Does the ability of medical students to assess and relate to their patients through the psychiatric interview reflect in traditional academic achievement scores? A previous publication by our group offers data suggesting that certain personality traits of final year medical students have a fundamental importance in determining a successful performance on a psychiatry rotation. As the literature is vague and contradictory about the relationship between the psychiatric interview and clinical performance, a study was developed to determine what outcome variables are associated with the performance of clinical clerks on the Psychiatric Interview. An attempt was made to develop and test a new scale designed to measure the Psychiatric Interview. The Psychiatric Interview Scale was used to measure performance on 33 final year medical students on a 1 month psychiatry rotation at a University of Toronto teaching hospital, and subsequently, as a basis for analysis and training of these students in interviewing. Inter-rater reliability of the Psychiatric Interview Scale was rho = 0.72 and internal reliability was 0.74. Psychiatric Interview Scores were compared with other achievement scores, and with patients' ratings of the students interviews. A significant correlation emerged between the Psychiatric interview scores and final clinical examinations, but surprisingly, there was no relationship between the patients' evaluation of the Psychiatric Interview and the raters' evaluations and a strongly significant negative correlation between patients' ratings and clinical scores.(ABSTRACT TRUNCATED AT 250 WORDS)