Identifying distinctive groups in a college applicant pool

In a time when most post-secondary educational institutions must distribute limited resources more efficiently, segmentation analysis affords a way to direct planning to yield strategic benefits. The Automatic Interaction Detector is recommended to the institutional research community as an analytical tool for effectively identifying distinctive subgroups within an educational market. In this application, AID is used to segment the Boston College applicant pool according to subgroups' relative probabilities of enrolling. The findings illustrate that AID can make a useful contribution to market research and that the technique has broader applicability—to other student groups and to other educational policy questions.An earlier version of this paper was presented at the Twenty-First Annual Forum of the Association for Institutional Research, Minneapolis, May 1981.