Purpose: To compare in-training evaluations of residents by their peers with evaluations by faculty preceptors in an internal medicine residency.
Method: The study group consisted of 22 residents enrolled in the core (three-year) internal medicine program at the University of Calgary in 1989-90 and 1990-91. At the end of each rotation, ratings of the residents were requested from faculty preceptors and from peers for several categories of clinical competence. The peer ratings were paired with faculty ratings, for a total of 74 pairs of ratings. The Wilcoxon matched-pair signed-rank procedure was used to compare the paired ratings. One-way analysis of variance was used to compare the peer and faculty ratings with the residents' scores on three other kinds of evaluation used by the residency.
Results: While there was no significant difference between peer and faculty ratings for overall competence or for several components of competence, there were significant differences for some components, with faculty tending to rate higher than peers. The latter components were physical examination, team relationships, industriousness and enthusiasm, teaching, physician-patient relationships, and case presentations. External validation of the ratings by comparing them with other kinds of evaluation yielded little meaningful information.
Conclusion: That the faculty ratings were significantly higher than the peer ratings for some components of clinical competence suggests that there were differences in the quality of evaluation between the peers and faculty, or differences in the standards or expectations of the two groups.