PURPOSE: To determine whether medical school admission interviewers change their evaluations and impressions of applicants as a direct result of the interview. METHOD: In 1991-92, 419 applicants to the University of Virginia School of Medicine were interviewed by members of the admission committee in two separate half-hour sessions. After reviewing each applicant's folder, interviewers rated the applicant before the interview on six objective scales. After the interview, ratings were again made on the same six scales, on the same form, below the ratings made before the interview. Data were examined using paired t-tests, Pearson correlations, and stepwise multiple-regression analysis. RESULTS: Of the six scales, only the ratings of Commitment to Serve Others were not significantly changed by the interview; the ratings of Familiarity with Issues in Medicine changed the most (p < .01 by paired t-test). The ratings of Overall Impression increased for accepted applicants and decreased for rejected applicants. CONCLUSION: The interview did influence interviewers' ratings made before the interview, and in the direction consistent with admission decisions, which supports the continued use of the interview. Although the magnitude of the changes was not large, the changes validate the conviction that the interview aids in the selection of individuals for medical school.