Tools of the Trade

Logic Models: The PictureWorth Ten Thousand Words

Karen Denard Goldman, PhD, CHESKathleen Jahn Schmalz, RN, EdD, CHES

It's what in the 1970s we used tocall a "Kodak moment"—every-thing you wanted to communi-

cate conveyed in a single, memora-ble picture. Back in the 1970s, logicmodels first began to attract atten-tion . . . and they continue to holdour attention today, providing uswith the framework to plan and fund(!) powerful programs that producepowerful results.

A logic model is like a carefullycrafted curriculum. Everything con-nects. A thread runs through it thatlinks everything, explaining whatand why in no uncertain terms.Everything that comes makes sensebased on what came before. In termsof program planning, you knowexactly where you're going becauseyou know precisely where you'vebeen and what's coming up next onthe horizon. It's the "yellow brickroad" of quality program/interven-tion planning, except that it makesthe road map from Kansas to Ozcrystal clear to you and all whomyou involve in its development anduse. The "wizards" who helped uswith this tool include the wonderfulpeople at the University of Wiscon-sin Extension who designed and putonline the self-study course "En-hancing Program Performance withLogic Models" (http://www.uwex.edu/ces/lmcourse/#stay_put); BobbyMilstein and Marshall Kreuter whoquite some time ago shared with us a

Health Promotion PracticeJanuary 2006 Vol. 7, No. 1, 8-12DOI: 10.1177/1524839905283230©2006 Society for Public Health Education

8

summary outline adapted from theCenters for Disease Control and Pre-vention (1999) (http://www.cdc.gov.eval) and the W. K. Kellogg Founda-tion (2000). The road to hell, theysay, is paved with good intentions.The road to quality program plan-ning, implementation, and evalua-tion is a far more pleasant journeyalong a logic model.

> WHAT'S A LOGIC MODEL?

It is a picture that

• illustrates a logical sequence ofevents to occur through a pro-gram (initiative or intervention)or a system of related programs tobring about change in response toa specific situation;

• combines major program ele-ments into a picture of how theprogram is supposed to work;

• portrays the underlying rationaleof the program or initiative;

• is often displayed as a flow chart,map, web, network, or table toshow the sequences of steps (notalways linear) that connect activ-ities or processes to programresults;

• can exist as part of a "family" oflogic models to display differentlevels of detail of a program, differ-ent perspectives of a program, orto highlight specific elements of aprogram for different audiences.

In short, it is a practical pictorialdescription of how a program oughtto work.

> OTHER NAMES FOR ALOGIC MODEL• Roadmap• Conceptual map• Blueprint

The Authors

Karen Denard Goldman, PhD, CHES, is the assistant com-missioner for citywide personnel development for the City ofNew York, a health education and social marketing consul-tant, and an adjunct professor of health studies at New YorkUniversity in New York City.

Kathleen John Schmalz, RN, EdD, CHES, is an associateprofessor and associate chair of the Health &• Human Ser-vices Department, Health Education Program, at the Collegeof Mount Saint Vincent in Riverdale, New York.

'

Tools of the Trade

• Rationale• Program theory• Program hypothesis• Program framework• Framework for action• Theory of change• Model of change• Theoretical underpinning• Causal chain• Chain of causation• Weight of evidence model

• WHY USE LOGIC MODELS• Bring detail to broad goals; helps

in planning, evaluation, imple-mentation, and communications

> INTRODUCTION

One of the most important aspects of conducting a good evaluationis the development of a model or framework that communicates theessential elements of the program or initiative. Varied methodologiesand much competition has been generated over the years for theapproach that is the "right" one. Goldman and Schmalz (2005) arguethat a lot of similar approaches exist, but a useful approach is one thatcontains the elements that allow for good communication and replica-tion. The authors draw a line down the middle of previous approaches,and provide us with a very useful, attractive, and engaging synthesis.Readers should pay particular attention to this article as it will defi-nitely help us all be better program designers and implementers.

Fran Butterfoss, PhD, Associate Editor, Evaluation & Practice

PROGRAM DEVELOPMENTPlanning - Implementation - Evaluation

SituationNeeds and

Symptomsversusproblems

Stakeholderengagement

ConsiderMission

Vlwon

Values

Mandates

Resources

Local dynamics

CoSaboralocs

Competitors

Intendedoutcomes

Program Action - Logic Model

Outputs

Acth/tties Participation

Staff

Volunteers

Time

Money

Research base

Materials

Equipment

Technology

Partners

What we do

Conductworkshops,meetings

Deliverservices

Developproducts.curriculum,resources

TrainProvide

counselingAssessFacilitatePartnerWork with

media

Who we reach

Participants

Clients

Agencies

Decision-makers

Customers

Satisfaction

What theshort termresults are

Learning

Awareness

Knowledge

Attitudes

Skills

Opinions

Aspirations

Motivations

What themedium termresults are

Action

Behavior

Practice

Decision-making

Policies

Social Action

What theultimateimpacts) is

Conditions

Social

Economic

Civic

Environmental

Assumptions

EvaluationFocus - Collect Data -Analyze and Interpret - Report

FIGURE 1 Framework for a Simple Logic ModelSOURCE: © 2002 by the Board of Regents of the University of Wisconsin System. Reprinted with permission.

