A causal model of college attrition with a secondary school population: A path analysis

Frances Loretta Tobin, Fordham University


The present secondary analysis attempted to apply Tinto's causal model of college attrition (1975) to a high school population. According to his model, background variables are hypothesized to influence student persistence through two sets of endogenous variables: the academic system (academic integration and goal commitment) and the social system (social integration and institutional commitment). The model used included the modifications of Pascarella, Duby, and Iverson (1983) whereby background influences also affect persistence directly.^ Data were taken from a nationally representative study, High School and Beyond. All nonhandicapped students who had completed the sophomore year base study in 1980, and both the first and second follow-up surveys (1982 and 1984) were included in the sample. The total sample was divided into three groups according to their educational aspirations upon entering high school as follows: (a) students already decided to attend college, (b) students undecided as to college attendance, and (c) students already decided against college attendance.^ The outcome variables were two separate measures of educational perseverance: high school completion (or noncompletion) as of first follow-up and educational status as of second follow-up.^ Six path analyses were performed to allow comparison across groups and between perseverance measures. The results indicated that, for all three groups, both perseverance outcomes were influenced much more strongly by the academic system than by the social system. The model explained more variance in follow-up status than in high school completion, and more variance for the college-bound than for the other two groups. Background effects on perseverance were transmitted more indirectly for the college-bound group and more directly for the undecided group. For the non-college-bound group, background effects were transmitted chiefly through academic integration on high school completion, but chiefly through goal commitment on follow-up status.^ The model explained approximately the same amount of variance as it had in studies using college populations. While this amount was modest, it was nevertheless concluded that the model was useful in comparing the pattern of effects upon the perseverance of high school students of differing motivations. ^

Subject Area

Adult education|Educational psychology|Secondary education

Recommended Citation

Tobin, Frances Loretta, "A causal model of college attrition with a secondary school population: A path analysis" (1991). ETD Collection for Fordham University. AAI9136341.