The Logic Model GuideBook 2nd Edition

1
Introducing Logic Models
his chapter introduces logic models. There are two types: theory of change
and program. This chapter describes model benefits and uses and explains
the role of modeling in both program and organizational effectiveness.
The process of modeling begins with results. Regardless of type, quality models are
evidence based.
LEARNER OBJECTIVES
Explain the difference between models and modeling
Recognize the benefits and uses of logic models
Demonstrate how to read a logic model
Recognize types of models and their characteristics
Describe the ways that models can support effectiveness
Work in any sector, whether private, charitable or government, requires

design, planning, monitoring, and evaluation. Each of these functions solves
problems, and evidenced-based models are a great aid in any context. Perhaps
you have been asked to design a new program, lead a change project, create a
marketing strategy, or plan an evaluation. Did drafting a narrative to circulate
among colleagues feel fragmented or inadequate? Did you think, Where do
I begin? Logic models and modeling can be a potent option to resolve your
dilemma. The Guidebook provides the practical support you need to create and
use models. It will also enhance your understanding of the relationships
between actions and results. Step by step, we describe how logic modeling can
be used as both a tool and a process that resonate with learning and performance management.
2
Chapter IIntroducing Logic Models3
Basic Concepts
Models and Modeling
Logic models support design, planning, communication, evaluation, and learning.
They are often used when explaining an idea, resolving a challenge, or assessing progress. They can untangle and clarify complex relationships among elements or parts.
Logic models are a graphic way to organize information and display thinking.
They are a visual approach to the implicit maps we all carry in our minds about
how the world does or should work. Logic models are tools that convey a scheme,
program, or project in a brief, visual format. Logic models describe planned action
and its expected results. A model is a snapshot of an individuals or groups current
thinking about how their idea or program might work.
Modeling is a technique. The process of modeling encourages iterative development of an idea, program, or project. It can create a safe space to start a debate,
generate ideas, support deliberations, and allow one to think more clearly about
specific relationships. A single, coherent logic reflects a consistent thread that connects design, plans, execution, and evaluation. This thread of evidence-based logic
is critical to program and organizational effectiveness.
Modeling allows careful consideration of the relationship between activities and
results. When tackled by a team or small group of stakeholders, models can be
improved by engaging the knowledge and experience of others. We think modeling
is significantly underutilized as a valuable process with real benefits. We believe the
best models are socially constructed in a shared experience that is facilitated. The
shared understanding and meaning they produce among colleagues are valuable
and enable success in subsequent steps of implementation and assessment.
Logic Model Benefits

In addition to extraordinary execution, organizational effectiveness ultimately
requires design, planning, monitoring, and success measures. Logic models can
contribute to the quality of all of these. In Chapters 1 through 4, we address models
from the design and planning perspective. In Chapter 5, we offer more detail about
their use with monitoring and evaluation. Logic models also
Develop common language among stakeholders.
Offer highly participatory learning opportunities.
Document and emphasize explicit outcomes.
Clarify knowledge about what works and why.
Identify important variables to measure and enable more effective use of
evaluation resources.
Provide a credible reporting framework.
Lead to improved design, planning, and management.
When logic models and modeling are used as a standard technique, they can
influence an organizations effectiveness. Logic models offer the strategic means to
4PART I CONSTRUCTION
critically review and improve thinking. And better thinking always yields better
results. Modeling can happen well before resources are committed or final decisions
get made. This offers a way to pretest quality and limit risk.
Effectiveness is not limited tobut certainly depends ona clear vision, capable implementation, and the means to monitor both processes and results. Logic
models can be tremendous supports for creating and communicating a common
understanding of challenges, resources, and intended success. Moreover, models
can also be used to calibrate alignment between the big picture and component
parts. They can illustrate parts of or whole systems. Choosing a perspective can
influence the level of detail. When modeling, this specifies boundaries as well as the
breadth or depth of display. For example, a logic model can show the learning
objectives for an elementary Spanish curriculum, what a school district will do to
secure student achievement, or what the federal government will provide in educational resources for second-language learning.
Logic Models Defined

Logic models are a visual method of presenting an idea. They offer a way to describe
and share an understanding of relationships (or connections) among elements
necessary to operate a program or change effort. Logic models describe a bounded
project or initiative: both what is planned (the doing) and what results are expected
(the getting). They provide a clear road map to a specified end. The development
of models (or the modeling process) provides an opportunity to review the strength
of connection between activities and outcomes. Through the experience of critical
review and development, models can display participants learning about what
works under what conditions.
Models are the product of modelingwhich we believe is best done in small
groups of stakeholders with the aid of intentional facilitation. They complement
systems thinking as a tool and technique for achieving valid but simplified representations of real-world complexities. Common synonyms for logic models include
idea maps, frameworks, rich pictures, action, results or strategy maps, and mental
models. Although logical frameworks (logframes) and causal loop diagrams (systems dynamics) are used for purposes similar to logic models, they are fundamentally different but complementary tools.
Logic Model Uses

While often used in the nonprofit sector among large nongovernmental organizations
and foundations, logic models are of increasing interest among community-based
organizations and the private sector, too. Because models enhance learning through
the iterative exchange of information and experience, they offer important features to
organizations that value evidence, diversity, dialogue, feedback, inquiry, great planning, and teams. Models can be used in program design, planning, implementation,
and evaluation. For example, logic models can be used to design a marketing program,
display a purchasing process, describe a school districts education improvement plan,
create a community leadership program, or establish the best ways to resolve conflict.
Two Types: One Logic

We describe two types of models: theory of change and program. They differ by
level of detail and use but represent the same logic. A theory of change model is
simply a general representation of how you believe change will occur. A program
logic model details resources, planned activities, and their outputs and outcomes
over time that reflect intended results.
These two model types are different in their appearance and use. The level of
detail and features distinguish theory of change and program logic models.
Program logic models include more features than theory of change models. This
concept of view is important and is discussed further in Chapter 4 because it
influences the quality and utility of models. Theory of change models are conceptual, and program logic models are operational. Model types and their relative
features are indicated in Table 1.1.
Relative to time frame, level of detail, volume of elements, display, and focus, the
model types contrast. They are alike because they share the same research, theory,
practice, and/or literature. Essentially, the types are different views of the same
evidence-based logic that have a shared origin.
Model use differs in purpose(s). Theory of change models display an idea or
program in its simplest form using limited information. These models offer a
chance to test plausibility. They are the elevator speech or cocktail napkin outline
of an idea or project. Program logic models vary in detail but offer additional information that assists design, planning, strategy development, monitoring, and evaluation. Program models support a display that can be tested for feasibility. They are
the proposal version of an idea or project because they have fleshed out far more
detail that often includes activities, resources, outputs, and other elements of interest to those creating and/or using the model. The relationships between elements,
both the relative interaction and sequence, are valuable for understanding intended
work and causal connections. They can be a huge help in creating action plans.
Historical Background
Use of theory of change and program logic models began in the 1970s. Carol
Weiss (1995) and Michael Fullan (2001) and Huey Chen (2005) are among the
Table 1.1 Features of Model Types

Feature
Theory of Change Logic Model
Program Logic Model
Time frame
No time
Time bound
Level of detail
Low
High
Elements
Few (do + get)
Many
Primary display
Graphics
Graphics + text
Focus
Generic
Targets + specified results
pioneers and champions for the use of program theory in program design and
evaluation. U.S. Agency for International Developments logical framework
approach (Practical Concepts, Inc, 1971) and Claude Bennetts (1976) hierarchy of
program effectiveness were among the earliest uses of the types of visual displays
that have evolved into the program logic models we know today.
Logic models did not receive much recognition, however, until after the United
Way of America came out with its publication Measuring Program Outcomes in
1996. This publication promoted the structures and vocabulary of logic models.
The W. K. Kellogg Foundation also was instrumental in spreading the use of logic
models with its Logic Model Development Guide (2001). For those readers interested
in more detail on the historical evolution of logic models, see the references provided at the end of this chapter. Thinking about thinking, or metacognition, is
present in many new management and leadership texts. Because our thinking
affects our actions, its an area thats well worth understanding better.
Examples
In the examples that follow, we briefly explain the general concepts and terms
related to a theory of change and to a program logic model. Chapters 2 and 3 provide more depth. Although we show one of each type of model, it is important to
keep in mind that these are but two examples from a much broader continuum of
possibilities. There are many ways to express or display the ideas and level of detail.
Theory of Change Model Example

Figure 1.1 shows a simple theory of change model for leadership development.
Read from left to right, it suggests that some strategies, for example, curriculum and
experiences, can positively influence people so they can more effectively tackle
community challenges. This theory relies on the assumptions that training, experiential learning, and community orientation will have a substantial influence on
individuals skills and ultimately result in community development. It also relies on
a particular framing of the problem(s).
Chapter 2 focuses on creating theory of change logic models. They are the critical foundation for any change effort. Often, these models exist as part of an internal
mental framework that is dormant or undisclosed. They can also imply considerable knowledge, experience, research, and practice. The evidence base for theory of
change models typically is not made explicit.
Program Logic Model Example

Program logic models inventory, from start to finish, a specified program effort.
For example, a program logic model for a community leadership program (based
on the theory of change) would include the specified resources/inputs, activities,
outputs, outcomes, and impact. Resources or inputs are what are needed to ensure
the program can operate. Activities are the tactical actions (e.g., events, services,
Figure 1.1 Community Leadership Academy Theory of Change

Results
Strategies
Academy
Leadership
Curriculum
More and
Better
Community
Leaders
Community
Development
Academy
Leadership
Experiences
publications) that occur to fulfill the promise of each strategy. Together, activities
make up the program design. Outputs are descriptive indicators of what the specific activities generate. Outcomes are changes in awareness, knowledge, skill, or
behavior. The impact reflects changes over a longer period. Figure 1.2 displays a
simple program model for the same community leadership program shown as a
theory of change model in Figure 1.1.
This program model suggests desired results include more and better leaders
and community development. It implies the leadership development agenda is
about resolution of community challenges and that, if resolved, it contributes to
community development.
To read this model, first note the intended impact (ultimate aim) of the program: community development. Then, move to the far left-hand side, where
resources or inputs essential to the program are listed. Logic models employ an
ifthen sequence among their elements. When applied to the elements in each
column, it reads, If we have these resources, then we can provide these activities. If
we pursue these activities, then we can produce these outputs. If we have these outputs, then we will secure these outcomes, and so on.
This model is just one very simple representation of how a program might be
designed and planned for implementation. Many variations on this example could
represent program design and planning for community leadership development
that meets standards of logic and plausibility. We know that Figure 1.2, in fact,
represents a program with some definite flaws. More discussion about how the
Sponsors ($)
Marketing/
communication
campaign
Host and
facility
Participants
Faculty
Curriculum
and materials
Resources
Leadership
Experiences
Processes
Leadership
Curriculum
Content
Activities
Participant
satisfaction
Completion
rate
Participant
description
Outputs
Figure 1.2 Community Leadership Academy (CLA) Program Logic Model
Better Leaders
Increased
community
awareness and
action bias
New leadership
attitudes,
knowledge, skills,
and behaviors
Short Term
Graduates use
knowledge and
skills obtained
through the
program to
strengthen
the community
Intermediate/Long Term
Outcomes
Community
development
Impact
program could be improved through a mark up (or critical review) that tests the
program design is described in Chapter 4.
Program Logic Model and Evaluation Design

This guidebook also offers some support for using logic models to assist in evaluation design. This book will address only the framing of broad inquiry. At this level,
evaluation questions are the foundation for evaluation design and planning. If we
apply this to the community leadership program example, it is appropriate to focus
on the programs intended results. The summative evaluation question is, What
difference did the program make in the communitys development? Perhaps a place
to begin is in determining the contribution made by the program to the development of more and better community leaders. A clear, coherent program logic
model provides great assistance during evaluation design. A model points out the
key features and shows the relationships that need assessment.
In this example, an evaluation could consider both changes in the awareness,
knowledge, skills, and behavior of participants as well as the community development impact. Stakeholders (funders, participants, and other influentials) might
also want to know about the content selection and quality of training. They might
be curious about implementation fidelity and adaptation, too. Figure 1.3 demonstrates a program logic model with typical evaluation questions.
This program logic model is serving evaluation. The five key evaluation questions are applied at specific locations on the illustrated program model. Key questions for the Community Leadership Academy (CLA) displayed include
1. Is the CLA doing the right things?
2. Is the CLA doing things right?
3. What difference has the CLA made among participants?
4. What difference has the CLA made across the community?
5. What are the ways community needs can and should be addressed by the CLA?
Positioning questions on the program model identifies where evaluative evidence
might be found to address inquiry. Labeling on the model also helps to establish the
relationship between program, implementation (processes), results, and evaluation.
Question 1 tests the logic constructed during evidence-based planning. This question requires thoughtful connections be drawn across activity accomplishment,
implementation fidelity, and the attainment of desired outcomes/impact. It
addresses the overall effectiveness of the selected strategies and the related action in
achieving the desired results. Question 2 examines implementation fidelity/variance
as well as the scope, sequence, penetration, and quality of activities. Questions 3 and 4
focus on the extent to which outcomes and impact have been achieved. Question 5,
like Question 1, should span the whole model to surface program improvement
needs. Questions 1 and 5 are more reflective but are essential to improved effectiveness.
10
Is the CLA doing

the right things?
What are the ways that community

needs can and should be
addressed by the CLA?
Sponsors ($)
Marketing/
communication
campaign
Host and
facility
Participants
Faculty
Curriculum
and materials
Resources
Leadership
Experiences
Processes
Leadership
Curriculum
Content
Activities
3
What difference has the CLA
made among participants?
2
Is the CLA doing
things right?
Community
development
Impact
What difference has the CLA

made across the community?
Graduates use
knowledge and
skills obtained
through the
program to
strengthen
the community
Outcomes
Better Leaders
Increased
community
awareness and
action bias
New leadership
attitudes,
knowledge, skills,
and behaviors
Short Term
Participant
satisfaction
Completion
rate
Participant
description
Outputs
Figure 1.3 Community Leadership Academy (CLA) Program Evaluation Model
These evaluation questions can be very helpful in the initial design and development of the program, as they help to aim the program intervention. The next step
is establishing indicators. Models also help in guiding the conversation and exploration needed to determine indicators or the measures of progress for an effort.
These issues are addressed in greater detail in Chapter 5.
Limitations of Logic Models and Modeling

