74 0.21 BI4 0.85 0.30 BI5 0.91 0.07 USE1 0.82 0.36 USE2 0.85 0.42 USE4 0.78 0.44


The high IC values for the facilitating condition variable supports the retention of these

variables in the model even though their initial reliabilities were lower than 0.7.

Partial Least Squares

PLS-Graph 3.0 was used to analyze the dissertation model to examine the

hypothesized relationships for the acceptance of tablet PCs. The model represents the

relationship between the indicator items and the constructs they are intended to measure.

Its primary purpose is to evaluate the reliability and validity of the instrument. PLS uses

confirmatory factor analysis (CFA) to assess the measurement model, which requires the

researcher to designate the structure of variable relationships. CFA is used to validate the

dimensions and validity of the measures as well as provide a means to analyze

measurement issues (Garrison, 2005).

Model Analysis

PLS-Graph generates weights and loading factors for each item in relation to the

construct it was intended to measure. The loadings in the model were used to assess

individual item reliability (IIR) (Chin, 1998). Variables with IIR loadings greater than

0.70 are considered acceptable and explain the variance in a particular measure and

ensures the item measures the correct construct (Gerfen & Straub, 2005). The weights

calculated by PLS are used to calculate latent variable scores for the constructs, which

reflect the contribution of each variable to its construct. In addition, PLS generates an

internal consistency score for each indicator and construct indicating the correlation

between each variable and the composite score for the construct (Chin, 1996). Internal


consistency values of 0.7 and higher indicate a positive relationship between the indicator

and the construct. ICs lower than those imply that there is little relationship to the parent

construct. Overall the indicators and constructs had high IC scores which indicate that all

the measures met the reliability requirements suggested by Chin (Chin, 1998) (See Table

32).


Table 32. Individual Loadings, Weights, and Internal Consistencies (IC). Indicator Construct Construct Item Weight IC IC factor Performance PE1 0.30 0.81 0.86 Expectancy PE3 0.32 0.73 PE5 0.30 0.76 PE10 0.34 0.82 Effort EE1 0.15 0.84 0.96 Expectancy EE2 0.20 0.93 EE3 0.17 0.92 EE4 0.22 0.90 EE5 0.17 0.90 EE6 0.20 0.88 Attitude Toward ATUT1 0.35 0.83 0.89 Using Technology ATUT3 0.35 0.85 ATUT4 0.21 0.75 ATUT5 0.31 0.83 Social SI1 0.27 0.81 0.88 Influence SI2 0.31 0.86 SI3 0.36 0.83 SI5 0.30 0.73 Facilitating FC1 0.32 0.82 0.86 Conditions FC2 .029 0.77 FC4 0.25 0.71 FC5 0.41 0.83 Self Efficacy SE2 0.38 0.85 0.89 SE3 0.26 0.79 SE4 0.33 0.85 SE5 0.26 0.83 Anxiety ANX2 0.28 0.80 0.91 ANX3 0.47 0.93 ANX4 0.38 0.89 Behavioral BI1 0.35 0.85 0.91 Intention BI3 0.21 0.74 BI4 0.30 0.85 BI5 0.32 0.91 Usage USE1 0.36 0.82 0.86 USE2 0.42 0.85 USE4 0.44 0.78


PLS-Graph was used to measure the discriminate validity of the model.

Discriminate validity is the degree to which any single construct is different from the

other constructs in the model. The criteria for measuring discriminate validity is to

measure the average variance extracted (AVE), which indicates the average variance

shared by a construct and its indicators (Fornell & Larcker, 1981). Discriminate validity

is adequate when constructs have an AVE loading greater than 0.5 meaning that at least

50% of measurement variance was captured by the construct (Chin, 1998). In addition,

discriminate validity is confirmed if the diagonal elements are significantly higher that

the off-diagonal values in the corresponding rows and columns. The diagonal elements

are the square root of the AVE score for each construct. Table 33 contains the AVE

scores and a correlation matrix for the constructs. All constructs have AVE scores greater

than 0.5 indicating successful validation. The instrument has achieved acceptable levels

of validity and reliability. The instrument demonstrates adequate discriminate validity

because the diagonal, in bold, values are greater than the corresponding correlation

values in the adjoining columns and rows (Chin, 1998).


Table 33. AVE Scores and Correlation of Latent Variables. AVE PE EE SI FC SE ANX BI USE ATUT PE 0.61 0.78

EE 0.80 0.30 0.89

SI 0.65 0.31 0.22 0.81

BI 0.71 0.51 0.43 0.45 0.84

USE 0.67 0.27 0.27 0.24 0.30 0.82

FC 0.67 0.43 0.49 0.43 0.60 0.30 0.82

SE 0.66 0.09 0.25 0.14 0.27 0.09 0.16 0.81

ANX 0.76 (0.06) (0.30) (0.11) (0.20) (0.02) (0.25) (0.01) 0.87

ATUT 0.67 0.57 0.42 0.59 0.69 0.27 0.56 0.18 (0.14) 0.82

Structural Model Analysis

The dissertation model was evaluated with PLS-Graph to determine the

correctness of the model proposed in Chapter 3 (See Figure 5). Because PLS does not

require a normally distributed data it is evaluated with R-squared calculation for

dependent latent variables (Cohen, 1988) and the average variance extracted (Fornell &

Larchner, 1981). The first item that PLS-Graph provides to determine how well the

model fits the hypothesized relationship is the squared multiple correlation (R2) for each

dependent construct in the model. The R2 measures a construct’s percent variation that is

explained by the model (Wixom & Watson, 2001). The R2 values for each dependent

variable are Behavioral Intent (0.55) and Use Behavior (0.11). The interpretation of these

factors indicate that the model explains 55% of the variance of the dependent variable BI


toward adoption of the tablet PC and that the dependent variable Use Behavior explains

11% of the variance of the device usage. R2 values were also calculated for the freshman

and upper classman models.

