Developing a Framework or Model of Change is a process of mapping out what will be done regarding an intended effort for change. This process helps groups outline the relationship among inputs (i.e., resources), outputs (i.e., proposed interventions), impact (i.e., immediate results of those outputs), and outcomes (i.e., indicators of community change, behavior changes, and population-level improvements). When groups engage in this process, they are addressing "How will we get from here to there?" The process of Developing a Framework or Model of Change can help groups consider how to get from "where we are" (i.e., features of the problem, current resources) to arrive logically and systematically at "where we want to be" (i.e., a continuum of outcomes that describe resolution of the problem). Developing a Framework or Model of Change is a key process to help communities outline and connect a sequence of events for bringing about change, and can help a group propose an integrated [and testable] model for making a difference based on how they understand underlying mechanisms of change.
Developing a Framework or Model of Change helps us organize our thinking and orient program development by intended outcomes rather than by [often limited] resources. This process aims to develop a visual representation or roadmap (i.e., theory of the problem, theory of action, logic model, or framework). It takes participants on a journey of "walking back" from intended outcomes to clarify what needs to be in place to get there. Developing a Framework or Model of Change helps an initiative specify an intervention's proposed mechanisms for making a difference, and helps to articulate where and how change needs to happen for improvements in more distant outcomes to occur. It can also be helpful for naming indicators of success - the answers to "how will we know it when we see it?" - that can be useful for Documenting Progress and Using Feedback.
Although the exact mechanisms are unclear, participation in Developing a Framework or Model of Change can help a group to:
When logic models recognize and clearly link what needs to change with intervention strategies that can do so, they increase an intervention's chances for success. For example, teen pregnancy prevention interventions are now more directed toward factors contributing to the problem. During the 1980s, many schools implemented sex education programs to increase knowledge about sexuality and contraception; yet evaluations of these knowledge-based programs revealed that they did increase knowledge, but did not change behaviors. Interventions focused on personal factors such as knowledge that they were able to change, but one that was not markedly associated with adolescent sexual and contraceptive behavior (Kirby, 2001). By contrast, a more recent generation of programs focuses more clearly on specific behaviors and research and theory that identify environmental factors to be changed. For example, a change in clinic policy that authorizes more time for teen patients may enable staff to allocate and spend sufficient time for youth clients, thereby increasing youth use of birth control (Kirby, 2002). When intervention strategies are based on a clear logic model for change (e.g., behavior change resulting from environmental interventions), they are more likely to actually change behaviors - which is necessary for population-level outcomes to improve.
Similarly, logic models can be powerful tools to communicate grounded and integrative program strategies - for funding and evaluation purposes. For example, with limited time and no opportunity for a formal presentation, staff for Project WISEUP (a juvenile justice intervention) successfully used their logic model to articulate their ideas so government officials would understand and see its value as a preventive intervention. They used a visual representation to outline a step-by-step plan of expectations, action-strategies, and outcome measures. Their theory of action was that juvenile justice required an ecological model that involved parents, siblings, counselors, teachers, police, and community agencies; a set of programs to build good study skills (measured by grade point averages), self-discipline (decreased suspensions), and regular attendance at school could improve arrest outcomes and reduce incarceration. Consequently, they provided a logical and sound argument that was met with curiosity, support, and subsequent funding (Cato, Chen, & Corbett-Perez, 1998).
At present, much of the information available on Developing a Framework or Model of Change does not explicitly manipulate or test this process and its effects on community change and improvement. Although this process has been identified in empirical and experiential reviews as a key ingredient for advancing change, there is a need for more systematic evaluations of its effects. Such research would provide a better understanding of the factors that enable communities to come together and address shared problems and goals.
Some key research questions to consider include: (a) How can we better integrate knowledge from various disciplines (e.g., community psychology, public health, social work, urban studies) to develop multi-level program theories and models to guide this work of community change and improvement? (b) How can we use logic models to help us better assess how program outcomes are obtained from comprehensive interventions? And (c) What hypotheses for future research can we draw from candidate pathways outlined in logic model of comprehensive community change efforts?
Based on research and experience, we recommend (with qualifications) Developing a Framework or Model of Change as a key process to advance community change and improvement.The process is suggested by many researchers and practitioners as an important tool for designing, strengthening, and evaluation community change and improvement interventions. Yet, the prevalence of empirical and experiential evidence regarding its effectiveness is at an early stage compared to what is known about other Best Change Processes.