It is important to note that the proper reference, logic model, is no guarantee of
logic. While many models do demonstrate some modicum of logic, a logical representation does not equal plausibility, feasibility, or success. There is some danger in
seeing a graphic display on paper and considering it true. This notion of omnipotence stems from limited domain knowledge, vested interest, and lack of perspective.
Typically, models do not take unintended consequences into account, although every
program has side effects. The modeling process usually does not include program
critics, and most stakeholders are not likely to be grounded in the research literature.
Realistically, even when program theory and logic are constructed and build on
the insights of broad representative stakeholder groups, can anyone be sure who is
right? Every model should be considered a draft. They are deterministic, incomplete approximations of what usually are more open systems. They provide the
simple illustration that makes evaluation and program improvement more accessible to individuals and groups. The mere existence of a model does not mean that
the model or the plans it represents are ready for implementation or that it will
readily deliver the intended results!
Chapters 2 and 4 tackle model improvement and development in greater detail.
It is essential to note that a model is a graphic display of the program captured at
one point in time. Models, we believe, should change to reflect best thinking and
current evidence as these evolve. Creating and displaying variations of a model are
experiences that can develop thinking about strategies/activities and results. This
development is a critical process in model quality and, ultimately, in the feasibility
of the efforts described.
We believe the greatest value of logic models is their use in an iterative, intentional process aimed at improving the thinking they illustrate. This is best done
through a facilitated sequence with selected stakeholders. Obviously, logic models do
not ensure plan implementation fidelity or quality. Nor do they remedy any of the
many concerns about organizational structure and culture that can deeply affect
program and organizational effectiveness. Important action steps associated with
quality include identification of both assumptions and evidence used in models.
Models Begin With Results

Determining the results you desire is the first step in effectiveness, because knowing
where you are headed is critical to picking the best route to use. In our experience,
models begin with results. Results consist of outcomes and impact; each appears in a
sequence over time. While impact is the ultimate end sought, sometimes synonymous
with vision, outcomes are earlier indications of progress toward results. We think
results are the place to begin when you are struggling with choices about strategies
(with a theory of change) or activities (with a program logic model). It is important
to avoid moving prematurely to specify what you want to do. In any change work,
program design, or problem solving, specifying those outcomes most likely to occur
soon and then those that will take more time to emerge helps determine what route
(action path) might be best to use.
People commonly complain their work is both activity focused and frantic.
Considerable time and effort are spent on a flurry of tasks that frequently lack a
clear relationship to intended results. Logic models can assist in sorting priorities
because they both rely on and help build a visual literacy that makes action and
expected consequences clear. Through the models and modeling, stakeholders can
identify potent strategies/activities likely to contribute to the results sought. And
those with less (relative) value can be sidelined or discarded.
Logic Models and Effectiveness

In the workplace (and in life), almost everyone is interested in effectiveness. To that
end, we provoke important thinking when we ask these questions:
Are you doing the right work?
Can you make better decisions?
Are you getting superior results?
All of these questions apply in any contextwhether it is in government or in
the private or the nonprofit sector. They are among the most critical questions for
managers and leaders because they focus on key levers that influence performance.
We know from practical experience and assessment that doing the right work along
with great decisions secures superior results. Logic models can help with the design
that ensures the right work, the plans and implementation that reflect better decisions, and the evaluation that tests both pathways and progress toward success. For
these reasons, they are an exciting tool and process for anyone interested in more
effective programs, projects, and organizations.
Figure 1.4 demonstrates key points of the design, planning, implementation,
and evaluation that the two types of models can support. Theory of change models
are most helpful during the design of a program or project. As plans or evaluation
require greater detail, program logic models can make a substantial contribution to
these later stages of work. The types of models and their uses form a continuous
loop that can provide feedback about a program throughout its life cycle.
Logic models as both a tool and a strategic process offer considerable value to
programs and, subsequently, organization effectiveness. They can be used for different purposes at different times in the life cycle of an idea (program, project, or
change effort). Theory of change models can dramatically influence program
planning because they rely on knowledge to offer choices about doing the right
work. In this stage, the selection of strategies relative to results occurs. Program
Figure 1.4 Effectiveness and Logic Models

Can we make
better decisions?
Planning and
Implementation
Are we doing the
right work?
Start
Here
Program Logic
Model (PLM)
Improved Program
Effectiveness
Design
Theory of
Change (TOC)
Evaluation
Are we achieving
superior results?
logic models help with more precise decisions about which activities in a given
strategy are most effective. Program logic models can also be used to support
evaluation design. They can assist in pointing to optimal areas of inquiry and help
to determine whether progress is being made and what difference has occurred
relative to results.
Some organizations use logic models routinely. They can become a standard tool
that promotes alignment and synergy. For example, evaluation can be designed and
implemented more easily when a clear theory of change and program logic model
have already been determined. These tools and related processes can also assist
learning and dissemination in significant ways. Logic models and modeling can be
vital elements in performance management because they rely on evidence, support
informed decisions about strategy, and assist with assessment. Performance management seeks predetermined results and adapts actions to obtain them.
IN SUMMARY
Logic models are simply a visual display of the pathways from actions to results.
They are a great way to review and improve thinking, find common understandings, document plans, and communicate and explicate what works under what
conditions. We think theory of change models are distinct from program logic
models in several important ways. Theory of change models present a very highlevel and simple explanation of do and get. Program logic models offer a detailed
map that can be implemented when supplemented with work plans. In this chapter,
we also distinguished between models as tools and modeling as a process. A quality
feature of logic models is that they are evidence based. Logic models can be used
for learning, improving, and greater effectiveness.
LEARNING RESOURCES
Reflection
1. In what circumstances can you use logic models in your work or field of study?
2. What benefits does each type of model provide? And to whom?
3. What do logic models display? And what is missing?
4. How are theory of change models and program models alike? Different?
5. What kind of logic models have you seen before? Which are most commonly
used?
6. What current models/processes are commonly used for program design in your
organization? What work cultures are best suited for logic models?
Application
Select and draw one of the following: promotion of a new brand of ketchup, a
drivers training program, or a domestic violence awareness campaign. Have others
independently draw the same project you select. What do all the drawings have in
common? What areas are different? Why? When and how do these differences
become reconciled? How did the levels of detail differ among the drawings? What
can these drawings tell us about mental maps?
References and Supplemental Readings

Texts
Chen, H. (2005). Practical program evaluation: Assessing and improving planning, implementation and effectiveness. Thousand Oaks, CA: Sage.
Collins, J., & Hansen, M. T. (2011). Great by choice: Uncertainty, chaos and luckwhy some
thrive despite them all. New York: HarperCollins.
Fullan, M. (2001). The new meaning of educational change (3rd ed.). New York: Teachers
College Press.
Funnell, S. C., & Rogers, P. J. (2011). Purposeful program theory: Effective use of theories of
change and logic models. Hoboken, NJ: Wiley & Sons.
Kane, M., & Trochim, W. M. K. (2006). Concept mapping for planning and evaluation.
Thousand Oaks, CA: Sage.
Murray, D. K. (2009). Borrowing brilliance: The six steps to business innovation by building on
the ideas of others. New York: Gotham Books.

Murray, D. K. (2011). Plan B: How to hatch a second plan thats always better than your first.
New York: Free Press.
Practical Concepts, Inc. (1971). The logical framework. Approach and training materials developed for the U.S. Agency for International Development. Washington, DC. Unpublished
manuscript.
United Way of America. (1996). Measuring program outcomes: A practical approach.
Alexandria, VA: Author.
W. K. Kellogg Foundation. (2001). The logic model development guide. Battle Creek, MI:
Author. Retrieved December 1, 2011, from http://www.wkkf.org/knowledge-center/
resources/2006/02/WK-Kellogg-Foundation-Logic-Model-Development-Guide.aspx
Weiss, C. H. (1995). Evaluation (2nd ed.). Upper Saddle River, NJ: Prentice Hall.
Journal Articles
Bennett, C. (1976). Analyzing impacts of extension programs, ESC-575. Washington, DC: U.S.
Department of Agriculture, Extension Service.
Dwyer, J. (1996). Applying program logic model in program planning and evaluation. Public
Health and Epidemiology Report Ontario, 7(2), 3846.
Fitzpatrick, J. (2002). A conversation with Leonard Bickman. American Journal of Evaluation,
23(3), 6980.
Funnell, S. (1997). Program logic: An adaptable tool for designing and evaluating programs.
Evaluation News and Comment, 6(1), 517.
Julian, D. (1997). The utilization of the logic model as a system level planning and evaluation
device. Evaluation and Program Planning, 20(3), 251257.
Julian, D. A., Jones, A., et al. (1995). Open systems evaluation and the logic model: Program
planning and evaluation tools. Evaluation and Program Planning, 18, 333341.
Sartorius, R. (1991). The logical framework approach to project design and management.
Evaluation Practice, 12(2), 139147.
Internet Resources
Davies, R. (2008). The logical framework: A list of useful documents. Retrieved December 1, 2011,
from http://mande.co.uk/2008/lists/the-logical-framework-a-list-of-useful-documents/
Duigan, P. (n.d.). Brief introduction to program logic models (outcome models). Retrieved
December 1, 2011, from http://outcomescentral.org/programlogicmodels2.html
Evaluation logic model bibliography. (n.d.). Madison, WI: University of Wisconsin Extension
Service. Retrieved December 1, 2011, from http://www.uwex.edu/ces/pdande/evaluation/
evallogicbiblio.html
Israel, G. D. (2001). Using logic models for program development. Retrieved December 1, 2011,
from http://edis.ifas.ufl.edu/wc041
List, D. (2006). Program logic: An introduction. Retrieved December 1, 2011, from http://
www.audiencedialogue.net/proglog.html
McCawley, P. F. (1997). The logic model for planning and evaluation. Retrieved December 1,
2011, from http://www.uiweb.uidaho.edu/extension/LogicModel.pdf
2
Building and Improving Theory
of Change Logic Models
his chapter identifies the basic elements of a theory of change logic model. They
are evidence based and plausible. This chapter describes the steps to create and
improve a theory of change model. It also names criteria for a good model.
LEARNER OBJECTIVES
Identify basic elements of a theory of change model
Identify the contributions a theory of change model lends to a change
effort
Create a simple theory of change model
Apply critical review for theory of change model plausibility
Logic models offer an exciting way to combine narrative and graphics to display
the mental maps people hold about a specific program or change initiative. These
mental constructs are also sometimes called idea maps. While the process of creating a model can be solitary, there are significant benefits when models are generated in a small group among stakeholders with a shared agenda. Logic models can
be used over the entire life of a change efforttheir boundaries should be consciously determined by the participants who create the model.
Building a Theory of Change Model

Getting Started
While logic models can be used for many purposes, there are two basic types: theory of change and program models. Understanding these types is important to
their development and use. The choice of which to use reflects whether the model
16
Chapter 2Building and Improving Theory of Change Logic Models17
needs to describe broad and general concepts about change or more detailed
operational elements essential to design, plans, and management. It is possible to
begin with either a program logic model or theory of change model.
We believe it is important that a program model always accompany a theory of
change because the assumptions held in the theory of change have fundamental
value for program operations and success. These assumptions should be consistent
and anchor choices made in the development and selection of strategies to fulfill
intended results. When assumptions are evidence based, then a single coherent
logic and alignment can occur that enables success. Relying on knowledge, whether
theory, research, practice, and/or literature, is essential to a good model.
Preferences and Styles

People vary considerably in what level of detail they prefer to describe their mental
maps. This is an important consideration for those who lead the modeling process.
Invariably, any small group will include people with a strong preference to start at a
broad, high level and those who feel far more comfortable beginning with detail. Both
approaches have value because the best program or change effort design eventually
should display models with these features. Accommodating differences in how any
individual approaches the display task is a common tension in the creation process.
Our aim is to guide you consciously from big ideas to finer points. For this reason, we provide a theory of change model description and example first, then a
parallel representation of a program logic model in the next chapter. So that content matter does not confound the process, we have chosen to use community
leadership development and health improvement for all model content in Chapters
1 through 5. A broad range of subject content is offered in the models found in
Chapters 6 and 7.
Evidence Based and Plausible

Theories of change can be grounded either in an established claim with proof or in
a hypothesis. Programs based on proofs are a replication of something that has
worked. Hypotheses are rationales based on research literature that show promise
of working and are therefore something worth trying. Programs based on hypotheses are innovations. If the theory of change is supported by a body of evidence,
there is a stronger chance that the strategies chosen will secure the desired results.
Frequently, however, this standard is overlooked. In the urgent fever to get to
implementation, the design and plan quality can be shortchanged and rely, instead,
on faulty assumptions, old practice, or little or no evidence.
We suggest practitioners construct the theory of change model with grounding
it in literature, experience, or other evidence that promotes plausibility. Most theories of change will exhibit some degree of logic. Plausibility, however, is a more
appropriate litmus test for work that has inherent opportunity cost. Later, in
Chapter 3, we suggest that the program model must also be feasible if there are
authentic intentions of securing results. During the construction of a change
model, it is important to explore or discover what works under what conditions.
This is about the choices made in selection of strategies relative to anticipated and
therefore planned results. When constructing a program logic model, the realistic
criterion of limited resources is also in play. Any program, project, initiative, or
organization has some limits on time, talent, and financial resources. In the migration from theory of change to program logic model, users can shift their thinking
from what could work to what should work.
The Big Picture