Bootstrap method was used in PLS-Graph to determine the strength of the

relationships between two dependent constructs in the model (Wixom & Watson, 2001).

The dependent constructs in the dissertation model are behavioral intent and use

behavior. The structural path diagram shown in figure 16 provides evidence that supports

many of the six hypotheses.

Figure 16. Tablet PC Structural Model The positive coefficient values for all constructs, except anxiety, indicate that the

participants in this particular study had a positive inclination toward tablet PC use which

supports hypothesis one. The same information supports hypothesis two. Hypothesis four

states “Computer self efficacy does have an impact students acceptance of the Tablet PC”


and it is supported by the positive path coefficient of 0.124 of the self efficacy construct.

The anxiety constructs negative coefficient of (-0.073) does not support hypothesis five.

The final hypothesis that states “Students’ use of the Tablet PC does impact student’s

acceptance of the device” is partially supported by the low R2 of 0.11 and the path

coefficient of 0.10.

Freshman vs. Upper Classmen

Examination of the survey population earlier in this dissertation uncovered a

statistically significant difference between freshman and upper class survey participants.

Differential Analysis of the two population groups is necessary to understand what this

means in the context of technology acceptance in this environment. Data extracted from

the survey representing just freshman students (N=127) with PLS-Graph is depicted

graphically in Figure 17. A graphical representation using PLS-Graph of the upper

classmen, just juniors and seniors (N=65), is displayed in Figure 18. The comparison of

the results for the two analyses is summarized in Table 34 and 35.


Figure 17. Freshman PLS-Graph (N=127).

Examination of the PLS-Graph for the freshman analysis is the result of the

contributions of the various independent construct to the dependent constructs of

behavioral intent and use behavior. The strength of the contributions of performance

expectancy, effort expectancy, social influence, self efficacy, and anxiety are very similar

to the whole group but of a higher magnitude in attitude toward technology use and

anxiety. Also the explanation of the variance of behavioral intent and use behavior are

markedly higher at 64% and 22% respectively. That is in contrast to 55% and 11% in the

model using all students surveyed.


Figure 18. Upper Classman PLS-Graph Model (N=65).

The junior and senior survey participants were more comparable with the total

survey results with a predictive efficiency of 49% and 7% for the dependent construct of

behavioral intent and use behavior. In this group both self efficacy and anxiety were

found to have negative correlation with behavioral intent. The researcher believes that

this is a direct result of actions taken to provide training in TPC use after the TPC

initiative. After the university required students to lease a TPC the introduction to

computers course, which every new student at this university is required to enroll in, was

re-engineered to comprehensively include TPC use. New freshman students and transfers

from other universities would be required to enroll in this class. The new class was

completely restructured to indoctrinate students into the use and function of a mobile pen

based computing device. The upper level students, juniors and seniors, would have


already completed this course before the major changes were made. This resulted in the

upper level students’ low self efficacy score, low facilitating conditions ratings, and

negative attitude toward using technology scores. The upper classmen were not given

sufficient support or education in the use of the TPC. The upper classmen have

significantly higher scores for social influence because they voluntarily joined the

initiative in compliance with the initiatives key stakeholders.

Table 34. Comparison of Freshman and Upper Classman Model Contributions Freshman Upper Classman Construct Contribution Contribution Performance Expectancy 0.144 0.197

Effort Expectancy 0.159 0.218

Social Influence 0.039 0.322

Attitude Towards Using Technology 0.513 0.194

Self Efficacy 0.101 -0.021

Anxiety -0.123 -0.122

Behavioral Intent 0.021 0.106

Facilitation Conditions 0.457 0.206

Freshman N = 127 & Upper classmen N = 65


Table 35. Comparison of All, Freshman, and Upper Class Models.

Behavioral Use Model Intent Behavior Total All Participant Model 55% 11% 66%

Freshman Model 63% 22% 85%

Upper Class Model 49% 7% 56%

All Participant N = 263, Freshman N = 127, Upper Classmen N = 65

The difference between the freshman and upper classman scores was completely

unexpected. This valuable finding is a bonus derived from the research.


CHAPTER 5: SUMMARY & CONCLUSIONS

The research conducted for this dissertation represents a significant contribution

toward understanding the acceptance of tablet PCs (TPC) in a higher education academic

environment. The study measured college students acceptance of tablet PC’s at a small

mid western university. TPC acceptance was measured using a modification of a

technology acceptance model recently published in 2003. This model is derived from

Viswanath Venkatesh’s dissertation which was completed in 1998 under the direction of

his advisor, Fred Davis, the originator of the technology acceptance model. Participants

in this study were undergraduate students enrolled during the summer 2006 semester.

This study confirms the ability of the unified theory of acceptance and use of

technology (UTAUT) model to determine user’s acceptance of a technology tool. The

UTAUT model was modified in this study by including the constructs of attitude toward

using technology, self efficacy, and anxiety because of their significance in other

technology acceptance models.

The survey tool used to measure technology acceptance contained 53 statements

pertaining to the various constructs used in popular technology acceptance models. The

participants indicated the strength of their agreement with each statement by responding

to the statements with a seven item likert scale ranging from strongly disagree (1) to

strongly agree (7). The data was then analyzed to determine statistical validity with PLS-

Graph to determine the participants’ perceptions and behavioral intent toward use of

tablet PCs. The data produced by the statistical analysis provides a basis for responding

to the individual research hypotheses.


University students, in this environment, accept the tablet PC.

