In Defense of Medical School Grades
With residency application season getting underway, I've been thinking a lot about medical school grades, and why I wish we'd bring them back. As someone who reviews applications and interviews candidates for residency, I feel that we’re doing a disservice to students by removing all comparative performance benchmarks from residency applications.
First it was preclinical grades, which admittedly makes sense, as performance in preclinical biochemistry is probably not a reliable predictor of residency performance. But then it was Step 1 scores. The purported reason was to reduce medical student stress — an aim I whole-heartedly support — but that simply kicked the evaluative can down the road to Step 2. When Step 1 was scored, if a student did not do as well as they hoped, they had a second opportunity to demonstrate their mastery of medical material on Step 2. With pass/fail Step 1, students only have one shot to showcase their abilities in an objective, comparative way.
Since COVID, the new trend has been removing clinical clerkship grades. In lieu of a traditional transcript, schools produce lengthy Dean’s Letters that contain summaries of and excerpts from students’ clerkship evaluations. While the theory is that this allows for a more holistic review, clerkship evaluations are subject to the same linguistic arms race that plagues letters of recommendation. If every student is “excellent” or “outstanding,” and there is no grade to provide a more objective, comparative assessment, it is nearly impossible for an application reviewer to determine how a student actually performed in medical school.
This concern over the purge of comparative metrics in residency applications also applies to schools that have grades but fail to use them in a meaningful manner. If an applicant receives all honors on their clinical clerkships, but the grade distribution for the school shows that 80-90% of students receive honors, I’m not sure how to assess that students performance — all I know is they weren’t in the bottom 10% of their class, which isn’t particularly helpful when trying to distinguish between many impressive, immensely qualified applicants.
I’m all for reducing student stress and encouraging a more comprehensive approach to residency application review, but those efforts may be hurting medical students in the long run. Without more objective tools, I fear reviewers will focus on other, less meaningful heuristics, like the prestige of the applicant’s medical school or college. Utilizing more comparative metrics would be a win-win — applications would be easier to review residencies and medical students would have more opportunities to clearly demonstrate their academic achievement.