A theory of change logic model offers the big picture of strategies that could generate your intended results (or impact). This construct is illustrated by Figure 2.1.
A basic theory of change model contains just two elements: strategies and
results. The intent is to illustrate the connection between what you will do with
what you hope to get.
Strategies reflect a choice of optimal actions (via activities or tactics) to secure
intended results. They represent an allocation of resources focused on a clearly defined
objective. Marketing, training, political advocacy, and fund development are examples
of common strategies. This element describes your actions or what you plan to do.
Strategy is the overall plan that gives coherence and purpose to the specific
actions that organizations undertake. For some nonprofits, however, the meaning
can be murkier and framed as an ambiguous aspiration. When we use the term
strategy, we rely on an implicit but evidence-based assertion that connects means
and ends. Wherever the word strategy appears in our illustrations and narrative, we
assume that the case for selection is soundin other words, that it has strong
potential to secure impact. Later, in Chapter 3, we also use the term as an umbrella
for nested clusters of activities (or tactics) that aim at specific single outcomes or
clusters of outcomes.
Results reflect the long-term effect of strategies. They are the get from what we are
doing. Results are ultimately secured through the change(s) generated by the preceding
strategies. They can reflect a single outcome or multiple outcomes over time.
Multiple Strategies and Results

In reality, many programs are more complex than is shown in Figure 2.1. Most
often, several strategies combined (over time) in a particular sequence yield results.
And we generally both create and experience results as the net yield of several strategies working together. When displaying theory of change models, this can be
Figure 2.1 Basic Theory of Change
Do
Get
Strategies
Results
challenging both to conceptualize and to represent. For example, to become proficient in a new language, it is most likely the combination of instruction, practice,
and cultural immersion that generates proficiency. Likewise, to be a profitable
cereal company may require a high level of competency in research/development,
marketing, production, and distribution strategies. In addition, great health outcomes for neurosurgery may rely on expertise in diagnostics, surgical techniques,
pre- and postsurgical care, and rehabilitation therapies. Because multiple strategies
often contribute to results, a more complex representation of a theory of change
might look like Figure 2.2. We call the path from each strategy to result a strand.
Recall the theory of change model for the Community Leadership Academy
(Figure 1.1). In that example, the outcome desired was more and better community
leaders. This model described two simple strategies as essential contributions to the
recipe: curriculum and experiences. An applied example of a multi-strategy model
for securing improved health is displayed in Figure 2.3. Read from left to right, the
theory of change suggests if we provide exercise, nutrition, stress reduction, and
some other key strategies, then we will secure improved health for participants who
follow the program. It is also important to note that strategies may interact
(although not shown here). This theory of change represents a generalized construct for many health-improvement programs.
Figure 2.3 simply provides a gross summary of strategies and intended results for
a health-improvement program. A theory of change model displays some of the
underlying assumptions about change and is a view at 65,000 feet. This view is how
a farm looks from an airplane window in contrast to the view from a tractor seat. It
simply shows the specific strategies that the designer believes will achieve a desired
result. Theory of change models are distinct in that they include assumptions (either
explicitly or implicitly), offer the big picture of the bounded mental map, and name
impact. Theory of change models do not provide the detail essential to action planning, implementation, or evaluation. They simply state what you plan to do and
Figure 2.2 Multiple-Strategy Basic Theory of Change
Do
Get
Strategy 1
Strategy 2
Results
Strategy 3
Strategy 4
Figure 2.3 Health-Improvement Theory of Change

Do
Get
Exercise
Nutrition
Improved
Health
Stress
Reduction
Recruitment
and Retention
what you expect to get. It is easier to explore ideas and manipulate them at this stage.
The why and how of these models are embedded in assumptions and eventually
reveal themselves in the strategies selected. For example, in Figure 2.3, some of the
underlying assumptions might include the following:
Increased exercise and improved nutrition are known to contribute to
improved health.
Only those who participate in the program will achieve results. Participants
need to be recruited.
Stress may be a contributing factor to poor health.
Participants fidelity to the program is critical to achieve results. They will
need parallel increases in awareness, knowledge, and skill in order to change
behaviors that most impact health.
Realistic Models
Theory of change models should demonstrate plausibility. This means they could
work. Given the realities of limited time, as well as human and social resources,
logic alone is inadequate. In fact, the logic displayed in a model can be uninformed
or misinformed. For example, world peace is a tangible and clear desired result, but
a theory of change that relies solely on communication (e.g., newsletters and websites) is not plausible in securing world peace. Or consider the desired result of
hiring more mid-level scientists at your research institute. Are outreach strategies
with local math and science teachers and students logical action steps? Yes, but
meetings with those targets can be helpful only in a pipeline that can tolerate a
decade of delay. It is not a best strategy given urgent human resource needs this
week and next month.
Knowledge and Assumptions

So far, we have described a basic theory of change model for improved health that
is specifically composed of doing (strategies) and getting (results). Each of us brings
along some other contributions to our theory of change that are more closely held.
While not often named, we commonly bring what we believe (our assumptions) to
theories of change, too. The most viable assumptions used to select strategies are
rooted in knowledge, and that knowledge generally includes research, practice, and
theory. Figure 2.4 illustrates the knowledge base for beliefs that precedes assumptions and strategies in a theory of change.
It is critical to recognize the role of beliefs. They are important determinants in
choices about strategies for both creating and improving a theory of change model.
Figure 2.4 illustrates how knowledge and beliefs contribute to a programs underlying
or driving assumptions. Assumptions are often informed by knowledge, which can
include research, practice, and theory. We find that making assumptions explicit can
improve our chances for program success. Sometimes assumptions are informed by
experiences, habits, or values that do not also reflect knowledge. Mediating or moderating factors such as program context are useful to consider as barriers or facilitators to
program success at this stage. Dogma, misinformation, ignorance, and wishful thinking
are hazards here. Often, assumptions can differ significantly among and between both
stakeholders who create and those who execute. They can also dramatically affect how
problems are identified and framed. For model utility, its important to cite what
problem(s) were trying to solve and find a way to frame a problem so that it is meaningful to others.
Modeling can help surface vital differences among stakeholders and offer a disciplined process for resolution based first on plausibility, then on feasibility during
subsequent versions. This is why, in part, modeling offers considerable value
beyond the construction of models alone. Its important to note that dialogue is
critical to exploration of knowledge and assumptions that are embedded in models.
Engaging multiple stakeholders is critical to quality as well as meaning.
However, modeling can be an uncomfortable process because it nearly always
raises differences among participations perceptions, experience, knowledge, training, and other factors. Identifying and negotiating these can be challenging. This
navigation is most easily done with external assistance. If not, then it can be useful
to explicate the criterion for decisions about model content and display. Simply
who participates in modeling can be loaded with politics, since it will very likely
influence the model content.
Action Steps: Creating a Theory of

Change Logic Model
We recommend that people begin building a theory of change model by specifying
their intended results. Most often, it is easiest to be clear first about the intended
results. Our experience with clients is they know what they want to accomplish.
They often label this as desired results.
Figure 2.4 Informing a Theory of Change
Research
Practice
Believe
Assumptions
Do
Get
Strategies
Results
Theory
Knowledge
Once results are named, we suggest identifying the strategies required to achieve
the results you seek. Strategies are about how intended changes will occur. And
assumptions are the preceding knowledge: the research, practice, and theory that
inform choices about strategies. They significantly influence which strategies are
chosen as pathways to your intended result.
So the steps to generate a theory of change logic model are ordered in
this way:
1. Identify results desired.
2. Name the strategies that will deliver your intended results.
3. Define the assumptions (see Figure 2.4) that support your specified strategies.
Figure 2.5 displays these actions in a three-step sequence. Most theory of change
models generally do not display underlying beliefs or assumptions. Nevertheless,
these are important elements to explore consciously when creating a theory of
change. We suggest those assumptions are named in association with the theory
of change. Assumptions can simply be a bulleted list on the same page or reverse
side. Remember, a theory of change model is simply one representation of the
truth, not a substitute for it. The model draft becomes a place for starting discussion and testing meaning, coherence, assumptions, and plausibility. Engaging
others in modeling offers the opportunity for critical review and improvement
over time through the generation of versions. By starting with a theory of change
model, it is easier to arrive at shared understanding of what your program will do
and can achieve.
Figure 2.5 Steps in Creating a Theory of Change Model
Step 3
Define
Underlying
Assumptions
Strategies
Results
Step 2
Name
Strategies
Step 1
Identify
Results
Improving Theory of Change Models

We offer several common processes to consider as you explore iterations of your theory
of change model. While improvement is definitely not limited to these suggestions, an
application of these four will likely contribute to the development of any attempt:
Engage multiple stakeholders.
Share explicit assumptions.
Toggle or test alternative content in model versions.
Explore promising practices and consider benchmarking.
Multiple Perspectives
People hold and operationalize theories of change in both their work and personal
lives. Most experienced parents, for example, have a recipe that contains the primary strategies they believe are vital to parenting a good kid. Parents can vary
considerably, however, in what they mean by a good kid. Likewise, even if we agree
on what a good kid might know and be able to do, it is highly likely that from one
parent to the next, there will be many variations on parenting strategies to ensure
the good kid result. This example suggests the considerable importance of ensuring that all stakeholders in your program or change effort are specifying results and
the strategies needed to get there with the same meaning and level of specificity.
Developing and improving the theory of change for your program is one way to
begin the conversations needed to reach shared understanding.
In the health example we started this chapter with (see Figure 2.3), we identify
improved health as the result sought. It is important to ensure that everyone has a
highly consistent understanding of what improved health means. To one participant, it may be weight loss. Another could interpret it as normal blood pressure.
Others may feel improved health is a combination of several positive outcomes. If
you ask a half-dozen people what improved health means to them, it is quite likely
there will be variation in their individual answers.
Specifying what the results mean, such as improved health in this example,
becomes critical for your program design as well as essential for measuring progress
toward and determination of results. If the meaning and measures of results are
shared and understood similarly, then it is more likely strategy choices will align with
your intended impact. It is more likely indicators of progress will be appropriate, too.
Unpack and Share Assumptions

The most significant opportunity to improve theory of change models lies in
unpacking the knowledge and beliefs employed in assumptions. This means, in
practice, that any theory of change for a program or social change effort should be
grounded in knowledge. If results are connected to strategies that reflect research,
practice, theory, and experience, there are far greater chances for success than with
strategies that lack this grounding. Figure 2.6 displays a combination of the elements found in knowledge. A combination of little or no practice, experience,
research, and theory in your model means the effort it represents is an idea that may
be highly innovative but is not likely to succeed. A combination of practice, experience, research, and theory in your model suggests the effort it represents is more
likely to succeed.
The best theory of change model deliberately pursues alignment among research,
theory, practice, and experience. The stronger models build on the knowledge and
good work that precede them. These substantiated models and their associated
programs or social change efforts gather and then use codified knowledge from
prior efforts to inform effective program design. Figure 2.6 describes the geography
of choices and emphasizes that relative success relies on a depth of practice, experience, research, and theory.
Figure 2.6 Success in Theory of Change Models
Research and Theory
Practice and Experience

Low
High
Low
High
Least Likely
to Succeed
Most Likely
to Succeed
Toggling
Another practical way to improve models is what we refer to as toggling. We
define toggling as finding the optimal fit between a selected set of strategies and
plausible results. For example, options to improve school nutrition could include
planting a garden, removing vending machines, or changing the lunch menu.
Toggling tries on options and makes a best choice. In this critical review, the
model builder is experimenting with the best combination of strategies to secure
the results sought. Inviting others to join in this iterative tactic in real time can be
very productive. Sometimes it is best to refine or focus the specified results. For
example, a program or social change effort could specify one of these results: end
childhood obesity or create schools with improved nutrition choices for children.
During toggling, it may become apparent the result is not plausible. Often, ambitions are greater than what is feasible. It is important to guard against grand ambitions. They are possible hazards that can result in flawed models. Figure 2.7
demonstrates the interactions between strategies and results as choices are made in
the final determination.
This figure displays the testing that occurs in an effort to determine the best
combination of strategies to secure your intended results. Once a preliminary theory of change model is drawn, the modeling process begins. The model is tested
though iterative cycles of inquiry. The basic questions addressed are Are the results
focused and narrow enough to discern optimum strategies? and Is the connection
between the strategies named and the results desired as strong and direct as needed
to be effective with the population of interest?
Toggling can also involve a review of both the duration and sequence for chosen
strategies. The objective is to specify a model that is plausible. The specifics of what
is feasible are developed in the program logic model (and are discussed in following
chapters). People sometimes mention confusion when they talk about the fog of
war. In our experience, there is considerable fog or ambiguity in program design
and planning. A clear and plausible theory of change is the foundation of intended
work and requires considerable attention and scrutiny. Just as with the Cheshire Cat
in the story Alice in Wonderland, if you do not know where you are going, any road
will do. Fuzzy, ambiguous theories of change rarely net the success intended.
Figure 2.7 Toggle: Improving Theory of Change Models

Do
Get
Testing Design/Plans
Strategies
Results
Testing Effectiveness
Promising Practices and Benchmarking

