Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)

First Advisor

Richard G. Lomax

Second Advisor

Diana G. Pounder


The intention of this dissertation was to evaluate empirically the generalization of a theoretical model. The model (Tinto, 1987) studied was intended to predict departure from college. In this study the model was extended to departure from career. The basic premise of Tinto's theory is that voluntary departure from the college community is the result of a decreased integration between an individual and the institution. Concepts of the Tinto model were used to develop the career persistence model, which depicts how the preoccupational attributes of academic achievement and educational attainment are a function of precollege goal commitment. These variables plus kinship responsibilities continue to affect the individual's job satisfaction, intentions, academic integration, goal commitment, and persistence throughout the career span. To test the model, a sample of 270 nurses who were part of the National Longitudinal Study-1972 (NLS-72) from 1972 to 1986 was studied. The analyses of data were completed with a FORTRAN-77 program, GEMINI (Wolfle & Ethington, 1985). The program calculates indirect and direct effects and their significance for a specified recursive model. The results of the study revealed the following significant findings. Goal commitment had positive effects on achievement and attainment in college, however it inversely affected academic integration. Achievement had direct effects on academic integration and indirect effects on job satisfaction through goal commitment. The level of attainment directly affected intention and later job satisfaction. The degree of academic integration had major importance across time as a contributor to job satisfaction. The first measure of Kinship responsibility inversely affected intention and persistence, but nurses with higher family responsibilities had higher job satisfaction later in career. However, ten years after college graduation, the only direct predictor of the final measure of career persistence was earlier persistence. The study findings demonstrate the importance of employing methods that analyse the significance of both direct and indirect effects. Although the model did not show a significant causal relationship between most variables in the model and persistence, relevant findings were shown for other departure issues.