Program evaluation and spillover effects

This paper is a practical guide for researchers and practitioners who want to understand spillover effects in program evaluation. The paper defines spillover effects and discusses why it is important to measure them. It explains how to design a field experiment to measure the average effects of the treatment on eligible and ineligible subjects for the program in the presence of spillover effects. In addition, the paper discusses the use of nonexperimental methods for estimating spillover effects when the experimental design is not a viable option. Evaluations that account for spillover effects should be designed such that they explain the cause of these effects and whom they affect. Such an evaluation design is necessary to avoid inappropriate policy recommendations and neglecting important mechanisms through which the program operates.