Provides a guide to running randomized impact evaluations of social programs. This book explains why and when randomized evaluations are useful, in what situations they should be used, and how to prioritize different evaluation opportunities.