Council on Medical Student Education in Pediatrics


Search This Site

COMSEP 2010 Albuquerque Meeting


Jared P. Austin, MD; Chad K. Brands, MD; Robert G. Voigt, MD, Mayo Pediatric Residency Program, Mayo Clinic, Rochester, MN; Sophie P. Gladding, PhD, Center for Medical Education Research, Cleveland Clinic, Cleveland, OH


Background: Third year clerkships are a time of rapidly expanding clinical knowledge and skills for medical students.  Many different components are used to evaluate their progress, including subjective evaluations by residents and faculty and objective measures via the National Board of Medical Examiners (NBME) subject, or shelf exams.  However, few studies have been done to evaluate the correlation between these measures.  We hypothesized that students who perform better on shelf exams will receive higher evaluations from faculty and residents. Objective: To determine the correlation between subjective and objective evaluations of medical students on their pediatric clerkships. Methods:  Medical students who completed a pediatric clerkship between July 1, 2008 and June 30, 2009 were enrolled in the study.  Subjective data, consisting of resident and faculty evaluations of medical students, were obtained through the Mayo Integrated Scheduling and Evaluation System.  This electronic database uses a 5-point Likert scale for all evaluations (1=needs improvement, 5=exceeds expectations).  The NBME pediatric shelf exam was administered at the end of each 6-week clerkship.  Exam percentile scores, based on national standards, were obtained from the pediatric clerkship director. Results:  A total of 39 medical students were enrolled in the study.  Three were excluded due to insufficient data.  There was a moderate correlation between resident evaluations and exam scores (r=.401, p=.008), and between composite evaluations (faculty plus residents) and exam scores (r=.458, p=.002).  There was a much weaker correlation between faculty evaluations and exam scores (r=.255, p=.067), and between resident and faculty evaluations (r=.123, p=.237). Discussion:  Overall, subjective evaluations moderately predicted objective evaluations.  However, resident evaluations were a better predictor of medical student performance on the shelf exam than faculty evaluations.  This may be due, in part, to the fact that residents spend much more time with medical students than faculty, and hence, are able to more accurately evaluate their clinical skills and fund of knowledge.  It may also reflect basic differences in how faculty and residents evaluate students.  Given these results, it is important that programs continue to use a multifaceted approach in assessing performance.