Sixty-nine graduate students enrolled in a statistics class participated in this study, which investigated the effects of self-monitoring on learning and attitudes toward learning. With protocols designed for this study, subjects in a self-monitoring condition recorded frequency and intensity of their learning activities and rated selfefficacy of solving statistical problems. Subjects in an instructor-monitoring condition evaluated the instruction provided in the class. Subjects in a control condition took the course without any research activity. Scores obtained from class examinations and attitude inventories were the dependent variables. As predicted, the selfmonitoring group performed better than did the instructor-monitoring and control groups. No attitude difference was found. Implications for teaching graduate-level statistics are discussed, and future research on self-monitoring is suggested.