This study examines turnover among direct-care staff of community residential facilities. Turnover is of concern as the projected rate indicated by direct-care staff is 34%. A review of personnel records project an annual turnover rate of 40%. Stress is examined for its relationship to turnover. The Maslach Burnout Inventory is used to measure the perceived stress level of staff. Results indicate direct-care staff are not stressed to the point of burnout in two of the three subscales of the Maslach Inventory. Further analysis reveals no significant relationship between stress and turnover intention. Role conflict, role ambiguity, and role overload are examined for any relationship to degree of stress and turnover. No relationships were found between these sources of stress. Although no relationship exists between perceived stress, roles, and turnover, direct-care staff's reasons for leaving may be related to more money and better management. In order to reduce turnover, potential strategies for administrator's may to be to clearly define the job of direct-care staff and provide sufficient recognition. / Ed. D.