It is valuable to explore strategies and results of programs (or social change efforts)
similar to yours. A better understanding of the rationale for their strategies and
related results can deeply inform your design choices. In the private sector, this is
often referred to as benchmarking, a systematic discovery and comparison process that can be a simple way of both clarifying and improving your design early on.
In benchmarking, one simply looks around at promising practices to inform and
then make good choices about your own. Benchmarking may include a review of
documents, a survey, and discovery with peers/competitors. It establishes the status
of other efforts, programs, or organizations on specified features or issues. The big
questions benchmarking can help answer are what others are doing and why?
Chapter 7 provides some examples of archetypes, general recipes that are evidence
based, that can also help in your early efforts to construct a model.
School-improvement efforts, for example, often include quality instructional and
assessment practices among their many strategies aimed at teachers to, ultimately,
improve student academic achievement. There is evidence these strategies can positively influence student academic achievement. It follows that a school-improvement
effort, then, might be more successful if these strategies (or some adapted version)
are included in the program plan. The converse is also true. A school without these
strategies as standard operating procedure or as part of a new plan is less likely to
secure improved student academic achievement. A theory of change can show what
you are and are not thinking about. The accompanying modeling process provides
the opportunity to reflect and improve on the underlying logic for your program.
Group Process
Consider involving others in co-creating a theory of change model. Lets build on
the improved health example from earlier in this chapter and aim at obesity prevention. How could you guide a group in exploring a countywide program design
intended to maintain healthy weight and prevent obesity? In tackling this question,
its important to anticipate the need for data prior to the convening. Gathering and
sharing information about research, practice, and theory makes for a much smarter
dialogue. Its also possible to include experts who bring data and field experience
literally to the table. In general, a guided group process could follow these action
steps in a daylong work session or over a series of meetings.
Remind participants, again, of the intention of the work to establish a theory
of change that articulates a single relationship between results and strategies. The
assignment is to identify strategies most likely to get the planned results given the
context, target audiences, and other factors. So its important, first, to secure a
shared understanding of the results intended. Ask all the participants, on their
own, to identify the result they want the program to achieve in the next 3 years.
Its vital to specify a period to bound the program effort. Have participants post
directly (or transfer) their intended results for public sharing. This first posting
will likely display a range of expectations and assumptions about what results are
desired. Reconcile those that are similar and do discovery on whats underneath
the postings.
Through dialogue, find the result that the group believes is most feasible given
the context. Features of context might include historical and current rates of obesity and overweight, definitions of those terms, an inventory of physical fitness
options and their physical proximity, socioeconomic data for the county population, and access to healthcare and weight loss resources, along with aspects of prevailing culture. Create a list of resources, including specific funds that could be
designated for the program. Your participants can probably name many other features of context. These are the influences as well as data that help to inform the
current reality. It may help to post facts and features of context so they are present
to dialogue. This portion of the process should rely on facts as well as perception.
Then, consider your target audience(s). Will your program effort be designed to
influence males, females, teens, young adults, all residents between 10 and 50 years
of age? Or some combination of these characteristics? Employ learning from the
context discussion to inform your choices. Be aware the selection you make may
require you to adjust the groups intended result. The effort to name and understand the results is well worth the effort because it frames subsequent action steps.
Last, ask participants to name strategies that the program should include. Post
them. Often, people will name tactics or specific activities. Getting to the same level of
detail just requires some modification. This is another great opportunity to insert
more information. For example, identify independent research, practice, and theory
shown to influence weight management. Share some benchmarking information from
effective programs that have already tackled this same challenge and those that failed
to make progress. Be sure to include their costs and related organizational resources.
Ultimately, the group should determine a clear list of strategies and specified results
that are not simply feasible but optimalthat is, highly likely to secure the impact.
This may require some toggling. Use the Guiding Questions (below) to critically
review the work of the group. Look forward to Chapter 6 and review the New York
state Healthy Weight Partnership. It offers some great ideas about strategies and results
(defined by their mission and vision). The NY model cites target sectors/settings to
segment their program plans since the work is focused on all state residents.
As you construct, then review a theory of change, the following questions may
be useful:
Guiding Questions for Reviewing

a Theory of Change Model
1. Are the results specified with shared meaning among all stakeholders?
2. Did we uncover our assumptions and carefully examine research, practice, and theory as the grounding for our choices in strategies?
3. Did we toggle between strategies and results to ensure plausibility
given our assets and limitations?
4. Have we carefully reviewed similar programs to learn what strategies
worked under what conditions to secure results?
5. Does the model clearly show the relationship of strategies to results?
Nonlinear Theory of Change Models

Theory of change logic models are not always displayed in a linear fashion (as they
have been in the text so far). Realistically, few theories, programs, or change efforts
occur in a precisely linear sequence. The world is much more complicated and
integrated than the simple, step-by-step actions and reactions as drawn here. Most
change occurs iteratively, or in cycles, and with multiple interactions among many
features. Any change is also connected to a much larger system than illustrated by
the theory of change model. In our experience, systems and holism can be difficult
to manage and even harder to evaluate or communicate. In generating a theory of
change, it is important to represent how change occurs as an aspect of a far more
comprehensive and vast geography. When using a systems view, it is important to
consider the key leverage points or strategies that are most influential given time,
expertise, and resources. A simple example of a nonlinear theory of change logic
model is shown in Figure 2.8.
In Figure 2.8, the four strategies that contribute to results occur in a specific
sequence over and over again. Their interaction contributes to the center target:
results, an aggregate of progress over time. The intersection and influence of external issues and/or the environment can be illustrated as well.
Figure 2.8 Nonlinear Theory of Change Model
Strategy 1
Strategy 2
Results
Strategy 3
Strategy 4
In the nonlinear theory of change model, Figure 2.9 represents an alternative

view of the Community Leadership Academy theory of change model shown earlier. Through two strategies identified herecurriculum and experiencesthis
change effort expects to produce more and better community leaders and, ultimately, community development. The curriculum, the experiences, and the participants interact. Chapters 6 and 7 also provide some additional examples of
nonlinear models.
Doing the Right Work

In Chapter 1, we mentioned effectiveness and three critical questions:
The first question was about the right work. This is about attending to making
the strongest, most direct and plausible connection between your strategies and
results. It is about the focus of time, energy, talents, and resources in relation to
Figure 2.9Community Leadership Academy Nonlinear Theory of Change Model
Academy
Leadership
Curriculum
Community
Development
More and Better

Community Leaders
Academy
Leadership
Experiences
your specified success. Eventually, right work is also about detailing those specific
activities that are subsumed by each strategy that is chosen for display in the program logic model. Giving conscious attention to the criterion used in selecting
strategies at this stage, and again later, will identify how implementation can make
a big difference in the likelihood that your program will secure results. The right
work is clarified and confirmed if there is a shared understanding of the problem
you plan to resolve and there is agreement on how it can be accomplished.
Specificity here, on the front end, contributes to the results you and your colleagues
intend to secure. Ambiguity can doom the best-intentioned efforts to failure.
If your end result in a construction project is a great house, then cooking and
sewing probably are not the most relevant strategies. However, planning with welldetailed blueprints as well as appropriate purchasing (e.g., quality lumber) and
contracting should be ripe for your attention. It is surprisingly easy to spend time
on the wrong work. It can be an unconscious or conscious choice. Theory of change
models should display planned results and specify the most relevant and influential
strategies to secure the results. The strategies are determined from a universe of
possibilities. Often, people include strategies (and later, activities) they have always
done or are most familiar with doing. If replication is intentional, then repeating
what has been done before might be appropriate. As time passes and knowledge
changes, however, results may require we use what has been learned about new or
different strategies (and activities) to be more efficient and effective. Remember, a
theory is only as good as its last test.
Tough Questions
Of course, there are many ways to secure a named and intended result. Discarding
strategies/activities that are peripheral, modest contributors or less than optimal in
potency can focus limited resources. Models and their iterations can develop a disciplined way of thinking that contributes to new understandings about what will
generate progress toward results. Once results are specified, the discovery and discussion that should be encouraged during your modeling attends to these two big
questions:
What are the many ways we could resolve this challenge?
Then, what are the most effective and efficient ways to secure results?
Subsequently, as model versions are explored, it is important to inquire further. For example, are we doing something that has a reasonable chance of working? Are we doing something that should be done? Are we clear enough about the
work that we have shared expectations for what it includes and can yield? How
does our model rely on research, practice, theory, and/or literature? We know that
the politics of power and dynamics of resources often preclude these conversations. A predisposition to activities and busyness can overwhelm a disciplined
and interactive process, too. However, the benefits and relative value of getting
things right at the start are considerable. The opportunity cost or waste in missing this step is huge.
IN SUMMARY
Logic models display mental maps people hold about cause and effect. Combined,
theory of change coupled with program logic models are the most potent design
prescription. Theory of change models specify and link strategies with results. Most
change efforts require multiple strategies. Knowledge is a critical input for models
and can include research, practice, and theory. What people believe affects the content and format of models. Improving theory of change models requires multiple
perspectives, unpacking assumptions, shared language, toggling, and the exploration of promising practices.
LEARNING RESOURCES
Reflection
1. What role do assumptions and beliefs play in a theory of change model?
2. How can you test a theory of change model for plausibility? Why bother with
this step?
3. Are there blind spots in the modeling process? If so, what are they?
4. What are the implications of a change model that relies on a hypothesis versus
one based on a claim with proof?
5. What are some ways that theory of change models can be improved and/or
developed?
6. What challenges do complex and highly interactive systems present in a theory
of change model? Where and how do you bound the presentation of a theory of
change model?
Application
1. Have a conversation:
A. Ask colleagues to share their beliefs about parenting (or their mothers or
fathers beliefs) to ensure a happy, confident, successful young adult. From
this conversation, draw a theory of change. What are their most important
strategies? Can you identify their beliefs, values, assumptions? Do they cite
any evidence for their choices? Is research, practice, or theory part of their
explanation? How are their views similar to or different from yours? Do they
have a shared understanding and agreement about parenting with their
spouse (or among their parents)? How does your response to these questions
influence the model?
B. Ask a friend or colleague to share a recipe for marketing a new car model. What
are the most important strategies for ensuring profit? What evidence supports
their choice of strategies? How do assumptions inform their theory of profitability? How does your response to these questions influence the model?
2. Ask several people to list the many ways that improved health might be
described. Why does this outcome/result have different meanings? Could these
differences influence modeling?
3. Find a news article that describes a change effort (in a government, nonprofit, or
private sector). Draw it. Can you detect the efforts underlying theory of change?
How was it informed: based on a claim or a hypothesis?
4. Considering the drawings from Questions 1 and 3, how do choices of strategies
influence the likelihood of achieving your intended results? What changes, if any,
could be made to improve the plausibility of these models?

Texts
Bickman, L. (1987). The functions of program theory. In L. Bickman (Ed.), Using program
theory in evaluation. New directions for program evaluation, 33, 518. San Francisco:
Jossey-Bass.
Bono, E. (1999). Six thinking hats (2nd ed.). New York: Little, Brown & Company.
Chen, H. T. (1994). Theory-driven evaluations. Thousand Oaks, CA: Sage.
Donaldson, S. I. (2007). Program theory-driven evaluation science: Strategies and applications.
Mahwah, NJ: Lawrence Erlbaum.
Porter, M. E. (1995). Competitive advantage: Creating and sustaining superior performance.
New York: Free Press.
Reisman, J., & Gienapp, A. (2004). Theory of change: A practical tool for action, results and
learning. Baltimore: Annie E. Casey Foundation. Retrieved December 7, 2011, from
http://www.aecf.org/upload/PublicationFiles/CC2977K440.pdf
Scheirer, M. A. (1987). Program theory and implementation theory: Implications for evaluators. In L. Bickman (Ed.), Using program theory in evaluation. New directions for program evaluation, 33, 5976. San Francisco: Jossey-Bass.
Weiss, C. H. (1995). Evaluation (2nd ed.). Upper Saddle River, NJ: Prentice Hall.
Weiss, C. H. (1995). Nothing as practical as a good theory. In J. P. Connell, A. C. Kubisch,
L. B. Schorr, & C. H. Weiss. (Eds.). New approaches to evaluating community initiatives: Concepts, methods and contexts (pp. 6592). Washington, DC: Aspen Institute.
Wholey, J. S. (1987). Evaluability assessment: Developing program theory. In L. Bickman
(Ed.), Using program theory in evaluation. New directions for program evaluation, No. 33
(pp. 7792). San Francisco: Jossey-Bass.
Journal Articles
Birckmayer, J. D., & Weiss, C. H. (2000). Theory-based evaluation in practice: What do we
learn? Evaluation Review, 24(8), 4043.
Bolduc, K., Buteau, E., Laughlin, G., Ragin, R., & Ross, J. A. (n.d.). Beyond the rhetoric:
Foundation strategy. Cambridge, MA: Center for Effective Philanthropy. Retrieved
December 7, 2011, from http://www.effectivephilanthropy.org/assets/pdfs/CEP_
BeyondTheRhetoric.pdf

Chen, H. T., & Rossi, P. (1983). Evaluating with sense: the theory-driven approach. Evaluation
Review, 7(3), 283302.
Connell, J. P., & Kubisch, A. (1998). Applying a theory of change approach to the evaluation of
comprehensive community initiatives: Progress, prospects and problems. Washington, DC:
Aspen Institute. Retrieved December 7, 2011 from http://www.dochas.ie/Shared/Files/4/
TOC_fac_guide.pdf
Donaldson, S. I. (2001). Mediator and moderator analysis in program development. In
S. Sussman (Ed.), Handbook of program development for health behavior research and
practice (pp. 470496). Thousand Oaks, CA: Sage.
Fullan, M. (2006). Change theory: A force for school improvement (Seminar Series Paper No. 157).
Victoria, Australia: Centre for Strategic Education. Retrieved December 7, 2011 from
http://www.michaelfullan.ca/Articles_06/06_change_theory.pdf
Kramer, M. R. (2001, May/June). Strategic confusion. Foundation News and Commentary,
42(3),40-46.
Monroe, M. C., Fleming, M. L., Bowman, R. A., Zimmer, J. F., Marcincowski, T., Washburn, J.,
& Mitchell, N. J. (2005). Evaluators as educators: Articulating program theory and
building evaluation capacity. New Directions for Evaluation, 108, 5771.
Rogers, P. J., Petrosino, A., Huebner, T. A., & Hacsi, T. A. (2000). Program theory evaluation:
Practice, promise, and problems. New Directions for Evaluation, 87, 513.
Sridharan, S., Campbell, B., & Zinzow, H. (2006). Developing a stakeholder-driven anticipated
timeline of impact for evaluation of social programs. American Journal of Evaluation,
27(6), 148162.
Internet Resources
Sharpe, M. (2009) Change theory. Muncie, IN: Ball State University. Retrieved December 7,
2011 from http://www.instituteforpr.org/wp-content/uploads/ChangeTheory-1.pdf
For a comprehensive website and online suite of tools sponsored by ActKnowledge

and the Aspen Institute on Community Change, see the following:
Anderson, A. (n.d.). A community builders approach to theory of change: A practical guide to
theory development. New York: Aspen Roundtable on Community Change. Retrieved
December 7, 2011, from http://www.dochas.ie/Shared/Files/4/TOC_fac_guide.pdf
Theory of change. (n.d.). Retrieved December 7, 2011, from http://www.theoryofchange.org
3
Creating Program Logic Models
his chapter identifies the basic elements of a program logic model. Generally, these models have enough detail to support design, planning, management, or evaluation. This chapter describes a program logic model
example and the action steps to create a model with a small group.
LEARNER OBJECTIVES
Describe the relationship between theory of change and program logic
models
Identify basic elements for a program logic model
Create a simple model
Recognize limitations of display
From Theory of Change to Program Models

