Abstract
In less than 20 years, logic models have been established as a central and established method for laying the foundation for program planning and evaluation in many sectors, particularly in nonprofit work, government grant programs, and educational design. Through the dissemination efforts of organizations like the Kellogg Foundation and United Way, among others, logic models are now an expected feature of most program descriptions. Program funders, in particular, routinely require carefully developed logic models in grant applications and subsequent reports.
As a program evaluator who spends considerable time engaged in capacity building with novice evaluators, I am always on the lookout for books to support my work. The first edition of The Logic Model Guidebook: Better Strategies for Great Results, published in 2009, has long been an important resource for my capacity building efforts. With its clear and concise approach to developing and using logic models, I have found it to be a valuable aid for teaching logic modeling to the educators with whom I work. The second edition provides even greater utility through the addition of more complex and detailed graphics to support the text and the addition of a new chapter with a selection of actual program logic models that illustrate the multitude of ways that these models can be presented and used. These examples show the use of logic models to answer three questions put forth in the opening pages of the book: (1) Are we doing the right work? (2) Are we making the right decisions? and (3) Are we achieving superior results? The answers to these questions are foundational to our efforts to plan, implement, and evaluate programs, and this new chapter provides a real-world glimpse into the importance of well-developed logic models for these purposes.
The authors, both leaders in the field of logic modeling, present a thorough examination of the topic, from basic concepts through sophisticated program models that are complex, realistic, and understandable. Both authors have significant backgrounds in nonprofit work, notably with the Kellogg Foundation, and Cynthia Phillips was a primary writer for the Foundation’s influential Logic model development guide (W. K. Kellogg Foundation, 2004).
The book is written specifically for students and field practitioners, and assumes no prior knowledge with logic models or modeling (a distinction between the two is presented in the first chapter). It is organized into two sections, each containing four chapters. The first section focuses on the construction of logic models and the second on their application. Each chapter walks the reader through a focused and clearly presented overview of its topic, followed by considerable in-depth information. The chapters end with reflection questions, simple but informative exercises, and a comprehensive list of resources and readings for further learning. While the book can easily be used independently, it is also well suited for a group learning situation, especially because of the learning exercises, many of which will be most effective when used in a group. In my experience, developing skill in logic modeling takes guided practice, and while students new to the subject may easily grasp the concepts and be able to define inputs, outputs, and outcomes, they are often at a loss when translating this knowledge into real logic models for their own work. I have used the first edition as part of an evaluation learning circle and will continue to do so with the second edition.
Numerous pertinent examples and illustrations accompany the text, and readers are informed of the learning objectives for each chapter. However, no matter how simply logic models are presented, they are not always easily understood by someone with no prior experience, and there are places where advance knowledge is needed. For example, the second chapter focuses on logic models as representations of theories of change and it describes how to use and improve these theories. For those unfamiliar with the definition and purpose of program theory, much of this chapter could be somewhat unclear to a novice logic modeler.
Knowlton and Phillips make an essential distinction between theory of change logic models and program logic models. The more familiar of the two is the program model, which focuses on the operational plan for the program, elucidating a series of if–then relationships that should lead to program success. However, the authors emphasize the importance of beginning with a theory of change logic model, which clarifies the plausibility of a potential program’s prospects for success. As the authors note, a theory of change articulates what one will “do” (strategies) and what one will thereby “get” (results). The need for evidence to establish the plausibility of this relationship is critical, and the authors help the reader understand that evidence comes from research and theory, yes, but also from practice and experience. For the on-the-ground program developer, this is a welcome acknowledgment of the mutuality between research and practice and helps to defuse the idea that logic models are an intellectual exercise with little practical value. The authors show how evidence and arguments can be established through replication of prior effective programs and through hypothetical models that illustrate a program’s potential success. The goal of the theory of change model is to show that a certain strategy could work; it can be thought of informally as a “cocktail napkin” description of the program. The authors present a useful elucidation of the nature of assumptions in theory of change models and suggest specific methods for improving these models.
