Council on Medical Student Education in Pediatrics

COMSEP Logo

Search This Site

Journal Club

Boehler M, Rogers DA, Schwind CJ et al., An investigation of medical student reactions to feedback: a randomized controlled trial. Medical Education, August 2006; 40(8):pp. 746-749.


Boehler M, Rogers DA, Schwind CJ et al.,

An investigation of medical student reactions to feedback: a randomized controlled trial. Medical Education, August 2006; 40(8):pp. 746-749.

What is the problem and what is known about it so far?
As pediatric clerkship directors, we all have "hawks" and the "doves" among our faculty and we spend much of our time trying to convince both of the value of feedback to students. We all recognize the value of feedback, but most of us struggle with the question: What is the best format for that feedback?

Why did the researchers do this particular study?
In this simple but well-designed study, the authors measured the differences in student performance and satisfaction in two randomized groups: one group received specific feedback directed at deficiencies while the other experienced non-specific praise.

Who was studied?
Thirty-three M1 and M2 students participated.

How was the study done?
The study focused on a simple motor task: the two-handed surgical square knot. All student attempts were videotaped. Study participants were first videotaped tying a knot with no prior instruction (the pre-test). An academic surgeon then demonstrated how to correctly tie the knot. Students were then taped a second time (post-instruction, pre-intervention). Half the students were then assigned to the generic compliments group. During their third knot-tying exercise, these students heard scripted praise such as "great job" or "outstanding." The other students were assigned to the performance-based feedback group. These students were given 2 specific instructions addressing their deficiencies while tying the third knot. Students were then taped tying a 4th knot. At the conclusion of the session, students were asked to rate their satisfaction with the experience. All videos were reviewed in random order by 3 blinded reviewers who rated performance using a validated instrument.

What did the researchers find?
The feedback group performed better but students were was less satisfied with the experience. The generic compliment group did not improve their performance but were more satisfied with the experience.

What were the limitations of the study?
The study looked at a simple motor task and included a very small number of only M2 and M3 students at a single medical school. Feedback sessions were very brief; this might have contributed to student dissatisfaction with the feedback.

What were the implications of the study?
Individualized feedback improves performance. When evaluating education systems or programs, performance improvement is a better measure of the effectiveness of feedback than learner satisfaction. Praise does not improve performance, but it does lead to more satisfied learners. This study lends credibility to the "sandwich" technique of giving feedback: praise / constructive criticism / praise.

(Editorial comment: This is a fascinating small study that confirms what many have felt for a long time; students crave specific formative feedback but also don't like bad news. Educators have to take solace in fact that while the faculty or teaching session student evaluation could take a hit, specific feedback will improve student performance-which is the ultimate goal. Bill Raszka)

Return to Journal Club