Theory of change logic models are literally the foundation for program logic models. When well developed, they can ensure intellectual rigor for program logic
models. Figure 3.1 illustrates the relationship of a theory of change model (composed of strategies and results) to the primary elements of a program logic model.
Strategies reflect the resources, activities, and outputs needed to achieve results.
Results reflect the sequence of outcomes over time through impact. Outcomes (for
individuals) are generally progress in changes in awareness, knowledge, skill, or behavior among targeted audiences. There are also outcomes for organizations and systems.
Although a plausible and evidence-based connection can be established, impact is
often well beyond the scope (or feasibility) for the program being modeled. Together,
outcomes, which are closer to the effort, of multiple strategies plus impact (further
away) make up results. While program logic models are often built on a theory of
change, it is also possible to infer a theory of change from a program logic model.
34
Chapter 3Creating Program Logic Models35
Figure 3.1 Relationship of Program and Theory of Change Models
Resources
Do
Get
Strategies
Results
Activities
Outputs
Short-Term
Outcomes
IntermediateTerm
Outcomes
Long-Term
Outcomes
Assumptions Matter
It is important to be aware that specific assumptions are not illustrated in Figure 3.1.
Recall that assumptions are informed by beliefs and knowledge. Too often, program
models are built without the benefit of explicitly naming the assumptions and
underlying theory of change. This omission can help explain why tremendous conflict, even chaos, can erupt during program development, planning, implementation, or assessment. In the absence of explicitly named assumptions, either a clear
theory of change does not exist or people hold multiple and conflicting variations
that reflect their deeply held views about what should/could work and why. This can
lead to diffuse or dilute programs that lack the focus and intensity needed to produce intended results. Because of these implications, omitting this foundation for
your idea, program, or social change effort undermines its potential for success.
As noted previously, conceptualization and learning styles differ from person to
person. Organizational culture can also affect how design, planning, monitoring,
and measuring occur. Given these practical issues, we strongly suggest that both
theory of change and program logic models eventually be created to form the foundation of shared meaning for all aspects of the program. The sequence in which
they are developed certainly should and will reflect the stakeholders preferences.
Key Elements of Program Logic Models

Program logic models display what an existing idea, new program, or focused
change effort might contain from start to finish. The elements in a program logic
model consist of the recipe for a bounded investment of financial and social capital
for a specified result. The level of detail increases so that the relationships shown by
the model illustrate essential linkages needed to make a plan fully operational for
each of the strategy strands identified in the theory of change. The primary elements for each strand of a program logic model include resources, activities, outputs, outcomes, and impact. Figure 3.2 is a template of the elements for most
program logic models.
Impact
Figure 3.2 A Basic Program Logic Model
Resources
Activities
Outputs
Short-Term
Outcomes
IntermediateTerm
Outcomes
Long-Term
Outcomes
Impact
These program logic model elements are defined as follows:

Resources are essential for activities to occur. They can include human, financial,
organizational, community, or systems resources in any combination. They are
used to accomplish named activities. Sometimes resources are called inputs.
Activities are the specific actions that make up the program. They reflect tools,
processes, events, technology, and other devices that are intentional in the program.
Activities are synonymous with interventions deployed to secure the desired
changes or results.
Outputs are what specific activities will produce or create. They can include
descriptions of types, levels, and audiences or targets delivered by the program.
Outputs are often quantified and qualified in some way. They simply characterize
the application of activities with selected audiences.
Outcomes are about changes, often in program participants or organizations, as
a result of the program. They often include specific changes in awareness, knowledge, skill, and behavior. Outcomes are dependent on preceding resources, activities, and outputs. Sometimes outcomes are parsed by time increments into short,
intermediate, and long term. Time spans for outcomes are relative and should be
specified for the idea or project described. However, short term is often 1 through
3 years, intermediate-term outcomes 4 through 6 years. Long-term outcomes might
be achieved in 7 through 10 years. The intervals specified for any given model
would depend on the size and scope of the effort.
For example, a small-scale project such as an adult education typing class in
one location might produce knowledge and skill outcomes in 6 weeks, where
behavioral changes such as use or changes in employment might take somewhat
longer. Alternatively, a program targeting changes in global water quality might
specify changes in the awareness and knowledge of international policymakers
within 1 to 3 years; actual environmental improvements might not occur within
decades. Typically, dividing the project duration into thirds works pretty well as
a starting point. Relying on a literature or other evidence base can help inform
what is feasible.
Being clear about timing and expected results is important. The time span for
outcomes is project specific. Time is one of several important considerations. The
logical sequencing of any given outcome chain matters, too. Think about what will
happen first, then what is likely to happen next. Also keep in mind that the sequence
may or may not be lockstep. Under some conditions, there may be different points
of entry into a sequence. The important thing is to explore the interconnections
and dependencies that do exist among the outcomes and impact you specify.
Impact is the ultimate intended change in an organization, community, or other

system. It carries an implication about time. It varies in its relative timing to the
actual program or change effort. Sometimes impact occurs at the end of the program, but more frequently, the impact sought is much more distant. For some
efforts, this may mean impact can be cited in 7 through 10 years or more. This can
have important implications, as it is well beyond the funding cycle for many typical
grant-funded programs or the patience of many managers or politicians. The logic
model is one way to show how the work you can do within these constraints may
contribute to a larger, grander impact.
The planned work of a program logic model includes resources, activities, and
outputs. These are the essential elements that are used to secure results or make
change happen. The intended results include what the program produces: outcomes and impact.
Nonlinear Program Logic Models

Just as in theory of change models, very few ideas, programs, or projects actually
occur in a linear progression. Purposely, to aid learning, we simplified the display
of elements as a straight sequence. Reality suggests cycles, iterations (additional
attempts), and interactions are very common. This more organic development is
shown in Figure 3.3.
In this circular display, there is no specific starting point. Although the logic
model elements are constant, the work of design, planning, managing, or evaluating might begin with any element. In addition, this view shows how cycles of the
same activity might occur over time. Keep in mind that the illustration groups
activities together. A more detailed view could be staggering to portray. Sometimes,
capturing reality in a display impedes communication.
Hidden Assumptions and Dose

A program logic model displays the elements that are most critical to establishing
and operating a program or social change effort. It specifies the activities and their
often interdependent relationship as well as what they are expected to generate.
Program logic models do not necessarily include assumptions, but they rely on
them. They offer a view of the map that can inform action planning and, later,
implementation. Program logic models can also define the dose (e.g., number,
type, and duration of activities), quantify and describe the effects and benefits of
the program for a given dose and the ultimate change expected. Dose is an important concept in effectiveness. A dilute dose can have the same impact as none at all.
For example, if your intended result is a large voter turnout in an election, a classified ad is not an adequate communication strategy. A comprehensive media plan
coupled with free transportation to the voting booths has greater chances of success. So, it is important to design a program with enough of the right activities to
secure the outcome you intend.
Figure 3.3 Nonlinear Logic Model
Activities
Activities
Activities
Resources
Outputs
Outputs
Outputs
Resources Resources
Short-Term
Outcomes
Impact
Short-Term
Outcomes
Impact
Short-Term
Outcomes
Impact
Intermediate
-Term
Outcomes
Short-Term
Outcomes
Intermediate
-Term
Outcomes
Short-Term
Outcomes
Short-Term
Outcomes
IntermediateTerm
Outcomes
Building a Program Logic Model

Program Logic Model Example
An example of a simple program logic model for securing improved health is displayed in Figure 3.4. Read from left to right, this program model suggests that if we
recruit and retain participants and provide exercise, nutrition, and stress reduction,
then we will secure improved health. Note the development of detail connecting
strategies to results in this model compared to the theory of change (see Figure 2.3).
The program logic model provides detail for the theory of change by explicating the
elements from a basic logic model for each strategy strand. In a program model, the
details relative to resources, activities, and other elements are named.
Although still an overview and incomplete, this illustration provides a more
detailed view of what this health improvement program wants to do, plans to measure, and hopes to achieve. Beginning on the left with resources, this model includes
funds, facility, faculty, and coaches, as well as eligible and willing participants, among
its requisite inputs. To keep it simple, the strategies contain implied clusters of
39
E
Eigible and
willing
participants
Coaches
Faculty
Facility
Funds
Resources
Retention and
recruitment
activities
Stressreduction
activities
Nutrition
activities
Exercise
activities
Activities
Do
Coaching
tools and logs
Messages
and media
Participant
data
Curricula
and staff
Outputs
Figure 3.4 Improved-Health Program Logic Model
Awareness
Repeats for each strategy
Knowledge
Short-Term
Outcomes
Motivation
Adherence
Skill
IntermediateTerm Outcomes
Get
Retention
Relaxation
Flexibility
Fat/Calories
Nutrients
Endurance
Strength
Long-Term
Outcomes
Improved
health
Impact
activities in this illustration. The specific activities that contribute to outputs are not
named. Outputs from the intervention strategies and associated activities (exercise,
nutrition, stress reduction) could be numerous. For this illustration, we show only
the overarching categories of information that could be considered. Each category
would be repeated for each of the strands. These would include details about the
scope, sequence, and quality of the curriculum; staffing qualifications; and information about participants and their participation. Activities inside these strategy
strands contribute to changes in knowledge, skill, and adherence. Eventually, they
can contribute to increases in strength, endurance, nutrients, flexibility, and relaxation. Concurrently, over time, these same strategies also yield reduced fat/calories.
The retention and recruitment strategy strand also generates some outputs and
outcomes. Aggregated, activities within this strategy secure and keep participants in
the program. Note that this model uses arrows to show relationships. Sometimes
they reflect a cluster (indicating synergies) rather than just one:one relationships.
As is typical of many programs, several strategies may be shown as contributing
collectively to outcomes rather than each strategy making its individual contribution to distinct outcomes in isolation. Collectively, the long-term outcomes generate improved health, which could be measured in a variety of ways (e.g., blood
pressure, blood lipid and sugar profiles, weight).
In contrast to the big-picture view that theory of change models offer, program
logic models provide a closer, more detailed picture of operations. This view of the
program provides adequate detail to then create work plans. Program models can
provide a reliable outline for work plans that are used to implement and manage a
program or larger change effort. Just like theory of change models, program models
are often logicalbut here, feasibility, given limited time and resources, is the
appropriate standard for assessing their value. A common question about program
logic models focuses on their level of detail. Essentially, the level of detail in program logic models should be determined by their intended use and users. Although
somewhat situational, program logic models build out strategies to activities.
Sometimes they can even get to the fine detail of tasks, although more often that is
described in an operations or action plan.
From Strategy to Activities

Some program logic models can be extremely complex, but the steps to create them
are generally the same as for more simple efforts (see Figure 3.2). Large-scale programs
or multiyear change efforts (sometimes called initiatives) often are composed of
many strategies aimed at target audiences across many sites over considerable time.
Program logic models usually do not display underlying beliefs or assumptions.
They are nevertheless important elements in the conscious exploration of multiple
target audiences. Sometimes programs or change efforts are implemented in a cascade with some overlap in time, which requires a particular sequence of strategies
and associated activities. When this is the circumstance, it can be helpful to focus
on a function, a given strategy, or one partners designated work. The task is often
simplified by thinking about a single aspect and then connecting it back to the
whole with some of the inherent complexity reduced. Ultimately, program execution relies on integrated actionbut the work that precedes it may require focused
developmental attention on smaller parts.
Using the health improvement program example, Figure 3.5 provides an orientation to how the exercise strategy strand might be reduced to activities. It breaks
the strategy into greater detail for the purposes of selection and design.
In Figure 3.5, it becomes evident that exercise as a strategy is made up of several
key activities. They include physical exercise (strength and endurance), education,
and assessment. Together, all of these activities represent a comprehensive strategy,
exercise, that is just one means to improved health. Recall that the whole theory of
change for this example also includes stress reduction, nutrition, and retention. It
is the combination of strategies reflected in the whole program that is most likely
to secure results. Each strand of a comprehensive program logic model needs to
illustrate the contribution of each strategy as well as the interdependence.
As you specify the activities content of your strategy, you are naming more precisely
what makes up the given strategy. Later, the whole model is tested for feasibilityboth
practically before implementation and literally when the program is evaluated.
In Figure 3.6, we provide a view with greater detail for only the exercise strategy.
In this illustration, we show the detail of activities within the exercise strategy. It
also suggests the many decisions hidden in program design and planning. In choosing activities, it is critical that the relationship among strategies and activities is
intentional. The strategies and the cluster of appropriate activities should also be
chosen with reference to a target audience. Remember, logic models use ifthen
sequences from left to right in the columns and among the features as you read
from left to right.
Figure 3.5 One Strategy With Multiple Activities