The authors encourage readers to explore and understand the benefits of sound logic models, highlighting in particular their usefulness for discovering specific priorities for program evaluation through the articulation of key program features, processes, and relationships that need assessment. The use of logic models to set the stage for quality program evaluation is especially important, as training in logic models alone has often failed to prepare learners to conduct evaluations. The idea of “placing evaluation questions on the model”—for example, identifying specific relationships or predictions within the model that can be confirmed through evaluation activities—is an approach I have used in trainings, with great success. This process is well described in the opening chapter and effectively supports the principle that logic models are intended to set the stage for evaluation.
Through many useful examples, the reader is taken through a step-by-step process of developing a high-quality program logic model. Of particular use is the emphasis on beginning with the intended results, then working backward to activities and inputs. This process deserves emphasis because of the natural inclination to read models left to right. Experienced program planners begin with the end in mind (the right side of the model) and work backward from there, articulating the if–then relationships that need to happen in order to reach the desired outcome. As Michael Quinn Patton has advised, “we have to dream big” and then figure out how to get there by working backward to where we are now.
In my experience, the linear characteristic of a simple logic model can sometimes disrupt its deeper purpose and use, resulting in some learners dismissing the whole idea as overly simplified; for others, beginning with the simple linear model sets the stage for more complex understanding. The authors present many clear and concise images of nonlinear logic models. Such presentation is particularly useful for those of us who teach logic models to a wide variety of learners, by providing new and innovative illustrations that do not look like the standard line of connected boxes with which so many evaluators are familiar.
Going deeper, Knowlton and Phillips take us into the process of logic modeling with the ultimate goal of creating a detailed, quality program model. The authors examine potential errors and omissions in describing a program’s processes, as well as the importance of considering the program context in terms of program culture, politics, and the potential use of evaluation knowledge within the organization. The goal is to help the reader see logic modeling as an inherently iterative process that, at its best, contains input from multiple perspectives.
The authors argue that careful attention to articulating program outcomes and processes are key ingredients to developing sound logic models and, toward those goals, they present the useful techniques of SMART (specific, measurable, action, realistic, time), commonly used in developing program outcomes, and FIT (fidelity, intensity, targeted), which is a method for articulating program processes. Of particular use is the annotated illustration of a “mark-up” logic model, which expands and clarifies the information that is often presented cryptically, with minimal description, in the logic “boxes.”
Knowlton and Phillips acknowledge that the progression from logic models to evaluation requires a level of evaluation knowledge that is beyond the scope of their book. Nonetheless, the application sections do a nice job of making the case for the contributions of logic models to evaluation, at all stages of program planning and delivery. Perhaps, my favorite part of the book is the detailed presentation of different visual forms of logic models. These presentations are useful for encouraging learners to “think beyond the boxes,” and they set the stage for releasing the potential creativity inherent in logic modeling. In addition, the authors spend considerable time exploring the use of archetypal models, which they portray as a recipe for success based on what has come before. An example might be a program logic model designed to include components and strategies that most experts would consider essential for program success based on accumulated research, practice, and other evidence. Much like we know what goes into making good banana bread, we also often know the key ingredients and processes that make a high-quality “recipe” for program change. Readers are encouraged to build their logic models based on established program theory and key elements from past research, rather than continually reinventing the recipe from scratch.
Given that logic models have become a central aspect of program planning and evaluation, evaluators must continue to develop their basic design and utility. In many instances, logic models are too easily seen only as a necessary requirement for obtaining program funding, rather than an inherently valuable part of program design, implementation, and evaluation. In order to move beyond a mere exercise, our understanding and use of these models need to be continually expanded, to stay relevant to the ways that programs really look and operate. The basic logic model of the past has been simultaneously informative and limiting. The designs and applications of the models presented in The Logic Model Guide Book: Better Strategies for Great Results take us further down the path toward authentic logic model development and use. The second edition of this book sits prominently on my desk these days as a valuable tool for guiding beginning evaluators. It is, as well, a competent guide for more advanced journeys into the nuances of program planning, and the inherent rewards that await the traveler at the end of such treks to deeper understanding. With or without a guide, this book will be an essential resource for all who want to plan effective programs, or to teach others how to do so, in ways that go beyond an exercise in filling out boxes.