The results of this study support this hypothesis by indicating behavioral intent

(55%) among students to use the TPC. While the overall group exhibited a low predictive

value for use behavior of 11%, there was a significant difference between the freshman

students and upper class students on this construct. By extracting the different

populations from the combined data pool and examination of the results derived from

PLS-Graph, the data showed that the freshman group, who received focused training on

the use of the tablet PC, had a much higher behavioral intent (63%). Freshman students

also had a much higher use behavior (22%). These results suggest that the appropriate

training and expectations for TCP use enhances the adoption and use of technology. The

model containing only responses from the upper classmen (N = 65) had behavioral intent

values of 49% and use behavior of only 7%. The key difference between the two groups

is the amount of educational opportunity available to each group to learn about the

correct use of the pen based computing device (TPC). The freshman students received

focused, continued instruction on how to use, maintain, and apply the device in CSC105

(Introduction to Computers). All freshman students must enroll in that course. While the

upper classmen did not receive any focused training on the device or the available

facilitating condition for device use. The university did provide opportunities fro the

upper classmen to get specific tablet PC training through elective courses but none of the

offered courses received the necessary enrollment of ten students. Academic

organizations adopting tablet PC will have a much higher probability of success if they


require specific training for the proper use of the device. This training must be mandatory

and focus on the advantages of pen based computing.

The Unified Theory of Acceptance and Use of Technology (UTAUT) does predict the

successful acceptance of the Tablet PC.

This hypothesis is supported by positive correlations between most of the model

constructs toward behavioral intent and use behavior. Anxiety does show as a negative

factor in this analysis (-7%), which would indicate a negative correlation with TPC

adoption, but anxiety was not a component of the original Venkatesh UTAUT model.

The model, as it is constructed in this study, supports 55% of the intention to use the

tablet PC and 11% of the use of the tablet PC. The freshman students, who had the most

complete support of tablet PC use, had much higher percentages for behavioral intent

(63%) and use behavior (22%). The researcher believes this is a result of the tablet PC

infrastructure innovations such as the CSC 105 course enhancements and guidance to

university support services.

The constructs of the UTAUT will demonstrate an effect on user acceptance of the Tablet

PC.

This hypothesis is supported by examination of the PLS-Graph results which

indicate that the original UTAUT constructs all support the dependent constructs of

behavioral intent and use behavior with the following predictive percentages.

Performance expectancy, effort expectancy, and social influence, all original UTAUT


constructs, contributed 19, 12, and 6 %respectively toward behavioral intent. While the

dissertation model added constructs of self efficacy, attitude toward using technology and

anxiety contributed 12, 47, and a negative 7% towards behavioral intent. It is interesting

to see that attitude toward using technology has a higher contribution to behavioral intent

than all of the constructs from the original model at 47% versus 46% for performance

expectancy, effort expectancy, and social influence. The freshman students, who received

the most complete training and support in this initiative, had UTAUT construct values of

14%, 16%, and 4% for performance expectancy, effort expectancy, and social influence.

Their very high values for facilitating conditions of 46% the researcher believes are a

direct result of the training and support they received versus the lower value of 21%

exhibited by the upper classmen who did not receive that type of support and training.

Computer self efficacy does have an impact students acceptance of the Tablet PC.

This hypothesis is supported by the 12%contribution to behavioral intention by

the self efficacy construct. This self efficacy value is as influential on the model as the

original model construct of effort expectancy at 12% significantly larger than the social

influence value at 6% in this study. The influence of self efficacy on behavioral intention

suggests a reconsideration of the decision to remove this construct from the unified

model (Venkatesh et al. 2003). Comparison between freshman students and upper

classman students for self efficacy give reason to question the components of self

efficacy. The result for freshman students of self efficacy was very close to the whole

group with a 10% contribution compared to 12% for the population as a whole. The


upper classmen had a negative contribution of self efficacy to behavioral intent of a

minus 2%. Self efficacy will be further studied in a pilot study of K-12 students and

teachers in the next academic year.

Anxiety about computer use does have an impact on student’s acceptance of the Tablet

PC.

The negative contribution of the anxiety construct to the dependent variable

behavioral intent in this study indicates that anxiety has an negative effect on the

behavior intent to use the tablet PC in this environment. This finding is slightly in

disagreement with the Venkatesh model’s contention that anxiety is a non-factor such

technological contexts (Venkatesh et al. 2003, p. 461). Both freshman and upper

classmen populations have a negative correlation of anxiety to behavior intent. The

results of the two groups are very consistent at a negative 12%for both the freshman and

upper classmen groups.

Students’ use of the Tablet PC does impact student’s acceptance of the device.

This hypothesis not supported by analysis with PLS-Graph. The dissertation

model measured a 10% contribution of behavioral intent to use behavior. When PLS-

Graph is used to reverse the link between the two dependent constructs and the factor

analysis is regenerated use behavior contributes a very low percentage to behavioral

intent. The statements used in the survey to determine this construct all are linked to use

of the stylus pen and digital screen for course work and general PC operating system

navigation. The lack of a contribution from use on behavior is puzzling. This is in stark


contrast to findings from this path reversal for the freshman and upper classmen with a

7% and 16% contribution from use behavior to behavioral intent respectively. This leads

the researcher to question the device use behavior construct in the context of this study.

This discovery is will result in the addition of some more questions into the K-12 survey

tool pertaining to use behavior.

Discussion

The researcher is interested in the acceptance of table PC adoption in educational

institutions. Presently there are a small number of universities who have implemented

Tablet personal computing in their campus classrooms. A relatively large number,

perhaps seventy, of institutes of higher education are planning policies to mandate

student purchase of a laptop or tablet PC. The primary reason this researcher instigated

this study was to learn more about the subtleties of tablet PC adoption in the higher

education domain. This study has helped educate the researcher as to the key components

of technology acceptance. The two dependent constructs used in this model are

behavioral intent and use behavior. In the environment of this study the variables of

performance expectancy, effort expectancy, attitude toward using technology, self

efficacy are key components of behavioral intent. Social influence and anxiety do not

appear to have much contribution to behavioral intent. However, these factors are shown

to have a different impact on different social groups and there fore should be included in

an acceptance study.


Social influence, sometimes referred to as social norms, was only significant to

the upper classmen in this study but it was fairly significant for that group with a 32%

explanation of the variance of behavioral intent. What is the contribution our beliefs of

what people expect from us in technology acceptance? Do our peers and superiors affect

our perceptions of technical innovations? These are questions that need time to formulate

a plan of study.

