Purpose. This study assesses the relationship between students' confidence and competency in breast screening practices, compares different measurement approaches to these competencies, and assesses the effect on clinical breast examination (CBE) performance of an additional training session with a standardized patient. Method. In the spring of 1998, 96 third-year medical students (47 men and 49 women) at the University of Massachusetts Medical School completed knowledge items on breast cancer epidemiology and on perceived confidence in their counseling and CBE skills. The students were also rated on their performances of counseling and CBE skills. Results. Higher levels of confidence in mammography counseling skills and higher interview skills scores were associated with higher mammography counseling scores. Confidence levels in performing CBE, however, were not related to the actual performances of CBE. Students' knowledge of breast cancer was modestly related to their CBE scores. Finally, the students with one extra training session with a-standardized patients performed significantly, better on the CBE. Conclusions. It is important to directly measure students' breast cancer screening and physical examination skills because students' reported self-confidence for some skills might not accurately reflect their actual levels of performance. Additionally, while knowledge may be an important step to acquiring skills, knowledge may not be a good indicator of students' performances of risk assessment, counseling, or physical examination. Although confidence in counseling was related to counseling performance, the stability of this perception and how that translates into future cancer-control practices remain in question.