The Impact of Peer Assessment on Academic Performance: A Meta-analysis of Control Group Studies

TLDR

Peer assessment has been studied for decades and widely advocated for classroom use, yet most syntheses have been narrative reviews. This study conducts a meta‑analysis of 54 experimental and quasi‑experimental studies (k = 141) to evaluate the impact of peer assessment on academic performance across primary, secondary, and tertiary levels. Meta‑regressions examined how feedback and educational characteristics (e.g., online vs offline, frequency, education level) moderate the effect. The analysis found a small‑to‑medium overall effect (g = 0.31), with peer assessment outperforming no assessment and teacher assessment but not differing significantly from self‑assessment, and the effect was robust across diverse contexts, supporting its use as a formative practice.

Abstract

Abstract Peer assessment has been the subject of considerable research interest over the last three decades, with numerous educational researchers advocating for the integration of peer assessment into schools and instructional practice. Research synthesis in this area has, however, largely relied on narrative reviews to evaluate the efficacy of peer assessment. Here, we present a meta-analysis (54 studies, k = 141) of experimental and quasi-experimental studies that evaluated the effect of peer assessment on academic performance in primary, secondary, or tertiary students across subjects and domains. An overall small to medium effect of peer assessment on academic performance was found ( g = 0.31, p < .001). The results suggest that peer assessment improves academic performance compared with no assessment ( g = 0.31, p = .004) and teacher assessment ( g = 0.28, p = .007), but was not significantly different in its effect from self-assessment ( g = 0.23, p = .209). Additionally, meta-regressions examined the moderating effects of several feedback and educational characteristics (e.g., online vs offline, frequency, education level). Results suggested that the effectiveness of peer assessment was remarkably robust across a wide range of contexts. These findings provide support for peer assessment as a formative practice and suggest several implications for the implementation of peer assessment into the classroom.