OBJECTIVE: In a recently published survey of alumni of the Columbia University public psychiatry fellowship, respondents who were medical directors reported performing a greater variety of tasks and experiencing higher job satisfaction than those who were staff psychiatrists. Both medical directors and staff psychiatrists believed that job satisfaction was most dependent on clinical collaboration activities. Survey data were reanalyzed to determine whether there was a relationship between the frequency of tasks performed and overall job satisfaction, and whether the tasks that actually predicted overall job satisfaction were the same as those that respondents believed contributed to job satisfaction. METHODS: The survey was distributed to all public psychiatry fellows and alumni in active practice (N=89), and 72 forms (81 percent) were returned. The survey consisted of 16 self-administered items divided into three categories of job tasks: direct service, clinical collaboration, and administration. Results and conclusions: Despite respondents' beliefs that clinical collaboration activities contributed most to job satisfaction, performance of administrative tasks was found to best correlate with overall job satisfaction. Furthermore, overall job satisfaction was related to the performance of administrative tasks and not to the job title of medical director alone. Most of the medical directors in the survey had program-level, rather than agency-level, responsibilities. The findings indicate that the role of program medical director can serve as a crucial next step for staff psychiatrists, offering the opportunity to perform administrative tasks, which, according to the results, improves job satisfaction in public-sector positions.