Daugherty, S.R.; Nora, L.M.; Schmidt, J.L.; Goodman, L.J. Identifying Poor Preclinical Performers Who Do Well in Clerkships, Academic Medicine, 67;S72-S74, 1992.

PURPOSE: The authors' goals were to identify those students who performed poorly preclinically but excelled in their clerkships and to attempt to discover possible generalities for this population.

METHODS: Data were drawn from 523 students from five successive classes at Rush Medical College. Clinical performance was assessed by a composite that was formed by weighting grades from the seven core clerkships by length of experience. Preclinical performance was assessed by forming a composite score analogous to a computation of a weighted grade-point average. To obtain this average, courses were weighted by the number of contact hours. The final composite scale for both preclinical and low clinical performance was standardized to range from a high of 100 to a low of 1. Each performance index was split at the median to provide a group of "high" performers and a group of "low" performers. Combining the groupings derived from these median splits created a four-fold classification: Group1, students who had low preclinical and clinical scores; Group 2, students with high preclinical scores and low clinical scores; Group 3, students with low preclinical scores and high clinical scores; and Group 4, students with both high preclinical and clinical scores. Group 3 was the focus of this study. Independent variables included MCAT scores, GPAs, admission committee ratings, NBME Part I and II scores, whether English was a second language, and various demographic measures. Analysis of variance techniques were used to compare performance across groups.

RESULTS: The authors reported that while preclinical grades were generally predictive of clinical clerkship performance, they tended to be less predictive with Group 3 students. In comparison with the other groups, Group 3 students were less likely to speak English as a second language and more likely to be females. They also scored higher on theVerbal Reasoning section of the MCAT than those students who performed poorly both preclinically and clinically (Group 1).

CONCLUSIONS: The authors suggested that Verbal Reasoning scores may have some predictive value in determining which students performing poorly preclinically will later improve during clerkships. Group 3 consisted of students who, for various reasons, simply performed poorly on standardized examinations. The authors hypothesized that such students may perform better preclinically in a problem-based curriculum, since evaluations would be based on a broader range of criteria rather than on examination proficiency.

Contact Us    © 1995-2008 AAMC    Terms and Conditions    Privacy Statement