The success of any technical innovation requires planning. Adequate planning

should include three components that will improve the success of the initiative. First there

should be an opportunity for all key stakeholders to meet and discuss the initiative.

Second is adequate provision of educational opportunities to provide the skills necessary

to correctly use the device. And third, adequate support of the device including a help

desk and repair center. This three pronged plan encompasses the constructs contained in

the model. The meetings and training will benefit the constructs of effort and

performance expectancy, and the participants’ attitude towards using technology. All

which are contributors to behavioral intent. The availability of support and training will

increase the use behavior and increase the benefit of using the technology. The end result

is significantly higher probability of success which is the desired end result of any project

implementation.

A second reason for conducting this research is to study differences in technology

acceptance between students who are mandated to use the devices and those who use

them at their own discretion. The university environment used in this study is a

mandatory use environment. But since only first and second year students were required


to lease, or otherwise acquire, tablet PCs the upper class students used the device at their

own discretion. Many upper classmen did voluntarily join the initiative and appears that

in this study context those participants had a positive disposition toward tablet PC use as

indicated by their positive response level to attitude towards technology and moderate

response toward the influence of performance expectancy and facilitating conditions The

very high score for the upper classmen for social influence indicates that they based their

decision to obtain a tablet PC on the influence, or their perceptions of the influence of

key stakeholders of the initiative.

Moderating Conditions

This model includes moderating variables that impact the independent variables

that, in turn, influence the dependent variables of behavioral intention and use behavior.

Those are gender, age, experience, and voluntary use. In the context of this study gender

was found to not have any effect. Other studies have found effort expectancy to be more

important in women (Morris et al., 2000), but examination of a gender bias for the effort

expectancy variables did not find a significant difference between males and females.

The range of ages for traditional college students did not provide enough population

segments to study differences caused by the age of the participants, but in a study being

planned to survey K12 teachers this moderating variable be investigated in the near

future. Experience with computers was shown to have a significant impact on the

acceptance of technology indicated by the significance found between freshman and

upper classmen in the study. In fact, interpretation of the model with the data extracted


for the freshman students has a much higher determination of acceptance than the

population as a whole. This is due to good planning and support of this population

segment.

Voluntariness is measured with the variables used to build the behavioral intent

construct. Variables 1, 3, 4, and 5 had a very direct influence on behavioral intent. The

voluntariness indicators are very positive for the freshman students with internal

consistencies of 0.86, 0.84, 0.92, and 0.89 respectively. The upper classmen had internal

consistencies (IC) of 0.66, 0.74 and 0.90 for BI 1, 4, and 5 respectively. The upper

classmen had a very low IC value for BI3 of 0.43 resulting in it removal per PLS-Graph

normal operating procedures. It does not appear that voluntariness is statistically different

for the two populations.

One significant moderating variable was found to be computer experience,

particularly if the computer use began in elementary and middle school for the

participants. This moderating factor should be further evaluated to determine if it is a

factor in other populations. This moderating factor will be measured in the K-12 study

planned for this next year to measure the impact of first computer use on both the

teachers and the students in the 30 pilot organizations.

Limitations

The instrument records self reported results. To get an accurate picture of

participants’ ideas of TPC use several questions addressed each construct of interest. Yet,

no matter how much vigor the investigator applies to the instrument design and analysis

the final result is only a proxy measure of participants self perceptions, and a threat to the


internal validity of the study exists (Campbell, 1969), (Campbell & Stanley, 1963). Self

reported use is a relative indicator of usage intention. Precise determination would have

required data mining of computer use logs which brings up the issue of privacy and

ethics.

The findings of this study should only be applied to this unique environment. The

conclusions must be carefully evaluated before any attempt is made to project these

findings on another university setting. Student populations in other university

environments may very well have very different model analysis and distributions then

this studies participant pool. In the Venkatesh article the predictive efficiency of the

model was proclaimed to explain as much as 70% of the variance in intention but this is

still a subjective measure of intent to use a technology that can only be used as an

indicator of personal choice.

In mandatory use situations a more direct determination of use may be

informative. The same institution, in which this study was conducted, maintains a

detailed log of everything the particular TPC user is doing with the device. The

researcher has considered the implications of examining this log to determine computer

use on this campus. The line of research will be re-evaluated and submitted for a future

research study during the fall 2006 and spring 2007 academic year.

Future research

As mentioned earlier in this discussion the, state of South Dakota has started a

pilot project for laptop/table computing involving 20 high schools around the state titled

“Classroom Connections” the program will put a notebook, or tablet PC, in the hands of


each high school student. The state incentive to districts is to match one dollar for every

three dollars spent on the hardware required to transition to the wireless computing

environment. This initiative is an optimal setting to apply the dissertation adoption

model. The opportunity to query the participants prior to their participation in ubiquitous,

wireless computing provides an opportunity to determine how attitudes, efficacy and use

evolve over time. The same participants, through an anonymous identifier, will fill out

the same survey one year after implementation of the initiative at their school district. In

addition the initial survey will be used to identify critical areas for success of the

initiative. Once cross tabbed reports are prepared and delivered to the South Dakota

Department of Education the necessary steps can be implemented to remedy any

weaknesses indicated by the report. This research is already started with a meeting

between the respective parties on June 22, 2006.

Other research which should be conducted is to use a similar tool to that used in

this study to survey faculties at the universities considering, or in various stages of

implementing, adoption of tablet PC technology. The research team could help support

the academic centers adopting the technology. Just last year Brown University was the

recipient of a $1.5 million grant from the Microsoft Corporation. The new “Microsoft

Center for Research on Pen-Centric Computing” was announced on Thursday, March 20,

2006 (Lawton, 2006). The three-year alliance will create innovative software for pen-

based computers like the TPC. This type of research could easily lead to commercial

applications or grant opportunities. This calendar year Microsoft is not funding any tablet

PC research grants but with the recent announcement of Warren Buffet’s contribution to


the Bill & Melinda Gates Foundation should increase the possible funding for research in

pen based computing devices. The focus of the Gates foundation is on technology and

education in the United States and focuses on public library proposals.


