Signaling Student Retention With Prematriculation Data

Abstract
Logistic regression is employed to develop a model that seeks to provide information to enhance early identification of freshmen at risk of attrition. The early identification is accomplished shortly after freshman orientation. The dependent variable of interest is the binary and nominal variable of persistence. Students who proceed from freshman matriculation to graduation without ever having dropped out are classified as persistors, and freshman matriculates who leave college either temporarily or permanently are classified as dropouts. The independent variables employed to predict attrition include demographics; high school experiences; and attitudes, opinions, and values as reported on a survey administered during freshman orientation. The model and its results will be presented along with a brief description of the institutional intervention program designed to enhance student persistence.