Abstract
Suppose that we wish to use several methods to obtain data from several sources for a program evaluation. Because not all data are equally valuable and because the costs of collection vary from method to method, the question arises as to whether it is possible to maximize information yield for a fixed, limited budget. How this problem was structured and solved for a particular program evaluation is described in this article. The approach, which uses a mathematical technique known as linear programming, is easily generalized to other program evaluations.
Get full access to this article
View all access options for this article.
