From the Poverty Reduction and Economic Management Network

The key question in program evaluation is whether the intervention is working or not. But to answer that question, first program outcomes must be attributed to the intervention, and not some other factor. The attribution issue arises in this context because a program, as a means-ends relationship, interacts with contextual factors that may strengthen, reduce, or break the intended causal link between the intervention and the outcomes of interest. These factors therefore represent alternative explanations for any association between a program and observed outcomes. This note reviews two basic frameworks for resolving the attribution problem, including their limitations. Both the counterfactual framework and the theory-based approach rely on the logic of causal inference, which entails establishing a plausible association between the intervention and the outcome of interest, and ruling out alternative explanations of that association. B. Essama-Nssah