Predicting Graduate Student Success in an MBA Program: Regression Versus Classification

The decision to accept a student into a graduate program is a difficult one, based upon many factors that are used to predict the success of the applicant. Typically, regression analysis has been used to develop a prediction mechanism. Unfortunately, as is shown in this article, these regression models can be ineffective in predicting success or failure. This article evaluates the ability of different models, including the classification techniques of discriminant analysis, logistic regression, and neural networks, to predict the academic success of MBA students. The conclusions of this study are that (a) classification techniques may be an appropriate approach to the problem, (b) predicting success and failure of graduate students is difficult using only the typical data describing the subjects, and (c) nonparametric procedures, such as neural networks, perform at least as well as traditional methods and are worthy of further investigation.