Entering and exiting Bachelor of Social Work students were compared to determine if participation in an accredited BSW program predicted both higher social work direct practice skill and higher direct practice self-efficacy, while controlling for experience and intent to enter direct practice. Direct practice self-efficacy was examined as both a predictor of direct practice skill and as a mediator of the relationship between BSW education and direct practice skill. Entering BSW students were compared with exiting BSW students on self-efficacy and skill performance. Direct practice skill was assessed by rating student performance in a fifteen minute initial assessment interview with a standardized client. Direct practice self-efficacy was assessed using the Social Work Direct Practice Self-Efficacy Scale, incorporating items from the Social Work Self-Efficacy Scale and the Helping Skill Self-Efficacy Scale. Ordinary least squares hierarchical regression found BSW education to be predictive of both higher direct practice skill and higher direct practice self-efficacy. Self-efficacy, however, was found to be a negative predictor of direct practice skill, after controlling for BSW education. Self-efficacy acted as a suppressor variable in mediating the relationship between education and skill. This study supports the assertion that BSW education is predictive of higher skill, suggests caution in the use of self-efficacy as a proxy measure for skill, suggests more consideration of standardized client interviews for assessment in social work, and affirms the need for further development of valid and reliable instruments for assessment of direct practice skill in social work.