REFERENCES

Abbott, K. (2004). 2003-2004 Notre Dame Tablet PC Initiative. Retrieved January 26, 2006, from http://learning.nd.edu/tabletpc/

Agarwal, R., & Prasad, J. (1998). A Conceptual and Operational Definition of Personal

Innovativeness in the Domain of Information Technology. Information Systems Research, 9(2), 204-215.

Anderson, J. E., & Schwager, P. H. (2006). The Tablet PC: Applying the UTAUT Model, Paper presented at the American Conference on Information Systems, Acapulco, Mexico.

Apple Newton. (2006). Retrieved February 9, 2006, from http://en/wikipedia.org/wiki/

Apple Newton Ajzen, I. (1991). The Theory of Planned Behavior. Organizational Behavior and Human

Decision Processes, 50, 179-211.

Ajzen, I., & Fishbein, M. (1980). Understanding Attitudes and Predicting Social Behavior. Englewood Cliffs, NJ: Prentice Hall.

Bandura A. (1977) Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall.

Bandura A. (1986) Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall.

Bandura, A. (2002). Social Cognitive Theory in Cultural Context: An International Review. Journal of Applied Psychology, 51, 269-290.

Barkhuus, L. (2005). Bring Your Own Laptop Unless You Want to Follow the Lecture:

Alternative Communications in the Classroom. Paper presented at the Proceeding of the 2005 International ACM SIGGROUP Conference on Supporting Group Work, Sanibel Island, Fl.

Brown, R. (2000). Notebook Universities. Retrieved July 26, 2005, from

http://itc.vcsu.edu/notebookinitiative/notebook_univ_listing.htm Burton, J. (2004). Higher Ed Goes Mobile, Lightweight. University Business, 7(12), 52-

55.

Campbell, D.T. (1969). Reforms as Experiments. American Psychologist, 24, 409-429.


Campbell, D. T., & Stanley, J. C. (1963). Experimental and Quasi-Experimental Designs for Research. Chicago: Rand McNally.

Carlsson, C., Carlsson, J., Hyvönen, K., Puhakainen, J., & Walden, P. (2006). Adoption of Mobile Devices/Services — Searching for Answers with the UTAUT. Paper presented at the 39th Annual Hawaii International Conference on System Sciences (HICSS'06).

Chin, W. W., Marcolin, B. L., & Newsted, P. R. (2003). A Partial Least Squares Latent Variable Modeling Approach for Measuring Interaction Effects: Results from a Monte Carlo Simulation Study and an Electronic-Mail Emotion/Adoption Study. Information Systems Research, 14(2), 189 - 217.

Chin, W. (1996). The Partial Least Squares Approach to Structural Equation Modeling.

In Advanced Structural Equation Modeling: Issues and Techniques (pp. 296-335): Lawrence Erlbaum Associates.

Ciampa, M. (2001). Designing and Implementing Wireless LANs. Boston, MA: Course

Technology.

Code of Federal Regulations Title 34 Department of Education PART 97 - Protection of Human Subjects. (2004). Retrieved April 14, 2006, from http://www.ed.gov/policy/fund/reg/humansub/part97-2.html#97.103

Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates.

Compeau, D. R., & Higgins, C. A. (1995). Computer Self-Efficacy: Development of a Measure and Initial Test. MIS Quarterly, 19, 189-211.

Cronbach, L. (1951). Coefficient Alpha and the internal structure of tests. Psychometrika, 16, 297-334.

Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User Acceptance of Computer

Technology: a Comparison of Two Theoretical Models. Management Science, 35(8), 982-1003.

Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319-340.

Decusatis, C. (2002). Fiber Optic Data Communication: Technology Advances and

Futures. Cambridge, MA: Elsevier. Dillon, P. (1998). The Next Small Thing. Fast Company (15), 97.


Dillman, D. (1999). Mail and Internet Surveys: The Tailored Design Method (2nd ed.).

New York: John Wiley. El-Gayar, O., & Moran, M. (2006, November 18-21). Students Acceptance of Tablet

PCs; an Modification of the UTAUT Model. Paper to be presented at the Decision Sciences Institute Annual Meeting, San Antonio, TX.

Fishbein, M., & Ajzen, I. (1975). Belief, Attitude, Intention and Behavior: An

Introduction to Theory and Research. Reading, MA: Addison-Wesley. Fischer, A., & Lee, K. (1999). User Interface. Retrieved February 8, 2006, from

http://cne.gmu.edu/itcore/userinterface/GUIHistory3.html Fornell, C., & Larcker, D. (1981). Evaluating Structural Equation Models with

Unobservable Variables and Measurement Error. Journal of Marketing Research, 18, 39-50.

Fornell, C., and Bookstein, F. L. “Two Structural Equation Models: LISREL and PLS

Applied to Consumer Exit-Voice Theory.” Journal of Marketing Research, Volume 19, 1982, pp. 440-452.

Francik, E., Rudman, S. E., Cooper, D., & Levine, S. (1991). Putting Innovation to Work:

Adoption Strategies for Multimedia Communication Systems. Communications of the ACM archive, 34(12), 52-63.

Fujitsu claims top tablet spot in EMEA. (2006). Retrieved March 18, 2006, from

http://www.cbronline.com/article_news.asp?guid=971B7B9D-ED45-4729-8C17-70CB97653BFC

Gall, M. D., Gall, J. P., & Borg, W. R. (2003). Educational research: An introduction

(7th ed.). Boston: Pearson Education, Inc.

Garfield, M. J. (2005). Acceptance of Ubiquitous Computing. Information Systems Management, 22(4), 24-31.

Garrison, G. (2005). A Model of Factors Impacting an Organizations Propensity to be an Early Adopter of Disruptive Technology. Doctoral Dissertation, University of Mississippi. (UMI No. 3190561).

