The systems used by the medical schools for evaluation of 106 applicants for 1984-85 pediatric level 1 (PL-1) residency positions at the Duke University Medical Center were studied by the authors to determine the possible relationship between contents of the applications and the rankings of these applicants for the National Resident Matching Program (NRMP). In the 56 U.S. medical schools studied, the grading systems of 24 medical schools used three passing grades (usually A, B, C), 14 used two passing grades (usually honors and pass), six used one passing grade (pass) and seven used a mixed system or number grades. The grading systems of five schools could not be discerned. Ten schools provided the class distribution of the awarded grades, and 23 identified the relative class standing of the applicant. Twenty-seven schools disclosed the position of the applicant in the class through grade distribution data or class standing; the 29 other schools did not make such information known. Students whose schools had identified their class positions achieved higher rankings from the residency selection committee than did students from schools which did not, in spite of the fact that schools which did not disclose the student's comparative positions were categorized as being more "distinguished."