Goldman, Schmalz / LOGIC MODELS

• Provide a frame of reference forone or more evaluations of theprogram

• Help to identify gaps in programlogic

• Make underlying beliefs explicit• Reveal assumptions concerning

conditions for program effec-tiveness

• Build understanding and pro-mote consensus about what theprogram is and how it willwork—build buy-in and team-work

• Help clarify what is appropriateto evaluate, and when, so thatevaluation resources are usedwisely

• Summarize complex programs tocommunicate with stakeholders,funders, audiences

• Can strengthen claims of causal-ity and be a basis for estimatingthe program's effect on endpointsthat are not directly measuredbut are linked in a causal chainsupported by earlier research

• Enable effective competition forresources

• Many funders request logic mod-els in their grant requests

• By relating program activities totheir effects, they keep stake-holders focused on achievingoutcomes while remaining flexi-ble and open to finding the bestways to do the work at hand

• Stakeholders experience rewardsas a result of working together tocreate and refine the models

• Allow stakeholder to clarify theprogram's basic strategies, thusimproving and focusing thedirection of the program

• Lead to better programs

MAJOR COMPONENTSAND ACTION STEPS

1. Situation/Priorities

• Assess the sociopolitical, envi-ronmental, and economic con-text in which the problem exists.

• Basis of all future action; must becorrectly understood.

• Ask:1. What is the problem/issue?2. What causes this problem?3. For whom (individual, house-

hold, group, community, soci-ety in general) does this prob-lem exist?

4. Who cares whether it isresolved or not?

5. What do we know about theproblem/issue/people that areinvolved? What research,experience do we have? Whatdo existing research and expe-rience say?

• Summarize in situation state-ment

• Determine priority based on• Size of problem• Seriousness of problem• Changeability of problem/

situation• Potential of possible inter-

vention to have an impact onhealth and/or

• PEARL (Vilnius & Dandoy,1990)

• Propriety: Is an interventionsuitable?

• Economics: Does it make eco-nomic sense to address thisproblem?

• Acceptability: Will the com-munity accept an emphasison this problem and will theyaccept the proposedintervention?

• Resources: Are funding andother resources available orpotentially available?

• Legality: Do current lawsallow the intervention to beimplemented and if not, is itworth the time, energy, andresources to work for legisla-tive change?

2. Inputs

• Identify resources and contribu-tions needed

• Include time, staff, volunteers,money, materials, equipment,partners, research base, andtechnology

3. Outputs

Define "what we do" or "what weoffer"Create activities, services, events,and products for individuals,groups, agencies who participateor are the intended recipientsInclude workshops, meetings,direct services, curricula,resources, training, counseling,media campaign, advocacy cam-paign, etc.Decide whether to include whatyou plan to do or what got done(different people define this com-ponent differently)

4. Outcomes

• Determine direct results or bene-fits for individuals, families,groups, communities, organiza-tions, or systems to be achieved;may be positive, negative, neu-tral, intended, or unintended

• Include immediate results(awareness, knowledge, attitudes,skills, opinions, aspirations, moti-vations); mid-term results(changes in behavior, practice,decision making, policies, socialaction); and long-term results(changes in social, economic,civic, environmental conditions)

5. Assumptions

• Identify beliefs about the pro-gram and the people involvedand the way you think the pro-gram will work

• The problem or situation• The resources and staff• The way the program will

operate• What the program expects to

achieve• The knowledge base• The external environment• The internal environment• The participants: How they

learn, their behavior, motiva-tions, etc.

• Remember: Assumptions underlieand influence program decisions

10 HEALTH PROMOTION PRACTICE / January 2006

I

Tools of the Trade

6. External Factors

• Identify factors in the environ-ment that can influence programsuccess and/or can be influencedby the program

• Include cultural milieu, climate,economic structure, housing pat-terns, demographic patterns, pol-itics and politicians, backgroundand experiences of program par-ticipants, media influence,changing policies and priorities.

