STATISTICAL ALTERNATIVES FOR STUDYING COLLEGE-STUDENT RETENTION - A COMPARATIVE-ANALYSIS OF LOGIT, PROBIT, AND LINEAR-REGRESSION

While higher education researchers have long been concerned with the development and application of methods to adequately assess the impact of college on students, strong advances in statistical theory and computational practice have shifted this focus from the fundamental issues of research design to the application of appropriate statistics. This study focuses on the practical implications of applying logistic regression, probit analysis, and linear regression to the problem of predicting college student retention. Rather than simply assuming that one technique is analytically superior to others based on theoretical grounds, this study explores how these techniques compare in predicting student retention using data provided by registrars from a national sample of colleges and universities. Results indicate that despite the theoretical advantages offered by logistic regression and probit analysis, there is little practical difference between either of these two techniques and more traditional linear regression.