Do
Get
Strength
Activities
Endurance
Activities
Results
Exercise
Education
Fitness
Assessment
Exercise Strategy
42
Eligible and willing

participants
Coaches
Faculty
Facility
Funds
Resources
Implement
Exercise Strategy
Collect final
measurements
16-week
program
Participant
orientation
Collect baseline
measurements
Activities
Figure 3.6 Exercise Strategy Detail
Participant Data
and Exercise Logs
Participant logs
Baseline data
# and types of
participants
Outputs
Skill with
endurance
equipment
Skill with
strength
equipment
Increased
Adherence
Satisfaction and
enjoyment
Time (length/
session)
Intensity
(exertion/
session)
Frequency
(sessions/week)
Increased Knowledge and Skill
Understands
personal
exercise
prescription
General fitness
knowledge
Short-Term
Outcomes
Increased
endurance
Increased
strength
Long-Term
Outcomes
Improved
Health
Impact
Action Steps for a Program Logic Model

The practical construction of a program logic model often begins with one or more
information sources (e.g., research, interviews, documents). We recommend that
people begin both theory of change and program logic models with the named
ends. People are most clear about their intended results (outcomes and impact).
Our experience is that you do know what you want to accomplish. The results
sought reflect both the impact intended and the outcomes over time. Next, name
the changes or outcomes that will be part of your progress toward impact.
Unpacking this sequence is important because it makes it easier to see the strength
of the connection between what you do and what you can get. We suggest tackling
the activities required to achieve the outcomes you have specified in your third step.
Activities are about how intended changes will occur. Fourth, resources/inputs
become the essential ingredients of activities. And finally, outputs reflect the information needed to verify that activities named earlier in the process reach the right
audiences and are of the quality and quantity needed to produce results. So, the
steps to draft a program logic model are ordered in this way:
1. Identify the results that one or more strategies will ultimately generate.
2. Describe the stepwise series of outcomes (or changes) that will show progress
toward impact.
3. Name all the activities needed to generate the outcomes (for each strategy).
4. Define the resources/inputs that link directly to and will supply the activities.
5. Identify the outputs that reflect the accomplishment of activities.
Figure 3.7 illustrates these action steps and their sequence.
Creating Your Program Logic Model

The format of a logic model helps organize information in a useful way. Think of
an idea, project, or program you manage now or want to create and its results. For
each strategy, brainstorm elements that might be cited in short-term outcomes first
but are clearly linked to your intended results. Do the same for resources, activities,
and outputs. It is important to make choices about the outcomes that are feasible
with your limited resources. This is discussed in greater detail in the next chapter.
With some experience, you will begin to recognize commonly used strategies that
reflect knowledge from your field or discipline. For example, marketing/communications, recruitment, retention, professional development or education, advocacy,
and policy are strategies often found in program models. Examples of activities
under a marketing/communications strategy could include preparing a database of
target markets, generating news releases, creating and sending a newsletter, establishing a website, and distributing public service announcements. We suggest you tackle
one strategy at a time. Aim to define the same level of detail for each strategy.
Selected examples of archetypes, or tested recipes, are described in Chapter 7.
Figure 3.7 Steps in Creating a Program Logic Model

Do
Get
Strategies
Results
Step1
Resources
Activities
Step4
Outputs
Short-Term
Outcomes
Step5
Step3
IntermediateTerm
Outcomes
Long-Term
Outcomes
Impact
Step2
Guiding Group Process

We think the best method for generating a program logic model engages a small
group, especially if the members are stakeholders in the idea or program it represents. Stakeholders are situational, but they generally are those who have an interest
in or are likely to benefit from the program. Logic modeling often includes funders,
program staff, and program participants. Intentionally including stakeholders supports best contributions as well as some subsequent benefits relative to implementation. The facilitation of modeling requires some advance planning and a
commitment to both discipline and quality during the process.
If youve already constructed a theory of change, use it to catalyze the creation
of the program logic model. If not, defining shared understanding for specified
results gets your group process effort started. It is important to note that models
may need to be updated to respond to the dynamics of an external environment
(context). They also reflect living systems that are not mechanistic but changing.
For these two reasons (and others), it is necessary to expect models will be revised.
In association with some public specification of time, impact and outcomes can be
explored and selected. This can be accomplished a number of ways.
We have had success in using the action steps noted, particularly when each participant contributed to brainstorming the model elements by nominating contributions on sticky notes. This quickly generates a large number of possibilities for each
element. Redundancies should be noted and celebrated as commonly held. Then, the
group can sort them: those that must be kept, that could be kept, and those that are
not relevant. Once the results are named, then it is possible to compose content for
the other elements. In this disciplined sequence, each stakeholder contributes to the
whole, and each contribution has the benefit of an internal test relative to design.
There are several variations on this approach. From a group, you could invite
individuals or pairs to generate models in the sequence shown previously and then
reconcile the variations. This approach helps avoid groupthink but requires
strong process facilitation with content knowledge. A generic model or template for
a given program may be available. With some advance planning, its possible to
identify one of these archetypes and introduce it to your group. Then, the content
adaptations can focus on improving it so that the content is relevant to your purposes, conditions, and planned results. Chapter 7 includes examples of models that
can be used in this way. Sometimes a call for proposals or funding opportunity will
articulate the range of acceptable content in a model.
Regardless of the process, strategic decisions about activities and the relationships between elements should be made from among all the content generated. It is
important to consider criteria for choices that reflect context, target audience(s),
research, practice, literature, and program benchmarking, as well as resource
parameters. It can be very helpful to have draft models critically reviewed in a
mark up. This is described in the next chapter as a quality assurance process.
We often use Microsoft Visio to construct our models, but many other applications such as Word and PowerPoint have drawing options. These as well as
Inspiration software are all readily available. The Supplemental Readings list at the
end of the chapter identifies some examples of other free and commercial software
applications. Take care in using technology for model creation, because it can
exclude valuable participation.
IN SUMMARY
High-quality program logic models depend on the evidence base found in their
parallel but simpler theory of change models. Program logic models display several
important elements: resources; activities; outputs; short-, intermediate-, and longterm outcomes; and impact. To create a program logic model, start with the
intended results: outcomes and impact. Then, activities (which are consistent with
strategies in the theory of change model) are selected. Next, resources and outputs
are cited. We believe creating models with deep participation of stakeholders
improves their quality and encourages their use.
LEARNING RESOURCES
Reflection
1. What are the implications of a program logic model built without a specific
theory of change?
2. Think of a successful business and its product or service. What is the underlying
program logic that shows the explanations for profitability?
3. Feasibility relies on several aspects. Can you name some?
4. What are strengths and limitations of a linear or a nonlinear display? Would

individuals from different fields (and their relevant cultures) answer similarly or
differently? Why?
5. Why is being specific about results important?
Application
Specify the result of a shared program, project, or idea. Draw a theory of change
model for the program, project, or idea. Then, attempt a program logic model.
Using sticky notes or pieces of paper, brainstorm the outcomes that need to happen
to secure the result. Organize them into short, intermediate, and long term. Pick
one short-term outcome. Brainstorm what activities are critical to that outcome.
Organize the activities relative to a single or multiple strategies. For given strategies
and their activities, name the resources needed. From the activities, cite what outputs are possible. Organize these elements as one model.

Texts
Frechtling, J. (2007). Logic modeling methods in program evaluation. San Francisco: Jossey-Bass.
Green, E. L. (2005). Reinventing logic models: A stakeholder-driven group approach.
Unpublished doctoral dissertation, University of Cincinnati, OH.
Mayeske, G. W. (1994). Life cycle program management and evaluation: An heuristic approach.
Washington, DC: United States Department of Agriculture.
Stringer, E. T. (2007). Action research (3rd ed.). Thousand Oaks, CA: Sage.
United Way of America. (1996). Measuring program outcomes: A practical approach.
Alexandria, VA: Author.
Westley, F., Zimmerman, B., & Patton, M. Q. (2007). Getting to maybe: How the world is
changed. Toronto: Vintage Canada.
Wong-Rieger, D., & David, L. (1996). A hands-on guide to planning and evaluation. Ottawa:
Canadian Hemophilia Society.
Journal Articles
Cooksy, L. J., Gill, P., & Kelly, P. A. (2001). The program logic model as an integrative framework for a multi-method evaluation. Evaluation and Program Planning, 24(2), 119128.
McLaughlin, J. A. (1999). Logic models: A tool for telling your programs performance story.
Evaluation and Program Planning, 22(1), 6572.
Millar, A., Simeone, R. S., & Carnevale, J. T. (2001). Logic models: A systems tool for performance management. Evaluation and Program Planning, 24(1), 7381.
Porteous, N. C., Sheldrick, B. J., & Stewart, P. J. (2002). Introducing program teams to logic
models: Facilitating the learning process. Canadian Journal of Evaluation, 17(3), 113141.
Renger, R., & Titcomb, A. (2002). A three-step approach to teaching logic modeling.
American Journal of Evaluation, 23(4), 493503.
Rush, B., & Ogborne, A. (1991). Program logic models: Expanding their role and structures for
program planning and evaluation. Canadian Journal of Program Evaluation, 6, 95106.
Internet Resources
For comprehensive bibliographies and links to additional resources, see
Logic model resources. (n.d.). Atlanta, GA: The Evaluation Working Group of the Centers for
Disease Control and Prevention. Retrieved December 8, 2011, from http://www.cdc
.gov/eval/resources/index.htm#logicmodels
Jung, B. C. (19992012). Evaluation resources on the Internet. Retrieved December 8, 2011,
from http://www.bettycjung.net/Evaluation.htm
University of Wisconsin Extension. (n.d.). Logic model bibliography. Retrieved December 8,
2011, from http://www.uwex.edu/ces/pdande/evaluation/evallogicbiblio.html
For logic model development toolkits, see

Administration for Children and Families. (2006). Instructional guideCreating and using
the logic model for performance management. Retrieved December 7, 2011, from http://
www.acf.hhs.gov/programs/cse/grants/resources/logic_model/
Bureau of Justice Assistance. (n.d.). Developing and working with logic models. Washington,
DC: Author. http://www.ojp.usdoj.gov/BJA/evaluation/guide/pe4.htm
Centre for Development Innovation (2010). Logical framework approach. Wageningen, the
Netherlands: Author. Retrieved April 18, 2012 from http://portals.wdi.wur.nl/ppme/
index.php?Logical_Framework_Approach
Community toolbox: Developing a theory of change or logic model. (n.d.). Lawrence: University
of Kansas. Retrieved December 8 2011, from http://ctb.ku.edu/en/dothework/tools_
tk_4.aspx
Enhancing program performance with logic models. (n.d.). Madison: University of Wisconsin
Cooperative Extension Service. Retrieved December 8, 2011, from http://www.uwex
.edu/ces/lmcourse/
Graig, E. (n.d.). Logic models: A tutorial. Useable Knowledge, LLC. Retrieved December 8,
2011, from http://www.usablellc.net/resources/logic-model-tutorial
Jackson, B. (1997). Designing projects and project evaluations using the logical framework
approach. Geneva: The World Conservation Union. Retrieved April 18, 2012, from
http://cmsdata.iucn.org/downloads/logframepaper3.pdf
Point K: Practical tools for planning, evaluation and actionLogic model builder. (n.d.).
Washington, DC: Innovation Network. Retrieved December 8, 2011, from http://www
.innonet.org/index.php?section_id=64&content_id=185
For logic model development software, see

Duigan, P. (2011). Do View 3.0: Visual results and outcome planning. Retrieved December 8,
2011, from http://doview.com/
Milstein, B., & Chapel, T. (2011). Tool: Peer discussion of software programs to help you model.
Retrieved December 8, 2011, from http://ctb.ku.edu/en/tablecontents/sub_section_
tools_1877.aspx
ImpactMap. (2011). ImpactMap: Navigate social change. Retrieved December 8, 2011, from
http://www.impactmap.org/
Results Technologies, Inc. (n.d.) Results online: Outcome tracking software and outcomesbased applications. Retrieved December 8, 2011, from http://www.resultstechnologies
.net/index.php?process=solutions
4
Modeling
Improving Program Logic Models
his chapter focuses on improving models through simple processes that

test feasibility. With careful and deliberate review, models for an idea,
program, or project can change and mature in their quality. Logic models
that are accurate and realistic representations of what you do and will get can
increase the likelihood of effectiveness.
LEARNER OBJECTIVES
Apply simple review and improvement steps to models
Identify common errors in program logic models
Recognize the value of multiple versions of models
Recognize contributors to model quality
The process of modeling supports better thinking about a given idea or effort. It
can establish routines wherein alternative possibilities are considered and explored.
Modeling may be an important antidote to snap judgments. We see modeling as an
essential step that has tremendous potential to position a project, program, or initiative for greater effectiveness. It generally involves several versions or attempts at
models that result from a critical review of the information displayed.
Modeling and Effectiveness

Once a shared model is created, then those who created the model and others who
are external to the program should spend time on its critique and revision. Great
plans and programs come from great models. Great models require several rounds of
48
Chapter 4Modeling49
revision. Modeling is the most important content in this book because critical thinking is what refines the content of the model. Modeling is a means to illustrate where
you are headed, make better decisions about getting there, and get closer to the right
work. We think it is important to test both plausibility and feasibility through modeling. In Chapter 2, we stated that plausibility is the most important criterion for a
theory of change model. The process of toggling between combinations of strategies
and results to secure an evidence-based model tests and verifies plausibility. We provide techniques that address the aspect of feasibility once plausibility has been established or confirmed. These criteria reflect the widely used Program Evaluation
Standards (The Joint Committee on Standards for Educational Evaluation, 1994) that
include utility, feasibility, propriety, and accuracy. Given the strong relationship
between programming (strategy) and evaluation (results), we draw upon what constitutes sound evaluation practice to inform logic model development.
This chapter identifies some of the common context challenges that influence
models and mitigate their potential. Some result in flaws embedded in the models
and others occur in the modeling process. Our list includes blind spots, myths,
logic, and scale. We also name an overarching concern: culture. It can influence
all the preceding items and more. Next, we identify some quality review techniques
for program logic models. Through this sequence, we name the problems often
associated with models and attempt to offer some remedies.
Context Challenges
Common Pitfalls: Blind Spots and Myths
Generating high-quality models through modeling is not easy, but both the products and the process do offer significant benefits. Some discomfort among participants during modeling can be an indicator that the process is useful. This
discomfort probably reflects the twists, turns, flips, and angles of new or different
thinking. It is important to be aware of blind spots (which we all have).
Individuals have blind spots and small groups can have collective ones they protect and promote through interaction among participants. Blind spots are simply
unintentional omissions in our thinking or commissions of error that happen
because of habit (e.g., snap judgments, confidence, experience enables them).
Eventually, these blind spots show up in our models because models represent
how we think.
Prevailing myths are another pitfall to guard against. Common examples of
myths include access equals use, knowledge equals action, activities equals outcomes, and send equals communication. These myths end up as embedded
flaws in models because they are not precisely named and subsequently discarded.
Almost every organization, department, or unit perpetuates some myths. We think
its how people defend and rationalize bad habits. These habits are often a way of
coping with the organization, a boss, a funder, or another entity. Avoiding the pitfalls of blind spots and myths through disciplined and intentional discovery can
contribute to model quality and eventually to program effectiveness.
Logic, Scale, and Specificity

