Lessons Learned: Building a Logic Model

Logic models provide stakeholders with a road map that articulates how key program activities lead to the desired results, and they also support a new way for organizations to think about their impact. See our blog post on how to develop a logic model here.

Over the last impactED_10-2017_Jan-60-M.Edlowmonth, we’ve supported ten organizations in the Social Impact Collaborative* develop logic models. Below are a few lessons we’ve learned along the way.

Consider how your logic model fits into the broader organizational vision. We encourage organizations to develop a logic model for a specific program/project/campaign or initiative to ensure the necessary focus and depth. However, no program exists in isolation. Several of the organizations we’re working with have expressed the importance – and challenge – of aligning the work of any single program logic model with the broader organizational vision. We recommend:

  • Highlighting the parts of your strategic plan influenced by a particular program logic model to ensure alignment
  • Encouraging other members of your organization to create logic models for additional programs or projects
  • Incorporating the logic model way of thinking (i.e., linking resources to short-term goals to long-term impact) into organizational conversations

Use data to continually test your assumptions and refine your logic model when necessary. As a program grows and develops, so does its logic model. A logic model is a work in progress that can be refined as a program develops. Many of the organizations we’re working with have asked how to incorporate ongoing data collection and an evolving understanding of stakeholders’ needs into their logic model. We recommend:

  • Making research and data collection a key activity in the logic model
  • Setting times throughout the year to revisit the logic model based on new data to make revisions/additions as necessary

Record your key assumptions. One of the challenges of creating a logic model is being realistic about your organization’s contribution to ultimate outcomes. Many of the organizations we’re working with are part of broader coalitions, which means that achieving their ultimate outcomes require outside support and the presence of certain external conditions. We recommend:

  • Documenting the key assumptions necessary to lead to change at each transition point (e.g., from short-term to medium-term to long-term outcomes) in the logic model
  • Continually revisiting your resources (and in particular partnerships) to ensure you have the necessary supports to realize the desired change
  • Be realistic about what type of change your organization can achieve in a certain time period

Distinguish between outputs (what you do) and outcomes (the impact of what you do).  We found that distinguishing between outputs and outcomes can be challenging. Outputs capture implementation fidelity…. In other words, they tell you the extent to which you did what you set out to do as articulated in your activities. They capture things like the number of users, the number of trainings, or the number of social media followers. While these are important goals, they are not “outcomes.” Outcomes reflect the change that should occur as a result of whatever is implemented. When determining outcomes, we recommend considering participant/stakeholder/or constituent:

  • Knowledge learned or attitudes as a result of participation in a particular program
  • Skill to execute on what was learned and behavioral change as a result of the program
  • Sustained changes in decision-making or status as a result of the program

Don’t forget the mechanism of change.  This South Park clip about the underpants gnomes provides a funny – and relatable – example of what often happens when organizations create a logic model. Organizations typically start by developing short-term outcomes (for example, they may track a change in knowledge resulting from a program) and then expect that these short-term outcomes will lead to their long-term outcomes (for example, a change in decisions or status). But they forget the mechanism of change (or the medium-term outcomes!). We recommend thinking about change over time in the following way:

  • Short-term: Knowledge learned or attitudes
  • Medium-term: Skill to execute on what was learned and behavioral change
  • Long-term: Sustained changes in decision-making or status

 

What lessons have you learned from creating a logic model?

 

*The Social Impact Collaborative is being funded by the William Penn Foundation, but the opinions expressed in this blog post are those of the author(s) and do not necessarily reflect the views of the William Penn Foundation.