PREDICTING DROPOUTS OF UNIVERSITY FRESHMEN: A LOGIT REGRESSION ANALYSIS

Factors considered important for predicing attrition were extracted from the research done on university dropouts for the past twenty‐five years and integrated into a conceptual model. Stepwise discriminant analysis coupled with logit regression analysis of freshman data from Brandon University yielded six variables: student status, residence, financial sources, distance of hometown from the university, goal fulfilment, and satisfaction with overall university atmosphere, which were useful in prediction. Several recommendations are made for minimizing the withdrawal phenomenon.