Logic models often display a logical sequence or chain of events, but this is not the
same as feasibility. The content and order of activities as well as other elements in
a model can make sense, but scale is often an obstacle. Scale is about the relative size
of a given effort. In particular, we mean the relationship between the results sought
and the quality, volume, frequency, and other characteristics of the intervention or
doing that is described.
For example, a logic model could name a just, sustainable community as its
impact. Strategies to secure that intention might include some nonprofit capacity
development and leadership development. It is logical to think these strategies
might eventually contribute to a just, sustainable community. But these strategies
alone are insufficient to achieve the desired impact. There are other essential strategies necessary for this recipe to yield justice and sustainability. This theory of
change model is not plausible.
Subsequently, as activities are selected (in association with the capacity and leadership development strategies), they are far too dilute to have influence. A program,
if deployed as described, would be unable to make progress against such an ambitious impact. This is particularly true given the 3 to 5 years allotted to most grantfunded programs to secure results (outcomes and impact). Engaging a few leaders
and nonprofit organizations is inadequate to the huge ambition of a just, sustainable community given the enormous number of variables that can intervene. The
program logic is not feasible.
If the scale of effort does not match short-term outcomes, then its relative paucity
can overwhelm chances of success. Given limitations, it is best to revise the model and
indicate something that is appropriate. The first error in plausibility is compounded by
a second error of feasibility. This example illustrates why logic, alone, is not an adequate
quality test for models. It is important to note that sometimes theory of change models
are plausible although the program logic model is not feasible. We believe success is
more likely when plausibility and feasibility are evident in both models.
Given limited human and financial resources along with a time frame, its important to specify your program or change initiative carefully. Making these choices is
difficult but very important. We applaud audacious and substantial change agendas.
However, success is more likely secured if your work is discrete and seeks near-term
outcomes that, if accomplished, will contribute to results. For example, consider
ending childhood obesity. This is an important and big result that reflects a complex
systems challenge. Rather than taking on all of this substantial responsibility, perhaps the program effort that reflects your niche is focused on a single aspect such as
more physically active middle school children. A reduction in childhood obesity will
be more likely if you describe a selected population of children, their gender, their
school district, and details of the activities that will constitute your project. These
choices focus your results and intended outcomes. They also help inform the best
prescription to deliver the dose needed to secure success. We know from experience
that specificity is linked to success. Specificity does not preclude scale or ambition;
it can complement those issues. We suggest that efforts with limited resources and/
or relatively short time horizons be focused on outcomes that are earlier in the
Chapter 4Modeling51
sequence on the path to the vision of the future you want to create. These more
proximal outcomes can help define a contribution to the desired results. The
strength of logic models lies in articulating the contribution to desired impact.
Probability statistics underscore our point: With fewer variables, the likelihood
of the optimal combination increases. As the variables increase in number, the likelihood of success declines. We advocate for smaller bites. It simply means picking
which results are likely and moving backward from those results while employing a
feasibility lens to determine the activities needed to develop plans, programs, or
related efforts.
Politics, Persuasion, and Perception

Sometimes errors in scale are authentic oversights. Other times, the politics of context
are the origin. It is quite common for agendas that include marketing or positioning
to supersede whats feasible. Models can be very persuasive. Something that is simple
and clearly described in a graphic format is compelling. Models often help to persuade others because they clarify meaning and provide a logical organizer consistent
with how people think, from actions to consequence. Depending on intended use,
beware of models that are vague, general, and lacking in focus. Grandiose hyperbole
in a model can indicate it is a marketing tool. Because logic models are socially constructed, it is important to beware that they may represent a shared viewbut not
necessarily factual truth or reality. The model can display the truth or perception of
those who created itor a combination of both. It is an error to assume every model
is an evidenced-based prescription. The rigor of models in this matter varies considerably and reflects the context of their purpose and construction.
We indicated early in the book that models are co-created to ensure shared
meaning. Exchange with and among participants during the model creation and
modeling helps to generate common language. This process also ensures a display
that is both intentional and commonly understood. A model that conveys consistent messages among those who created it may not have the same meaning to others because they missed the dialogue, exchange, and iterative processes. An
important challenge in this group work is avoiding simple compromise. Compromise
may satisfy the intragroup dynamics and politics, but it is not the same as strategy.
Models need to reflect strategic thinking and choices. Keeping the focus on strategic
choices to secure the intended end results is crucial to model quality and ultimately
to program effectiveness.
Besides feasibility, we think program logic models should aim at utility. For
this reason alone, models will look different depending on the co-creators and
the intended model use. This situational aspect applies to the level of detail in
a given model, too. Program logic models can be complete but vary considerably in their content. What one group determines as a useful level of detail may
be inadequate for another. Others may choose to model certain aspects of their
program with different level of detail. Chapters 6 and 7 display some of this
variance in format and use in case profiles and examples. The social construction of models means they will inevitably reflect assumptions, expectations,
use, and other context features.
A Learning Culture and External Review

How models are created influences their format and content. Weve already mentioned the composition of the review group. In addition, the culture (norms, values,
beliefs) of those gathered to create and/or improve the model is very important.
The best culture for improving models is one focused on learning from experience.
This implies lots of variation and versions of models that will help secure the one
(eventually) used for plans and execution. Mistakes are critical to learning, and its
important that mistakes are welcomed. People often need the subjective processes
of critical review to identify errors and eventually secure remedies. Nobody draws
a perfect model on his or her initial attempt. We believe a perfect model is not
attainable regardless of the number of attempts. Any model simply represents a
snapshot of the best thinking at a given point in time.
Processes to test logic models can retain and engage the original stakeholders
who first drafted it. However, models improve most dramatically when introduced
to colleagues or external sources that are not directly invested in the models elements or their implementation. In this way, disinvested parties can offer objective
critiques, which often uncover blind spots and identify weaknesses, flaws, leaps of
faith, ambiguities, cockamamie, and fiction.
When inviting an external review, its important to be aware that people may see
and read models differently than those who created them. An external review can
help provide great insight to both improved model content (e.g., resources, activities, outputs) and relative display (by virtue of placement and arrows that indicate
relationship among content). Moving from general to specific, ask those doing an
external review a series of questions. For example, initial questions might discover
more about how the parts fit or do not fit together, what are other ways the same
results could be achieved, and if the resources are adequate. The review might also
inquire about the assumptions embedded in the model and ask about the underlying theory of change it has been built on.
Quality Techniques
Modeling
Most ideas, projects, or programs can be characterized in their life to include four
simple stages: design, implementation, evaluation and adaptation. We suggest that
modeling is most useful when done in the creation stage and during evaluation, but
models can be used at any stage for different purposes. Getting things right at the start
can be very important to ultimate results and is a key influence to subsequent stages.
Modeling can be thought of as a review process that occurs prior to implementation
or execution. It is done to improve thinking and the models that reflect thinking.
Time and effort spent in this work can have enormous return on investment through
the influence on the program itself. The steps in modeling are draw and test. This
construct is displayed in Figure 4.1.
As a program, project, or idea is created, we suggest it gets drawn as a model. The
draw step is satisfied when all elements of a program model (see Chapter 3) are present.
Chapter 4Modeling53
Figure 4.1 Relationship of Modeling to Program Life Cycle

Modeling
Design
Draw
Test
To IMPROVE
Implement
To ADAPT
Evaluate
Completion of this step means resources, activities, outputs, outcomes, and impact are
named. This provides an opportunity to graphically display the thinking behind how
the ideas framed in the theory of change will be implemented as a program. Many
efforts with logic models quit at this point. However, through modeling, you can move
quickly to dialogue to process the content and the tangles. Tangles represent areas of
confusion or where some in your group think a choice is wrong, confusing, or poorly
specified. Modeling is the process that guides model improvement.
In this chapter, we begin to name how and what can test (or explore) model
quality. We believe this testing can help improve models. The subsequent versions
of models that result from literal and figurative tests are products of modeling. This
process can yield benefits to the specific idea or project as well as the individuals
engaged as a work group. It is important to be aware that many external issues
influence modeling. We describe some of those issues, but our list is not exhaustive.
Testing Model Quality: SMART and FIT

In a conscious testing effort, one way to explore the quality of a model is to apply
SMART principles to it. SMART is a mnemonic used since the early 1980s to set
objectives:
Specific: what to do is clear enough to act on and is connected to outcomes.
Measurable: the content can be both quantified and qualified.
Action oriented: the content is selected to provoke change in awareness, knowledge, skill, and/or behavior.
Realistic: the content is both plausible and feasible.
Timed: the content specifies a duration and illustrates the time-dependent
sequence of outcomes for progress toward results.
If just a few people apply SMART principles to the outcomes portion of a program logic model, it will probably surface several flaws and, at the very least, some
good questions. It may also raise concerns that are unique reflections of the training, experience, perspectives, and style among those contributing to the review. We
suggest that a SMART analysis be applied to both the parts and the whole. In other
words, it can be used to examine selected individual strategy strands and activities,
as well as the model as a comprehensive effort.
A second way to explore or test the quality of a model is to consider FIT. FIT is
an acronym that was first coined by the American College of Sports Medicine as
FITT (frequency, intensity, time, and type). Although originally developed to prescribe exercise, we have modified it for use in our practice to quantify program
delivery. It stands for
Frequency of occurrence,
Intensity or strength of the given effort,
Targeted at a specified market or audience.
FIT principles offer an easy way to consider the process elements of the program,
project, or idea you display in your model. They translate to important questions
about the adequacy of dose relative to intended results. The frequency reference
implies this question: Does the project (idea or program) repeat, occur with appropriate volume, or happen often enough so that results are likely? The intensity reference implies this question: Does the project (idea or program) have enough depth
or concentration that results are likely? The target reference implies this question:
Does the project (idea or program) aim at an appropriate and specific audience or
market segment?
In effect, FIT can remedy treatment trickle, one thing known about resolution of problems or generating a change in awareness, knowledge, skill, or behavior: The treatment (or intervention) needs to be both appropriate (the right
choice) and adequate (the right dose). The wrong choice wont help get the results
you want. The right choice has more chance of securing results, but there needs
to be enough of it (volume) focused on appropriate targets to ensure results.
Treatment trickle means that a problem or challenge is underdosed or an effort
that is too broad or diluted to have impact. Feasibility is enhanced when we tackle
a desired change with the right strategies as well as appropriate depth of intervention. Treatment trickle is common because people prefer to give everything or
everyone some rather than make hard choices about who or what gets enough!
It is often a reflection of limited resources, vague goals, or organizations with an
activities focus (see Figure 4.2).
A Mark Up
In Figure 4.3, we revisit the logic model introduced in Chapter 1 for the Community
Leadership Academy program. We suggest a technique thats often used in the
Chapter 4Modeling55
Figure 4.2 Modeling as Quality Review
Context
identifies
challenges
FIT
tests activities
SMART
tests outcomes
Better model
Model
Modeling
Better program
legislative process as working drafts of language for a regulation or authorization

are generated. It is called a mark up. We adapt the legislative mark up to raise
important questions about model quality by applying SMART and FIT principles.
Other elements, including context and technique questions, can also be used. This
discovery is aimed at changing the model in constructive ways that reflect evidence,
strategic choices, and better thinking. Using a disciplined approach to modeling
captures an important opportunity for models to mature in quality.
Frequently, those who participate in developing a model judge how well it
addresses the SMART and FIT criteria differently from how it is assessed by an
external audience. This is because external reviewers may lack the context and
shared meaning that emerge during development. Not all the meaning in a model
is explicitly captured, displayed, or explained. The degree to which a model is
SMART or FIT is highly contextual and situational. Ratings will depend on who the
model is developed for, the models purpose, and the level of detail the model is
intended to communicate. If it is important that your model be self-explanatory
under all conditions and/or viewed as credible to a wide variety of external audiences, then consider supplemental narrative. Remember, the model is a summary;
it is usually not intended to tell the whole story.
In the Figure 4.3, questions generated from the application of SMART, FIT, and
other process principles are placed on the model for discussion and resolution. The
provoking dialogue that will likely occur will in turn result in changes to the model
content based on new understandings, evidence, and other contributions. Your
review of this model may raise different questions about its feasibility.
Figure 4.4 uses the same mark up process to identify questions and concerns
about content in the health model. As a constructive effort to critique and refine,
quality testing is an important but often overlooked step.
In both of these mark ups, FIT questions aim primarily at the quality of process or
intervention dose. SMART questions are focused on the feasibility of outcomes. It is
easy to see that as the model increases in complexity, the challenges to success increase,
too. In a comprehensive effort, more pieces or parts must align and work together to
56
Content
Faculty
Participant
satisfaction
Completion
rate
Does it make sense that the activities identified will contribute to

the outcomes specified? Is the program, as illustrated, feasible?
FIT
How often does the academy meet? How long are the sessions? Over
what period? Dose is important and should be included in the outputs.
What leadership and/or community development theories inform

