This review of methods used to assess the clinical performance of medical students focuses on four common assessment approaches: (1) the examination developed by the National Board of Medical Examiners (NBME); (2) systematic, multifactor evaluation methods; (3) observation techniques; and (4) problem based methods. Analyzed in conjunction with each approach were reliability and validity data as well as practicality of the assessment approaches. The reliability and validity data are extensive and high for the NBME. The NBME is the least complicated instrument to administer and score. However, it cannot assess client-clinician interactions utilizing live subjects. Reliability and validity data on observation methods are sparse; and where data exists, the coefficients are generally low. Multifaceted evaluation techniques have provided more accurate assessments of student competence, but require more time and more people to administer multiple assessments. A final issue related to the assessment of clinical competence involves the determination of a generally acceptable definition of competence. (Author/BW)