The Use of Test Scores for Classification Decisions with Threshold Utility

The classification problem consists of assigning subjects to one of several available treatments on the basis of their test scores, where the success of each treatment is measured by a different criterion. It is indicated how this problem can be formulated as an (empirical) Bayes decision problem. As an example, the case of classification with a threshold utility function is analyzed, and optimal assignment rules are derived. The results are illustrated empirically with data from a classification problem in which achievement test data are used to assign students to appropriate continuation schools.