A theoretical model of turnover in work organizations was applied to the college student dropout process at a major midwestern land grant university. The 854 freshmen women subjects completed a questionnaire that included measures for 14 independent variables: grades, practical value, development, routinization, instrumental communication, participation, integration, courses, distributive justice, campus organizations, opportunity, marriage, satisfaction, intent to leave. It was projected that 12 determinants would influence satisfaction, intent to leave and dropping out in a causal sequence. The model was estimated using path analysis and multiple regression. Nine variables were ranked by total causal effects. From high to low, these were: intent to leave, grades, practical value, opportunity, marriage, satisfaction, campus organizations, courses, and participation. The dropout data were obtained for spring and fall semesters 1979 and spring 1980. It is suggested that the attrition model, based on the Price/Mueller model of turnover in work organizations, was supported by estimation procedures, and that much can be learned about student behavior from the study of employee behavior. The chief difference between the findings for Price and Mueller and the present study was in the effectiveness of intention as an intervening variable to influence dropout or turnover decisions. One reason for this difference is probably due to the shorter time interval for students than for employees between gathering the data on the determinants and the data on dropouts. (SW)