The Actor-Based Change (ABC) framework provides a pragmatic approach to Theory of Change development in complex contexts. Unique about this approach is its focus on actors within a system. It draws from three main bodies of literature: complexity science, behavioural change literature, and the literature on models of causal inference. As an iterative process, it specifies the following steps:


1) Develop an actor-based system map with the relationships between the actors.

  • Identify the actors in the system. You may want to start with the primary actor group most affected by the development problem. As you build your understanding of this actor group (with the next two bullet points), identify additional actor groups to include in the system map based on their relationship with or influence on the primary actor group.
  • Build an understanding of how these actors behave with respect to one another and in response to their environment.
  • Break down the “current state behaviour” by describing the capabilities, opportunities, and motivations that allow this behaviour to persist. This way you also include the contextual conditions in your analysis that define the practices and relationships between actors.

2) Define the change agenda.

  • For each actor within scope, define the desired behavioural change and how it differs from their current state behaviour.
  • Also take into account that actors influence each other, thus allowing influencing the behaviour of one actor group through behavioural change in another actor group.

3) Model the unifying Theory of Change using causal impact pathways.

  • Model the Theory of Action per actor group, i.e., the pathways from intervention to behavioural change, including underlying assumptions. As template, you could use the “Capability, Opportunity and Motivation for Behaviour Change” framework by Mayne.
  • Expand each actor-level Theory of Action by addressing how changes might cascade through the system. For each actor-level behavioural change, use the system map (as this informs you about the connections between actors) to trace how that change might work to shift the behavioural conditions of other actors they are related to, thus leading to behavioural changes in these groups, which might then work to shift behavioural conditions in other actors they are related to, and so on. Also, keep adding causal assumptions.
  • Further expand the actor-level pathways, by describing how the collection of the actor-level changes over time would change the macro-level socioeconomic problems and thus address the development problem.
  • Unify the above actor-level impact pathways in a single Theory of Change.

As the actors in the system are interconnected, the result is that you might find multiple possible pathways by which change might cascade through the system, allowing for flexibility.

You can download presentation slides on the framework here on the ToC Academy. There’s also an article about the framework behind a payment wall:

Koleros, S. Mulkerne, M. Oldenbeuving & D. Stein (2018) The Actor-Based Change Framework: A Pragmatic Approach to Developing Program Theory for Interventions in Complex Systems. American Journal of Evaluation,