Decades of peer-reviewed research, including large-scale studies spanning tens of thousands of participants across hundreds of organizations and institutions, establish structured peer review as one of the most powerful, scalable, and cost-effective tools available to learning and development programs.
The most actionable finding in this body of research is that the person providing feedback learns more than the person receiving it. When peers review a colleague's work, they are incentivized to evaluate, think critically about criteria for success, diagnose problems, and explain their reasoning. This is what researchers call constructive learning, a more cognitively demanding and durable form of skill-building than passively receiving critique. In a meta-analysis of nearly 21,000 participants across 76 institutions, learning gains were found to be more closely tied to providing feedback than to receiving it.
→ L&D Implication: The cognitive work of reviewing produces measurable skill development in the reviewer, independent of any benefit to the person receiving the feedback.
Yu & Schunn (2023), Computers in Human Behavior; Wu & Schunn (2021), American Educational Research Journal; Zong, Schunn & Wang (2021), Computers in Human Behavior
A persistent concern about peer review is quality: can non-experts be trusted to evaluate work accurately? Research consistently answers yes, when the process is properly structured. Meta-analyses find an average correlation of 0.63 between peer ratings and expert ratings, with many well-designed studies reporting correlations between 0.70 and 0.91. Critically, the key to reliability is aggregation: when multiple peers assess the same work, their collective judgment closely tracks expert judgment. This mirrors how high-stakes professional evaluation works in grant review panels, 360-degree performance assessments, or editorial boards.
→ L&D Implication: Multi-peer assessment with structured rubrics is a credible, scalable alternative or complement to expert review.
Xiong et al., Computers in Human Behavior; Cho, Schunn & Wilson (2006), American Educational Research Journal; Li et al. (2016) meta-analysis
When the same issue is flagged by multiple colleagues independently, peers are substantially more likely to act on it. Research shows that feedback frequency is one of the strongest predictors of whether feedback is implemented. Agreement across reviewers functions as a social signal: it validates that a problem is real, reduces the ability to dismiss a single voice, and increases the perceived urgency of change. A controlled study found that employees who revised work based on feedback from a group of multiple peers showed greater improvement than those who received feedback from a single expert.
→ L&D Implication: Structuring programs so each piece of work receives feedback from multiple reviewers significantly increases the probability that feedback will actually change behavior.
Wu & Schunn (2020), Contemporary Educational Psychology; Patchan, Charney & Schunn (2009), Journal of Writing Research; Cho & Schunn (2007)
Multiple meta-analyses confirm that structured peer review produces measurable improvements in performance that transfer to new tasks and contexts over time. It strengthens the essential soft skills that define high-performing professionals, including critical thinking and the ability to give and receive clear, constructive feedback. These capabilities often distinguish employees who grow from those who plateau.
→ L&D Implication: Peer review is not a workaround for limited trainer capacity. It is a learning method that produces superior outcomes by engaging the full workforce as active developers of each other's capability.
Yu & Schunn (2023); Li et al. (2020) meta-analysis, Assessment & Evaluation in Higher Education; Li et al. (2021) meta-analysis, Applied Measurement in Education; Double et al. (2020)
Research conducted at the Learning Research & Development Center, University of Pittsburgh
All citations available upon request · © 2026 Panther Learning Systems Inc.