Gates, B. (2000). Tablet PC. Retrieved February 12, 2006, from

http://www.microsoft.com/presspass/features/2000/nov00/11-13comdex.mspx


Green, D. (2004). The 2004 Campus Computing Survey. Retrieved August 8, 2005, from http://www.campuscomputing.net/introduction

Gefen, D., & Straub, D. (2005). A Practical Guide to Factorial Validity Using PLS-

Graph: Tutorial and Annotated Example. Communications of the Association for Information Systems, 16, 91-109.

Groves, M. M., & Zemel, P. C. (2000). Instructional Technology Adoption in Higher

Education: An Action Research Case Study. International Journal of Instructional Media, 27(1), 57-65.

Haenlein, M., & Kaplan, A. (2004). A Beginners Guide to Partial Least Squares Analysis.

Understanding Statistics, 3(4), 283-297. Hair, J. F., Tatham, R. L., Anderson, R. E., & Black, W. (1998). Multivariate Data

Analysis (5th. ed.). Upper Saddle River, NJ: Prentice Hall. Henderson, T., Kotz, D., & Abyzov, I. (2004). The Changing Usage of a Mature

Campus-wide Wireless Network. Paper presented at the ACM MobiCom 2004, Philadelphia, PA.

Hoffman, B. (2006). Web Survey Design. Retrieved May 23, 2006, from

http://coe.sdsu.edu/eet/articles/websurvey/ Hu, P. J., Chau, P. Y., Sheng, O. R. L., & Tam, K. Y. (1999). Examining the Technology

Acceptance Model using Physician Acceptance of Telemedicine Technology. Journal of Management Information Systems, 16(2), 91-112.

Hwang, Y., & Yi, M. Y. (2002). Predicting the use of Web-Based Information Systems;

Intrinsic Motivation and Self Efficacy Paper presented at the AMCIS 2002. Igbaria, M. (1994) An Examination of the Factors Contributing to Technology

Acceptance, Accounting, Management and Information Technologies, 4(4), 205-224

Jarvis, C. B., MacKenzie, S. B., & Podsakoff, P. M. (2003). A Critical Review of

Construct Indicators and Measurement Model Misspecification in Marketing and Consumer Research. Journal of Consumer Research, 30, 199-218.

Johnson, N., Kotz, S., & Balakrishnan, N. (1994). Continuous Univariate Distributions

(Vol. 2). New York, NY: Wiley & Sons Jones, B. (2002). Introducing the Tablet PC, Retrieved February 8, 2006, from

www.developer.com/ws/other/article.php/1500121


Knowlton, D. (2004). DSU Announces the Wireless Mobile Computing Initiative for

2004. Retrieved August 28, 2005, from http://www.dsu.edu/WMCI/presidentmessage.htm

Lawton, W. (2006). Retrieved June 29, 2006, from

http://www.brown.edu/Administration/News_Bureau/2005-06/05-095.html Legris, P., Ingham, J., & Collerette, P. (2003). Why do People Use Information

Technology? A Critical Review of the Technology Acceptance Model. Information and Management, 40, 191-204.

Lowe, P. (2004). Bentley College Students Evaluate Tablet PCs. Retrieved January 14, 2006, from www.hp.com/hpinfor/newsroom

Lohmöller, J.-B. Latent Variable Path Modeling with Partial Least Squares. Heidelberg:

Physica-Verlag, 1989. Lynott, P., & McCandless, N. J. (2000). The Impact of Age vs. Life Experiences on the

Gender Role Attitudes of Women in Different Cohorts. Journal of Women and Aging, 12(2), 5-21.

Maconachy, V., Schou, C. D., Ragsdale, D., & Welch, D. (2001). A Model for Information Assurance: An Integrated Approach. Paper presented at the 2001 IEEE Workshop on Information Assurance and Security, United States Military Academy, West Point, NY.

Mathieson, K. (1991). Predicting User Intentions: Comparing the Technology Acceptance Model with the Theory of Planned Behavior. Information Systems Research, 2, 173-191.

Mathieson, K., Peacock, E., & Chin, W. W. (2001). Extending the Technology Acceptance Model: The Influence of Perceived User Resources. The Data Base for Advances in Information Systems, 32(2), 86-112.

Meister, D. B., & Compeau, D. R. (2002). Infusion of Innovation Adoption: An Individual Perspective. Paper presented at the Proceedings of the ASAC.

Microsoft. (2002). New Ways to Interact with your PC. Retrieved December 19, 2005, from htttp:www.microsoft.com/windowsxp/tabletpc/evaluation/overviews/

Mock, K. (2004). Teaching with Tablet PC's. Journal of Computing Sciences in Colleges, 20(2), 17-27.

Nunnally, J., & Bernstein, I. (1994). Psychometric Theory (3rd. ed.): McGraw Hill.


Pijpers, G. (2001). Executives' Use of Information Technology: An Examination of

Factors Influencing Managerial Beliefs, Attitude and Use of Information Technology. Information and Software Technology, 43(15), 959-971.

Pederson, P., & Ling, R. (2002). Modifying adoption research for mobile Internet service adoption: Cross Disciplinary Interactions. Paper presented at the Proceedings of the 36th Hawaii International Conference on System Sciences (HICSS’03).

Pederson, P., & Nysveen, H. (2003). Usefulness and Self-expressiveness: Extending TAM to Explain the Adoption of a Mobile Parking Services. Paper presented at the 16th Beld eCommerce Conference, Bled, Slovenia.

Pederson, P., Nysveen, H., & Thorbjørnsen, H. (2003). Adoption of Mobile Services. Model Development and Cross-Service Study. Retrieved June 23, 2006, from http://ikt.hia.no/perep/cross_service_jams.pdf

Plouffe, C. R., Hulland, J. S., & Vandenbosch, M. (2001). Research Report: Richness Versus Parsimony in Modeling Technology Adoption Decisions -- Understanding Merchant Adoption of a Smart Card-Based Payment Systems. Information Systems Research, 12(2), 208-222.

