This research tested a causal model of student retention on an academically disadvantaged community college student population. The model investigated the direct and indirect effects of four exogenous variables (family background, pre-college schooling, getting ready, and encouragement by significant others) and three endogenous variables (initial commitments, academic integration, and social integration) on the retention of academically disadvantaged students. The causal model was adapted from Tinto's theoretical model of student attrition. A longitudinal design was employed to explore the predictive validity of Tinto's model for this particular population. The method of data collection was a direct mail survey developed to measure the variables in the causal model. A principal components factor analysis was performed in order to estimate the discriminant validity of the instrument with this specific population. Cronbach alpha reliabilities were computed for each of the scales used to measure the variables. Zero-order correlation coefficients were calculated for all exogenous and endogenous variables. Path analysis was employed to examine the causal linkages among the variables in the model. In order to test the fit of the hypothesized causal model to the data a chi squared goodness of fit analysis was performed. [...]