The study presented in this document was designed to test, through path analysis, a modified version of the Bean and Metzner (1985) model of nontraditional student attrition in an effort to understand why older students frequently dropped out of a large public university in the Midwest. Four variables from Bean and Metzner's model were deleted and the modified version was used to survey 347 campus freshmen, aged 22 years and over, attending the university in fall 1988. The independent variable was their attrition outcome one semester later. The variables collected by the questionnaire included demographic data and single item Likert-type measures of significant other relationships, family income, work and study hours, institutional commitment, number of children, and intent to leave. Results revealed: (1) that intent to leave, followed by grade point average (GPA) and goal commitment were the most important variables in explaning attrition; and (2) that number of children, weekly study hours, and number of hours enrolled in school failed to contribute directly/indirectly to explaining attrition variance. Despite its limitations, the study concluded that the modified model may have some theoretical relevance in explaining why older students dropped out from a large residentia1 university. Includes 30 references. (LPT)