4.1: A caution about the linearity of Logic Models

Some people see the logic model as very structured and linear, with neat and tidy boxes lined up like a pipeline or a string of dominoes that fall forward in a linear progression. The program is designed, delivered, and produces outcomes while participants are viewed as passive recipients.

But we know that programs are NOT neat and tidy. Participants are NOT passive. Program development is a very dynamic and iterative process. There are likely to be a series of fits and spurts, following an iterative process of moving forward and then back two steps. Participants are active partners, interacting with the flow. Real programs are messy, as are the environments in which they exist.

To depict the nonlinear nature of programs, many logic models use vertical, two-directional, and circular arrows and loops to depict the more interactive nature of causal relationships (Funnell, 2000; Rogers, 2000). For example, an increase in knowledge can lead to a change in practice that in turn leads to the need for other or more knowledge; or a change in attitude may influence behavior that influences attitudes; or a policy change may create greater awareness that leads to behavioral change; or positive reactions to the program lead to increased attendance that leads to more services being provided; or an external factor causes a programmatic change that in turn affects the external environment.

The discussion in the previous two notes describes the logic model as a systems model: not a simple, “input causes output causes outcome” model but one where cause-effect relationships are connected in multiple and nonlinear ways.

Thus, as we saw in Section 3, logic models are usually not a single line of boxes connected by arrows. They are multiple chains with horizontal, vertical, and diagonal connectors between and among components, including the external environment. In fact, that’s the hardest thing about developing a logic model–depicting the lines and arrows that show connections and the circular feedback loops in a way that communicates to users.

Some people caution about the seeming linearity of logic models: they often are neat and tidy, with boxes lined up like a pipeline or like a string of dominoes that fall forward in linear progression. We know that programs are NOT neat and tidy. They are more likely to be a series of fits and spurts; to follow an iterative process of moving forwards and then back two steps. Real programs are messy, as are the environments in which they exist.

Concerns about the linearity of logic models

Some people see the logic model as a very structured, top-down approach to programming. It is equated with a program delivery model where the program is designed, delivered, and “produces” outcomes. Program participants are viewed as passive recipients in the flow of action. Rather, program development is a very dynamic, iterative process. Active participants are involved, interact with and influence the flow of action and outcomes achieved. They are partners, not objects, in program delivery. Possibilities and potential cause-effect relationships are numerous, not contained to predetermined boxes and arrows.


Support Extension