Ray, J. (2002). SAGE Cold-War Forerunner to the Information Age. Retrieved February 11, 2006, from http://www.eskimo.com/~wow-ray/sage28.html

Rawstorne, P., Jayasuriya, R., & Caputi, P. (2000). An Integrative Model of Information

Systems use in Mandatory Environments. Paper presented at the Proceedings of the twenty first international conference on Information systems.

Schmidt, J. (2000). DSU Ranks Again, Third Year In A Row. Retrieved March 19, 2006, from http://www.departments.dsu.edu/news/spotlight/00/dsu_ranks_again.htm

Second phase of DSU wireless network begins. (2002). Retrieved February 12, 2006, from http://www.departments.dsu.edu/news/spotlight/02/wireless_phase2.htm

Segars, A. (1997). Assessing the Unidimensionality of Measurement: A Paradigm and

Illustration within the Context of Information Systems Research. Omega, 25(1), 107-121.

Shandish, W., Cook, T., & Campbell, D. (2002). Experimental and Quasi-Experimental

Design for Generalized Causal Inference. Boston, MA: Houghton Mifflin.


Singleton, R. A., & Straits, B. C. (2004). Approaches to Social Research (Fourth ed.). Oxford, England: Oxford University Press.

Straub, D. (1989). Validating Instruments in MIS Research. MIS Quarterly, 13(2), 146-

169. Straub, D., Boudreau, M.-C., & Gefen, D. (2004). Validation Guidelines for IS Positivist

Research. Communications of the Association for Information Systems, 13, 380-247.

Singletary, L., Akbulut, A., & Houston, A. (2002, August 9-11). Innovative Software Use

After Mandatory Adoption. Paper presented at the Proceedings of the Americas Conference on Information Systems (AMCIS), Dallas Texas.

Steel, N. (2003). The Interactive Lecture. Paper presented at the CSCL. Surry, D., & Land, S. M. (2000). Strategies for motivating higher education faculty to use

technology. Innovations in Education and Training International, 1(37), 1-9. Tablet PC 2004 Quick Comparison. (2004). Retrieved December 22, 2005, from

http://www.tabletpctalk.com/faqs/comparison/2004.php Taylor, S., & Todd, P. (1995). Assessing It Usage: The Role of Prior Experience. MIS

Quarterly, 19(2), 561-570.

Taylor, S., & Todd, P. (2001). Understanding Information Technology Usage: A Test of Competing Models. Information Research, 6(2), 144-176.

Vallerand, R. J. (1997). Toward a Hierarchical Model of Intrinsic and Extrinsic Motivation. New York: Academic Press.

Venkatesh, V., & Davis, F. D. (2000). A Theoretical Extension of the Technology

Acceptance Model: Four Longitudinal Field Studies. Management Science, 46(2), 186-204.

Venkatesh, V., & Morris, M. (2000). Why Don't Men Ever Stop to Ask For Directions? Gender Social Influence, and Their Role in Technology Acceptance and Usage Behavior. MIS Quarterly, 24(2), 115-139.

Venkatesh, V., Morris, M., Davis, G., & Davis, F. D. (2003). User Acceptance of Information Technology: Toward a Unified View. MIS Quarterly, 27(3), 425-478.

Wachsmuth, B. (2003). Seton Hall University Tablet PC Project. Retrieved January 24,

2006, from http:www.sc.shu.edu/tabletpc/


Westland, J. C., & Clark, T. H. K. (2000). Global Electronic Commerce: Theory and

Case Studies. Cambridge, MA: MIT Press. Willis, C., & Miertschin, S. (2004). Tablet PC's as Instructional Tools or the Pen is

Mightier than the Board. Paper presented at the 5th. Conference on Information Technology Education, Salt Lake UT.

West, J. V. (2005). Tablet PC vs. Laptop: How Do You Choose? Retrieved February 12,

2006, from http://www.microsoft.com/windowsxp/using/tabletpc/expert/vanwest_05feb11tabvlap.mspx

Zhang, P., Li, N., & Sun, H. (2006, January). Affective Quality and Cognitive Absorption:

Extending Technology Acceptance Research. Paper presented at the Hawaii International Conference on System Sciences, Kauai, Hawaii.

Zolonowski, D. (2006). Personal Communication. Chief Information Officer. Madison.

APPENDIX A: SELECTION OF COURSES TO SURVEY

Course sections were selected based on the likelihood that students would have

between three months and one year experience using the tablet personal computer (TPC)

device. Every student, including new freshman student and transfer student of any status,

must register for CSC 105 (Intro to computers). The curriculum of this course is infused

with practical TPC use advice and knowledge. When students have completed the course

they are ready to efficiently use their leased TPC devices to continue the college

education. All students regardless of their major are required to enroll in this course. The

semester following this course, or concurrently for advanced students, the individuals

typically enroll in one of two programming language courses, either visual basic (CIS

130) or computer science I (CSC 150). Other courses that the enrollment services office

indicated these students register for are; computer hardware and networks (CIS 350), and

management of information systems (CIS 325).

Students enrolled in these courses during the summer of 2006 were given the

opportunity to complete the survey instrument that is used in this dissertation to measure

the student’s acceptance of tablet personal computing devices at this campus. Table A-1

indicates the courses surveyed, when they were surveyed (date & time), and the instructor

who granted permission to conduct the survey during normal class period times. It was

decided to conduct the survey in this manner to increase the participation of students in

the process.


