Evaluation of Programs With Numerous Jurisdictional Implementations

This article explores the problem of evaluating programs with numerous jurisdictional implementations. Typically, budget and time constraints may not allow evaluation of every instance of program implementation. Therefore, a method is needed to guide the selection of a sample of implementations, the analyses of which are generalizable to the program population as a whole. Two sample selection techniques are introduced: cluster analysis and an optimization technique. The techniques are applied to select a sample of states to evaluate the national benefits and costs of the Residential Conservation Service.