This study tested Pascarella's reconceptualization of Tinto's model of the persistence/withdrawal process for use in a non-residential institution. Subjects (n=372) were college freshmen who had attended freshmen orientation and had completed an opinionnaire both during orientation and at the end of their first year in a medium-sized, urban commuter institution. Path analysis and estimates of direct and indirect causal effects were used to test the predictive validity of the model. Results tended to confirm the importance of applying causal models cautiously even within the same types of institutions. Results further suggest that there is considerable between-institution variance within the broad category of commuter colleges. The influence of key concepts such as social integration and institutional commitment were found to vary significantly from the reconceptualized version of Tinto's model. The degree of fit between the individual student and the institutional environment was largely predicated upon the magnitude of relationship between academic integration and institutional commitment. Other things being equal, the greater the individual's level of academic integration, the greater his or her subsequent commitment to the institution. The combination of academic integration and subsequent commitment were seen as having a positive influence on persistence. (Author/ABL)