Council on Medical Student Education in Pediatrics


Search This Site

COMSEP 2007 San Antonio Meeting

Poster Presentation:

Does the number of potential responses in clinical performance rating scales affect student grades?


 PG McBurney; AV Blue; WTBasco Medical University of South Carolina, Charleston, South Carolina

 Background: Clinical performance ratings are standard in evaluation of medical students. We do not know the full affect of changes to the rating form.

Objective: Primarily, to compare mean clinical grades for pediatric clerkship students in 2004-05 (grading form with 4 point scale) to the grades in 2005-06 (7 point scale).

Methods: In both academic years, College of Medicine (COM) grading forms had 10 items (eg, professionalism). The 10 items each had either a 4 point or a 7 point scale. The student's assigned ratings for the 10 items were averaged per form. COM provided a conversion in 2005-06 for the 7 point scale back to a 4 point scale for the student's final grade. The clinical grade consisted of ratings from multiple attendings on general wards, subspecialty wards, nursery, and outpatient rotations. Means and ranges for grades were calculated and compared for the 2 academic years. To account for class capacity, we compared shelf scores. Further, we compared percentages of high ratings (>3.5) between years. T-tests and chi-square were used.

Results: 138 students completed all clerkship requirements in 2004-05, 129 in 2005-06. In 2004-05 the mean clinical grade was 3.3 (standard deviation .17; range 2.8-3.7); in 2005-06, mean was 3.3 (.16; 2.9-3.7) (p=.8). Student capacity was the same by the shelf scores means: 2004-05, 3.0; in 2005-06, 2.8 (p=.06). In 2004-05, 24 students earned a clinical grade >3.5; in 2005-06, 27 (p=.09).

Conclusion: Scale size appears to have little effect on final clinical grades.