Based on the theory of reasoned action, the present research built separate models of turnover for baccalaureate and nonbaccalaureate nurses and for married and unmarried ones. A representative sample of 755 registered nurses completed the research questionnaire. Six months later, information was obtained from their hospitals concerning whether each nurse remained or resigned. Results were supportive of the theory of reasoned action. Furthermore, significant differences emerged in the predictors for the four subgroups. Moral obligation was found to be a significant predictor for intention only for baccalaureate and unmarried nurses. Part-time versus full-time employment status was found to be a significant predictor of turnover only for unmarried nurses. These results indicate that individuals interested in understanding and reducing nurse turnover may wish to be sensitive to differences between various subgroups of nurses.