the choices about curriculum and experiences?
Leadership
Experiences
Processes
Participant
description
Outputs
Context
This model emphasizes the academy perspective. To what extent
was the community involved in developing it?
Sponsors ($)
Marketing/
communication
campaign
Host and
facility
Leadership
Curriculum
Curriculum
and materials
Participants
Activities
Resources
Graduates use
knowledge and
skills obtained
through the
program to
strengthen
the community
Outcomes
Community
development
Impact
What about impact? How far out is this?
What level of budget is needed/available to support the achievement

of these outcomes? How long is reasonable to expect that it will take
to see progress toward outcomes?
SMART
The short- and intermediate-/long-term outcomes are specific,
measurable, and action oriented.The impact is not.
Who are the participants targeted by this program? How important is

it to monitor recruitment and promotion activities? Neither is shown.
Better Leaders
Increased
community
awareness and
action bias
New leadership
attitudes,
knowledge, skills,
and behaviors
Short Term
Figure 4.3 Community Leadership Academy Program Logic Model Mark Up
57
Retention and
recruitment
activities
Stressreduction
activities
Nutrition
activities
Coaching
tools and logs
Messages
and media
Participant
data
Curricula
and staff
Outputs
Knowledge
Short-Term
Outcomes
Awareness
FIT
How are the frequency and intensity specified? Is there sufficient evidence that the program
dose will be enough to achieve outcomes? Do we assume that Participant Data will include
this information? Will baseline health data be collected?
Does it make sense that the activities identified will contribute to the outcomes specified?
Are these activities enough to change awareness, knowledge, skill, AND behavior? Is the
program, as illustrated, feasible?
What health-improvement and behavioral change theories inform the choices about
curriculum and activities (exercise, nutrition, stress reduction, and retention/recruitment)?
This model doesnt show how determined or if best approach is used.
Repeats for each strategy
Context
This model emphasizes the clinical perspective with little detail. To what extent were other
stakeholders involved in developing it? This may influence shared understanding about the
program and how its outcomes are defined and represented.
Eligible and
willing
participants
Coaches
Faculty
Facility
Funds
Exercise
activities
Activities
Resources
Figure 4.4 Health-Improvement Program Logic Model Mark Up
Retention
Relaxation
Flexibility
Fat/Calories
Nutrients
Endurance
Strength
Long-Term
Outcomes
Improved
health
Impact
What about impact? How far out is this?
What level of budget is needed/available to support the achievement of these outcomes?

How long is reasonable to expect that it will take to see progress toward outcomes?
This model shows knowledge, skill, adherence, and motivation as contributing to the
changes in health status. What else might help or hinder participants making progress
toward improved health?
In particular, the specifics of how each outcome and impact will be measured will
matter a lotfor example, Is it okay if people weigh more but are also more fit?
The short- and intermediate-/long-term outcomes and impact are specific, measurable, and
action oriented. They are appropriate for the overall level of detail expressed in this
design-framework-type model, but more detail would be needed to operate, manage, and
evaluate the actual program.
SMART
Who is the best target for this program? Who isnt? Eligible and willing may be insufficient.
What about able? Are the intended activities appropriate, safe, and effective for this target?
Do we assume these details are in the narrative that accompanies the model?
Motivation
Adherence
Skill
secure the intended impact. The real time and context between activities and outcomes
means many external issues also have the potential to influence the outcomes.
Typically, groups vary on how much emphasis they place on each of the characteristics of quality models described here. As with model interpretation and review,
context and composition of the group frequently determine how the difficult distinction is made.
This also points to crucial issues in execution. When a model (representing an
idea, project, or program) is evidence based, strategic, and well designed, implementation can still fail its potential. Although logic models are often used to inform
evaluation, modeling at this stage is rarely done (see Chapter 5). Although outside
the scope of this text, great execution is absolutely essential to results.
Quality Questions
The models you (and your colleagues) build reflect your collective thinking. Nobody
would argue with this assumption: Best thinking is a critical input to any work and
its likely success. As you consider the design or plans for your project (program or
idea), the following questions reflect quality considerations. They include application of the SMART and FIT principles. You might construct other questions based
on your own experience or particular subject matter expertise. Think about ways to
conduct discovery around blind spots, myths, and other culture concerns. Regardless,
it is important that deliberate efforts are invested in the maturation of a model.
Modeling provides an opportunity to test, improve, and revise models. This process
helps develop the thinking and the program that the model represents.
Quality Questions for Program Logic Models

1. Are the results specified with shared meaning among all stakeholders?
2. What research, practice, and theory are grounding for our choices in
activities? Is there a relationship between the program logic model
and a theory of change?
3. Did we challenge the recipe of activities and outcomes to ensure
feasibility given the challenges we seek to resolve?
4. What activities have worked under similar conditions? What current
conditions might influence selection of activities? Is there evidence
that these activities are best?
5. Is the model FIT?
6. Are target audiences clearly defined?
7. Does the interaction among activities contribute to intended outcomes?
8. Do the outcomes feasibly contribute to intended results?
9. Is the model SMART?
10. Do we have adequate resources and time to accomplish intended results?
Chapter 4Modeling59
A Quality Model
Figure 4.5 displays characteristics for model quality. It assembles key narrative from
this chapter into a graphic. The figure describes two important standards for model
quality: plausibility and feasibility. The quality characteristics for theory of change
models are noted (as in Chapter 2), where the focus is on the relationship between
strategies and results. The quality characteristics for a program logic model are
captured by FIT and SMART principles. We suggest the application of this in a
mark up is one way to work with colleagues in modeling. We think, taken together,
theory of change models that are plausible and program logic models that are feasible can contribute to effectiveness.
Better Decisions
Earlier in the text, we asked three questions about effectiveness:
The second, about making better decisions, is highly central to modeling. In
effect, we apply this query to strategy and activity selection relative to results. How
you choose to focus time, energy, talents, and resources in relation to your specified
success is related directly to effectiveness. There are many ways to secure a named
intention. Discarding strategies/activities that are peripheral, modest contributors,
Figure 4.5 Factors and Features of Quality Models
Plausibility
Quality
Characteristics
Co-created with
shared meaning
Evidence based
Appropriate
scale
Theory of Change
strategiesresults
Test Underlying
Idea for Plausibility
Better
Models
Feasibility
Quality Questions
FIT
SMART
Program Logic Model

activitiesoutcomes
Test Program Design

for Feasibility
Better
Programs
or less than optimal focuses limited resources. Further, specificity about strategies
and activities contributes to the results you and your colleagues intend to secure.
Ambiguity should be avoided. Eventually, right work is also about the strategies
and activities in the program logic model. Giving conscious attention to the criteria
employed in selecting strategies for the theory of change model and then the specific activities in the program logic model can make a big difference in securing
results. Right work gets clarified and better decisions considered through the
process of modeling. Models and their versions can develop discipline as well as
standard practices that contribute to new understandings about what will generate
progress on results.
IN SUMMARY
Although frequently overlooked, modeling is a helpful technique to improve your
chances of success, because great design and plans come from great models. We
suggest iterative versions of models are co-created through a disciplined process
that tests and retests quality. Models can be greatly influenced by unintentional
omissions (blind spots), myths, politics, persuasions, and perceptions. Logic, scale,
and specificity all have a bearing on models, too. A simple way to explore model
quality is the application of SMART and FIT principles. We suggest a mark up as a
good way to critically review your program logic model.
LEARNING RESOURCES
Reflection
1. Given how subjective program logic models are, what are the implications for
the outside reader of a model? What does a model that will be read and perhaps used by those other than those who constructed it have to communicate?
2. What role might politics, persuasion, or perception play in how a model might
be created, tested, and improved? How do these issues influence model quality
and use?
3. What prevailing myths might influence choices in your workplace or family?
How do blind spots influence choices?
4. How might the improvement process for a simple, single-site project model be
different from that for a more complex multisite, multilevel initiative? What concerns should the model development team be sure to address, and what aspects
of the model will be most important to communicate?
5. Can a complex, comprehensive program be effectively modeled with a single
diagram? Why or why not? How would you approach a task like this?
Chapter 4Modeling61
Exercises
1. Imagine what a strand focused on marketing might look like to promote an
innovative new hearing aid. Make an assignment for individuals to illustrate
this strand. Give them all the same amount of specified financial resources
over 12 months and tell them the desired outcome is maximum sales volume.
Then, ask them to name their target market and key activities in the marketing
strategy. Compare and contrast versions each participant creates. What
assumptions, knowledge, or evidence accounts for the differences in how the
marketing strands are constructed? What criteria are useful for reconciling
these different illustrations?
2. Consider the questions raised in Figure 4.3 about the Community Leadership
Academy program logic model. How would you answer them? What would the
model look like once those questions had your answers? Draw it and discuss the
implications of your response to the mark up.
Consider the questions raised in Figure 4.4 about the Health Improvement
Program. How would you answer them? What would the model look like once
those questions had your answers? Draw it and discuss the implications of your
response to the mark up.
3. Draw a logic model for learning a foreign language. Ask a colleague to conduct a
mark up using SMART and FIT on the models. What might change? How? Why?

Texts
The Joint Committee on Standards for Educational Evaluation. (1994). The program evaluation standards: How to assess evaluations of educational programs (2nd ed.). Thousand
Oaks, CA: Sage.
Van Hecke, M. L. (2007). Blind spots: Why smart people do dumb things. Amherst, NY:
Prometheus.
Journal Articles
Alter, C., & Egan, M. (1997). Logic modeling: A tool for teaching critical thinking in social
work practice. Journal of Social Work Education, 33(1), 85102.
Doran, G. T. (1981). Theres a S.M.A.R.T. way to write managements goals and objectives.
Management Review, 70(11), 3536.
Dwyer, J. (1996). Applying program logic model in program planning and evaluation. Public
Health and Epidemiology Report Ontario, 7(2), 3846.
Israel, G. D. (2010). Using logic models for program development (AEC 360). Gainesville, FL:
University of South Florida, IFAS Extension. Retrieved December 7, 2011, from http://
edis.ifas.ufl.edu/wc041
Julian, D. (1997). The utilization of the logic model as a system level planning and evaluation
device. Evaluation and Program Planning, 20(3), 251257.
Renger, R. (2006). Consequences to federal programs when the logic-modeling process is not
followed with fidelity. American Journal of Evaluation, 12(27), 452463.
Rush, B., & Ogbourne, A. (1991). Program logic models: Expanding their role and structure for
program planning and evaluation. Canadian Journal of Program Evaluation, 6, 95106.
Internet Resources
Burke, M. (n.d.). Tips for developing logic models. Retrieved December 7, 2011, from http://
www.rti.org/pubs/apha07_burke_poster.pdf
In addition to practicing the review steps on your own models, there are many
other examples of logic models to work from available on the Internet. For several
different approaches, see the following:
Duigan, P. (n.d.) Outcomes model listing. Retrieved December 8, 2011, from http://www.out
comesmodels.org/models.html
Tucson: University of Arizona Cooperative Extension. (2009). Logic models. Retrieved
December 7, 2011, from http://extension.arizona.edu/evaluation/content/logic-modelexamples
Capable communities: Examples. (2011). East Lansing: Michigan State University. Retrieved
December 7, 2011, from http://outreach.msu.edu/capablecommunities/examples.asp
SEDL. (2009). Research utilization support and help: Logic model examples. Retrieved December 8,
2011, from http://www.researchutilization.org/logicmodel/examples.html

The Logic Model GuideBook 2nd Edition

Uploaded by

Copyright:

Available Formats

The Logic Model GuideBook 2nd Edition

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Logic Model GuideBook 2nd Edition

Uploaded by

Copyright:

Available Formats

What are the benefits of using logic models?

What are the benefits of using logic models?

What are the two main types of logic models described in the text?

What are the two main types of logic models described in the text?

1

Introducing Logic Models

Work in any sector, whether private, charitable or government, requires

Chapter IIntroducing Logic Models3

Logic Model Benefits

Logic Models Defined

Logic Model Uses

Chapter IIntroducing Logic Models5

Two Types: One Logic

Table 1.1 Features of Model Types

Theory of Change Logic Model

Program Logic Model

Few (do + get)

Targets + specified results

Theory of Change Model Example

Program Logic Model Example

Chapter IIntroducing Logic Models7

Figure 1.1 Community Leadership Academy Theory of Change

Figure 1.2 Community Leadership Academy (CLA) Program Logic Model

Chapter IIntroducing Logic Models9

Program Logic Model and Evaluation Design

Is the CLA doing

What are the ways that community

What difference has the CLA

Figure 1.3 Community Leadership Academy (CLA) Program Evaluation Model

Chapter IIntroducing Logic Models11

Limitations of Logic Models and Modeling

Models Begin With Results

Logic Models and Effectiveness

Chapter IIntroducing Logic Models13

Figure 1.4 Effectiveness and Logic Models

References and Supplemental Readings

Chapter IIntroducing Logic Models15

Building a Theory of Change Model

Chapter 2Building and Improving Theory of Change Logic Models17

Preferences and Styles

Evidence Based and Plausible

The Big Picture

Multiple Strategies and Results

Chapter 2Building and Improving Theory of Change Logic Models19

Figure 2.3 Health-Improvement Theory of Change

Chapter 2Building and Improving Theory of Change Logic Models21

Knowledge and Assumptions

Action Steps: Creating a Theory of

Figure 2.4 Informing a Theory of Change

Chapter 2Building and Improving Theory of Change Logic Models23

Figure 2.5 Steps in Creating a Theory of Change Model

Improving Theory of Change Models

Unpack and Share Assumptions

Figure 2.6 Success in Theory of Change Models

Research and Theory

Practice and Experience

Chapter 2Building and Improving Theory of Change Logic Models25

Figure 2.7 Toggle: Improving Theory of Change Models

Promising Practices and Benchmarking

Chapter 2Building and Improving Theory of Change Logic Models27