"I have been searching for a book to train my students, research assistants, and field staff to design and implement social policy experiments. My search is over. Running Randomized Evaluations provides practical guidance covering all stages of an experiment--wisdom that previously could be acquired only by working directly for an evaluation expert."

-Jeffrey B. Liebman, John F. Kennedy School of Government, Harvard University

About the Book

Running Randomized Evaluations gives evaluators and practitioners the know-how they need to do valid randomized impact evaluations of social programs in developing countries.

The first three chapters explain how randomized impact evaluations can be a powerful tool for fighting poverty. They explore why it is so difficult to measure the causal effects of programs and how randomizing can help address those issues. They also explain the role the evaluator plays in program design, and how evaluators can choose the right time and context for conducting an evaluation.

The next five chapters take the evaluator step by step through the process of doing a randomized evaluation. They cover the choice of randomization technique, planning for data collection, designing the evaluation to have high statistical power, addressing threats to the validity of the experiment, and analyzing the data.

The final chapter concludes with an overview of how to interpret and draw policy conclusions from the results of randomized evaluations or generalize the results from one context to another. It explains the usefulness of comparing the cost-effectiveness of different programs based on the results of different randomized evaluations. It concludes with a list of potential pitfalls, and when to question the validity of the results of a randomized evaluation.