Health professions education researchers continually search for tools to measure, evaluate, and disseminate the findings from educational interventions. Clinical teaching, particularly teaching about the improvement of care and systems, is marked by complexity and is invariably influenced by the context into which the intervention is placed. The traditional research framework states that interventions should be adjudicated through a yes or no decision to determine effectiveness. In reality, educational interventions and the study of the interventions rarely succumb to such a simple yes or no question. The realist evaluation framework from Pawson and Tilley provides an explanatory model that links the context, mechanisms, and outcome patterns that are discovered during implementation of a project. This article describes the unique qualities of the realist evaluation, the basic components and steps in a realist evaluation, and an example that uses this technique to evaluate teaching about improvement of care in a clinical setting.