Association for Public Policy Analysis & Management

Most quantitative policy research starts with a question, and then finds data to answer it. The specification of which data should be found, and how much, is conventionally informed by power calculations, sample sizes, and ensuring statistical significance. Unfortunately, these methods are roundly dismissed by statisticians as directly useful for decision making. Instead, decision theory and statistical decision making can and should inform this process. A guide to this class of analysis, with both quantitative and qualitative examples, will be used to illustrate these methods, their potential, and some of the limitations.