Peer feedback quality matters for post-feedback outcomes: A meta-analysis

Peer feedback quality is key to enhancing learning through peer assessment. However, varied operationalizations of feedback quality across studies have led to inconsistencies in the literature. There are also contradictory findings regarding the relationship between feedback quality and students' use of feedback in current and future performance tasks. In this meta-analysis, we described peer feedback quality across five dimensions (i.e., amount, content, accuracy, features, and perceptions) and its relationship with other related feedback elements, drawing on recent conceptual frameworks and an integrative model of peer feedback. We then aggregated studies that examined the correlation between multifaceted peer feedback quality and post-feedback outcomes (i.e., revision, performance, learning, and course achievement) and analyzed potential moderating effects of several variables. Following a systematic literature search, 49 eligible papers with 53 independent samples and 241 effect sizes were identified. Using the robust variance estimation method, the synthesis of these effect sizes revealed a significant, positive correlation between peer feedback quality and post-feedback outcomes (r = .20). Moderator analyses showed that the strength of the correlations varied across post-feedback outcomes. Furthermore, post-feedback outcomes were more strongly associated with the quality of feedback students provided, rather than the quality of the feedback they received. Additionally, giving training prior to peer assessment strengthened the correlation between peer feedback quality and post-feedback outcomes. Overall, this meta-analysis corroborates the relationship between peer feedback quality and students’ use of feedback and highlights the importance of operationalizations, training, and peer feedback roles in this connection.