Table A-1. Courses on Which Survey Tool was Administered Course Name Time Date Instructor Hardware & Networks CIS 350

MWF 8am 6/5/06 Mark Moran

Hardware & Networks CIS 350

MWF 9am 6/5/06 Mark Moran

Intro to Computers CSC 105

MWF 9am 6/5/06 Pam Rowland

Intro to Computers CSC206

MWF 10am 6/5/06 Pam Rowland

CSC 206 Advanced Access

MWF 10:40 6/6/06 Mark Moran

Computer Science I CSC 150

MWF 9am 6/08/06 Nick Youngworth


MWF 2pm 6/06/06 Nick Youngworth

Visual Basic CIS 130

MWF 10am 6/09/06 Connie Daniel


MWF 11:05 6/09/06 Connie Daniel


MWF 11:40 6/09/06 Jill Schneider


TR 8:00 am 6/13/06 Connie Daniel


TR 9:05 6/13/06 Jill Schneider


TR 2:35 6/13/06 Jill Schneider


TR 10:30 6/13/06 Tom Farrell

The participants were allowed to not participate in the survey if they did not wish

to do so. The survey administrator (researcher) spent less than a minute guiding the

participants to the link that would take the students into the survey instrument. The

reason for the survey was explained, “to learn how students at DSU feel about use of the


Tablet PC”, and then asked to please contribute to the knowledge of technology

acceptance at DSU. The survey, in its entirety, can be viewed at

www.homepages.dsu.edu/moranm/research . Table A-2 indicates the course enrollment,

participation, and percent of the possible participants that completed the task.

Table A-2. Individual Course Section Participation Course Name Enrollment Participation Percent Hardware & Networks CIS 350

14 9 64%


24 12 50%

Intro to Computers CSC 105

15 15 100%

Intro to Computers CSC206

25 21 88%

CSC 206 Advanced Access

18 16 89%


15 7 47%


29 21 72%

Management of Info Sys CIS 325

27 23 85%


15 14 93%


34 25 74%


34 26 76%

27 23 85% Visual Basic CIS 130

24 22 92%


24 13 50%


36 21 58%

Totals 361 268 74%

APPENDIX B: SURVEY QUESTIONS

These are the statements included in the survey instrument. The survey

participants’ responses were expressed using a seven point likert scale where one

represented strong disagreement and seven corresponded to strong agreement to the

statement.

Performance expectancy (PE) Questions.

Using the Tablet PC in my classes would:

1. enable me to accomplish tasks more quickly.

2. hamper my performance.

3. would increase my productivity.

4. hamper my effectiveness in class.

5. make it easier to do my homework.

6. hamper the quality of the work I do.

7. because my classmates perceive me as competent.

8. increase the instructors respect for me.

9. decrease my chances of getting a good grade.

10. be useful in my classes.

Effort expectancy (EE) Questions.

1. Learning to operate the Tablet PC is easy for me.

2. I find it easy to get the Tablet PC to do what I want it to do.

3. My interaction with the Tablet PC is clear and understandable.


4. I find the Tablet PC to be flexible to interact with.

5. It is easy for me to become skillful at using the Tablet PC.

6. I find the Tablet PC easy to use.

7. Using the Tablet PC takes too much time from my normal duties.

8. Working with the Tablet PC is so complicated and difficult to understand.

Attitude Towards using Technology (ATUT) Questions.

1. Using the Tablet PC is a good idea. 2. I dislike the idea of using the Tablet PC. 3. Using the Tablet PC is pleasant. 4. The Tablet PC makes schoolwork more interesting. 5. Using the Tablet PC is fun. 6. I like working with the Tablet PC.

Social Influence (SI) Questions

1. People who influence my behavior think that I should use the Tablet PC. 2. People who are important to me think that I should use the Tablet PC. 3. Professors at this university have been helpful in the use of the Table PCs. 4. My advisor is very supportive of the use of the Tablet PC for my class. 5. In general, the university has supported the use of the Tablet PC. 6. Having the Tablet PC is a status symbol in my university.


Facilitating Conditions (FC) Questions

1. I have the resources necessary to use the Tablet PC. 2. I have the knowledge necessary to use the Tablet PC. 3. The Tablet PC is not compatible with other computer systems I use. 4. The help desk is available for help with the Tablet PC difficulties. 5. Using the Tablet PC fits into my work style.

Self Efficacy (SE) Questions

I could complete a task using the Tablet PC …

1. If there was no one around to tell me what to do as I go. 2. If I had seen someone else demonstrate how it could be used. 3. If I could call someone to help if I got stuck 4. If I had a lot of time to complete the job. 5. If I had just the built in help facility for assistance

Anxiety (ANX) Questions

1. I feel apprehensive about using the Tablet PC. 2. It scares me to think that I could lose a lot of information by using the Tablet PC and hitting the wrong key. 3. I hesitate using the Tablet PC for fear of making mistakes cannot correct. 4. the Tablet PC is somewhat intimidating to me.


Behavioral Intention (BI) Questions

1. Whenever possible, I intend to use the Tablet PC in my studies. 2. I perceive using the Tablet PC as involuntary. 3. I plan to use the Tablet PC in the next three months. 4. To the extent possible, I would use Tablet PC to do different things (school or not school) related. 5. To the extent possible, I would use Tablet PC in my studies frequently. Usage Questions

1. I use my Tablet PC in Slate mode. 2. I use my Tablet PC stylus for navigation. 3. I use my Tablet PC primarily as a notebook computer. I use Windows Journal with my Tablet PC. Other Information Questions 1. I am a male female 2. I am currently a at DSU. A. Freshman B. Sophomore C. Junior D. Senior E. Grad Student 3. My major is represented by the college at Dakota State University. A. Arts & Sciences B. Business& Info Sys C. Education. D. Graduate. E. Other 4. I began using computers regularly in school. A. Elementary B. Middle C. High D. College E. Do not use 5. Approximately how many months have you been using the Tablet PC? One Two Three Six Nine Twelve More than twelve months


APPENDIX C: CROSS TAB REPORTS

Cross tab reports were generated for all the different survey populations for each

survey variable. There were 53 statements in the final survey and five different survey

groups. The groups include gender, class status, college affiliation, first computer use,

and first tablet PC use. The cross tab reports for the variable PE1 and the five

subpopulations are included in the body of this dissertation under the titles of Table 16,

17, 19, 21 and 24. Contact the researcher if the other cross reports are desired for

examination.

Documents

A Modification of the Unified Theory of Acceptance Accepted and Signed and Use of Technolog