COMPARING THE ACCURACY OF DECISION TREES AND LOGISTIC REGRESSION IN PERSONNEL SELECTION

Abstract

Being able to make better personnel decisions is a problem that many organizations consider. Actuarial methods have been shown to make more accurate decisions than human decision making. This study examines the performance of two actuarial methods. (1) The decision tree method, a fast and frugal approach to decision making that has been shown to be equally as accurate as other actuarial models in making decisions. As well as, (2) logistic regression a decision aid that has been often used in selection assisting in making selection decisions. Study one investigates the accuracy of each method using a simulated data set where performance is known. Study two examined how these methods performed when predicting acceptance to a graduate school program. Study one and two found that the decision tree method was equally as accurate as logistic regression in both scenarios.