Across 2,421 students in 13 courses, Zong, Schunn, and Wang isolated which parts of the peer-feedback experience actually predict performance growth: comment depth over comment count, and providing feedback over receiving it.
Zong, Z., Schunn, C. D., & Wang, Y. (2021). "What aspects of online peer feedback robustly predict growth in students’ task performance?" Computers in Human Behavior, 124, 106924.
Peer feedback has several moving parts: how many comments a person gives or gets, how deep those comments are, how helpful they're rated, and whether the person is giving feedback or receiving it. Zong, Schunn, and Wang set out to test all of these against each other in the same dataset, to find out which specific aspects robustly predict growth in task performance and which ones are noise.
The study drew on 2,421 students across 13 different courses, seven universities, and six content disciplines. Using temporally lagged multiple-regression analysis, the researchers tested the unique contributions of comment quantity, comment depth, and comment quality — for both received and provided comments — to each student's growth in task performance across assignments.
Comment depth predicted performance growth far more strongly than comment quantity did; for received comments, the number of comments alone carried little predictive weight on its own. Providing comments contributed more to growth than receiving them did, and depth of the provided comment mattered more than how many comments a person wrote. Comment helpfulness ratings predicted the learning value of both provided and received comments. The study also found substantial variation in effect sizes across the 13 courses, meaning implementation details — not just whether peer review happens at all — determine how much it helps.
Two things follow directly for how a peer-review program should be designed. First, more comments is not the goal: prompts and rubrics should push reviewers toward depth — specific, detailed evaluation — over volume. Second, program design should weight toward every employee providing feedback, not just receiving it, since that's the side of the exchange this study ties most strongly to growth. Peerceptiv's reviewer prompts are built around this exact distinction, scaffolding reviewers toward depth rather than treating a longer comment count as success.
Chris Schunn co-authored this study isolating which specific aspects of peer feedback drive measurable performance growth. He is a Professor of Psychology, Learning Sciences and Policy, and Intelligent Systems at the University of Pittsburgh and a Senior Scientist at the University's Learning Research and Development Center (LRDC), where he has directed research projects backed by more than $80M in federal grants. As Peerceptiv's Chief Learning Scientist, his research directly shapes how the platform structures reviewer prompts, rubrics, and feedback workflows.