• Remember: External factors canaffect:

• Program implementation• Participants and recipients• The speed and degree to

which change occurs• Staffing patterns and

resources available

> WHEN TO USE A LOGICMODEL

During Planning to:

• Clarify the situation that drivesthe need for an initiative

• Demonstrate how investmentsare linked to activities

• Clarify program strategy• Identify appropriate outcome tar-

gets and avoid over-promising• Write a grant proposal or a

request for proposals• Assess the potential effectiveness

of an approach• Set priorities for allocating

resources• Estimate timelines• Identify necessary partnerships• Negotiate roles and

responsibilities• Focus discussions and making

planning time more efficient

During Implementation to:

• Provide an inventory of what youhave and what you need to oper-ate the program

• Develop a program managementplan

• Incorporate findings fromresearch and demonstrationprojects

• Make mid-course adjustments

During Evaluation to:

• Identify differences betweenthe ideal program and its realoperation

• Frame questions about attribu-tion and contribution (i.e., imme-diate links versus all prior links)

• Specify the nature of questionsbeing asked (i.e., boxes versusarrows)

• Determine which indicators will(and will not) be measured

• Document accomplishments• Organize evidence about the

program• Prepare reports and other media• Tell the program's story

During Advocacy to:

• Justify why the program willwork

• Explain how resource invest-ments will be used

During Training to Show Staffand Stakeholders:

• How the program works• Where they fit in• What they are expected to do• How they'll know if the program

is working

>ADVANTAGES OF LOGICMODELS• Offers the power of visual

communication• Appeals to stakeholders with

short attention spans• Puts program elements in context• Reveals assumptions by showing

desired connections

• Ensures that critical processesand outcomes are not overlooked

• Makes stakeholders accountablefor processes and outcomes

• Highlights types of data needed• Provides a framework for inter-

preting information• Prevents "program failures"

because of inadequate organiza-tion/management

• Integrates research findings andpractice wisdom

• Allows comparison of the "ideal"versus the "real" program

• Clarifies options for setting prior-ities and allocating resources

• Enhances learning and com-munication

> DISADVANTAGES OFLOGIC MODELS• Doesn't adequately capture the

program's context• Demands a high degree of speci-

ficity• Challenges assumptions, which

can create discomfort• Can be time consuming to create• Risks oversimplifying complex

relationships• Relies on the skills of graphic

artists

»• CRITERIA FOR A GOODLOGIC MODEL• Includes logically linked activi-

ties and effects• Includes forces known to influ-

ence the outcomes of interest• Visually engaging (i.e., simple,

parsimonious)• Aimed at a specific audience• Designed to communicate a spe-

cific set of main points• Includes an appropriate degree of

detail given the purpose (not toosimple, not too confusing)

• Provokes thought; triggersquestions

• Is useful to its intended users

Goldman, Schmalz / LOGIC MODELS 11

CDC's YOUTH MEDIA CAMPAIGN. VERB™ LOGIC MODEL

Short-Term and Mid-Term OuU-oines Ix>ng-Term Outcomes

YMC Vision:All youth leading healthy lifestyles

YMC Mission:To increase and maintain physical activity amo

(9-13 year olds).

FIGURE 2 Sample Completed Logic Model From the CDC's Youth Media CampaignSOURCE: Huhman, Heitzler, and Wong. (2004). © Centers for Disease Control and Prevention.

>• HOW TO CREATEA LOGIC MODEL

Two major approaches:

1. Reverse logic, which is drivenby "But how?" questions

2. Forward logic driven by "Butwhy?" questions of "If. . . then"thinking.

>* SUMMARY

Think of the logic model as your"road map."

What would happen if you ven-tured off on a trip without a map?Would you ever get to your final des-tination? Even if you did, how muchtime would you have spent in trying

to find your way, when mappingyour journey would have given youdirection from the beginning?

Logic models

• provide a graphic description of aprogram (process, event, commu-nity initiative)

• show the relationship of programinputs and outputs to expectedresults

• make explicit the underlying the-ory of a program

• are made up of six components:situation, inputs, outputs, out-comes, assumptions, externalfactors

• are useful for developing under-standing, improving program-

ming, clarifying outcomes, focus-ing evaluation, and communi-cating to stakeholders.

REFERENCES

Centers for Disease Control and Prevention.(1999). Framework for program evaluationin public health. Morbidity and MortalityWeekly Report, 48(RR-11), 1-41.

Huhman, M, Heitzler, C., & Wong, F.(2004). The VERB™ campaign logic model:a tool for planning and evaluation. Retriev-ed on July 2004 fromhttp://www.cdc.gov/pcd/issues/2004/jul/04_0033.htm.Vilnius, D., & Dandoy, S. (1990). A priorityrating system for public health programs.Public Health Reports, 105(5), 463-470.

W. K. Kellogg Foundation. (2000, March).Draft logic model development guide. BattleCreek, MI: Author.

12 HEALTH PROMOTION PRACTICE / January 2006