In this study we tried to answer the question of whether the three predictors from the theory of planned behavior provide a satisfactory explanation for intentions and behaviors in the context of voluntary employee turnover. We examined whether variables which have traditionally played a prominent role in the study of turnover, i.e., job satisfaction, organizational commitment, age, and tenure, were capable of explaining additional variance in turnover intentions and voluntary turnover. We conducted a longitudinal study in which 296 respondents serving as professionals in the Royal Netherlands Navy completed a questionnaire. Half a year later they were asked to answer a second identical questionnaire; data were obtained from 202 participants who answered all relevant questions in both questionnaires. Their actual behavior (staying versus leaving) was registered during a 2-year period following the completion of the first questionnaire. Behavioral intentions proved to be the best predictor of turnover in which the effects of all other variables were accounted for. Job satisfaction and tenure explained a significant proportion of the variance in intentions after the effects of the predictors from the theory of planned behavior had been taken into account. The results are discussed in relation to theoretical, methodological, and practical issues. Copyright © 2004 John Wiley & Sons, Ltd.