A Tekian1, Y Han, L Hruska, A J Krainik. 1. Department of Medical Education, University of Illinois at Chicago, College of Medicine, 808 South Wood Street 986 CME, Chicago, IL 60612-7309, USA. tekian@uic.edu
Abstract
BACKGROUND: In medical education, examinations must assess a logical progression toward problem-solving skills. Differences in cognitive development between underrepresented minority students (URMs) and non-URMs may affect examination performance and subsequent attrition rates. PURPOSE: The authors investigated URM and non-URM performances by retrospectively analyzing success rates on exam items of differing cognitive demand. METHOD: Mean correct responses to exam items classified as Recall, Interpretation, or Problem-Solving questions were calculated. Both URM and non-URM groups were stratified by grade point average (GPA) and scores on the Medical College Admission Test (MCAT). Differences were investigated with analysis of variance and general linear models. RESULTS: For all students, performance levels decreased as the cognitive demands of the exam items increased. When stratified by GPA and MCAT score, several important differences were found between URM and non-URM performance. CONCLUSIONS: Because cognitive measures fail to account for the majority of performance differences, noncognitive attributes must contribute to the poorer performance of URMs.
BACKGROUND: In medical education, examinations must assess a logical progression toward problem-solving skills. Differences in cognitive development between underrepresented minority students (URMs) and non-URMs may affect examination performance and subsequent attrition rates. PURPOSE: The authors investigated URM and non-URM performances by retrospectively analyzing success rates on exam items of differing cognitive demand. METHOD: Mean correct responses to exam items classified as Recall, Interpretation, or Problem-Solving questions were calculated. Both URM and non-URM groups were stratified by grade point average (GPA) and scores on the Medical College Admission Test (MCAT). Differences were investigated with analysis of variance and general linear models. RESULTS: For all students, performance levels decreased as the cognitive demands of the exam items increased. When stratified by GPA and MCAT score, several important differences were found between URM and non-URM performance. CONCLUSIONS: Because cognitive measures fail to account for the majority of performance differences, noncognitive attributes must contribute to the poorer performance of URMs.