Date: Tuesday, September 8, 2026
Greetings! I’m Elizabeth Devaney, an organizational learning consultant and evaluator in the San Francisco Bay Area. After 30 years in nonprofit leadership, I returned to school to study how evaluators can better help organizations strengthen their work in times of change. One of the most useful tools I discovered on my doctoral journey was the concept of double-loop learning.
Consider this scenario. An organization reviews the findings of a program evaluation and responds appropriately: staff adjust their outreach strategies, revise workflows, and improve participant communications. Activities, outputs, and outcomes are monitored according to the program’s logic model and theory of change. However, six months later the program’s results remain poor. Why? Because the organization may have learned from the evaluation, but only within the bounds of its existing assumptions.
Chris Argyris and Donald Schön describe organizational learning as occurring through single- and double-loop pathways. Single-loop learning improves existing actions, while double-loop learning examines whether the assumptions, beliefs, and governing variables behind those actions remain valid. Single-loop learning asks, “Are we doing things right?” while double-loop learning asks, “Are we doing the right things?” Both questions matter. But more precise data will not resolve a problem when the underlying explanation of how change is actually happening becomes invalid.
A theory of change or logic model shows a precise sequence of activities and outcomes, but each arrow contains an assumptive claim: If we do this, then that will happen. Those claims may take for granted that participants trust the organization, staff possess effective interpersonal skills, or that external conditions remain stable.
When results fall short of expectations, we often focus on the boxes: was the activity implemented with fidelity? Did our outreach reach enough people? Were our communications compelling? We check to see if we did everything right. This is single-loop learning.
Double-loop learning directs attention to the arrows: what do we believe connects this activity to that outcome, and what evidence could we gather to test that belief? Double-loop learning opens new opportunities for evaluation and, as a result, program improvement.
I invite teams to annotate each connection in their logic model or theory of change by completing this sentence: “We expect A to lead to B because we assume ______.” We then examine the assumptions using three questions:
Assumptions that are highly consequential and highly uncertain are strong candidates for testing.
Reframe assumptions as questions the evaluation can investigate. The assumption that “Participants will use the service if we communicate more frequently,” becomes, “What conditions influence whether participants view the service as relevant, accessible, and trustworthy?” The latter question opens inquiry beyond tactics and may reveal barriers in program design, organizational relationships, or the service model itself.
A well-designed logic model helps teams make their causal reasoning explicit and offers assumptions to test. For example, evaluators may introduce a set of constituent interviews to investigate how participants’ current experiences might be affecting their responsiveness to the organization’s communications. Ask how the evidence strengthens, weakens, or complicates the leaders’ understandings, and what must be reconsidered as a result.
Our challenge as evaluators is rarely convincing organizations to improve performance. Our greater opportunity is helping them identify which measures deserve examination – because sometimes the most useful finding is not a more precise answer, but a better question.
The American Evaluation Association is hosting Organizational Learning and Evaluation Capacity Building (OL-ECB) Topical Interest Group Week. The contributions all this week to AEA365 come from our OL-ECB TIG members. Do you have questions, concerns, kudos, or content to extend this AEA365 contribution? Please add them in the comments section for this post on the AEA365 webpage so that we may enrich our community of practice. Would you like to submit an AEA365 Tip? Please do so using the form on our website. AEA365 is sponsored by the American Evaluation Association and provides a Tip-a-Day by and for evaluators. The views and opinions expressed on the AEA365 blog are solely those of the original authors and other contributors. These views and opinions do not necessarily represent those of the American Evaluation Association, and/or any/all contributors to this site.