Predicting freshman persistence and voluntary dropout decisions: A replication

This paper describes the results of a replication of a study testing the predictive validity of a 34-item instrument designed to assess the fundamental constructs of Tinto's model of college student attrition. A design, variables, and analytical procedures virtually identical to those of the original study (done at a large independent university) were used, and this research was conducted at a large public university. The five-factor structure, found in the original study to underlie the 34 items, was replicated almost exactly. As in the earlier work, the Institutional and Goal Commitment Scale was a significant predictor of attendance behavior even after controlling for a variety of students' precollege characteristics. Potential institutional differences in faculty members' influence on retention were identified. A cross-validation classification procedure suggests the five factors are reasonably stable predictors of attrition.An earlier version of this paper was presented to the annual conference of the American Educational Research Association, Boston, April 1980.