Effects of online peer assessment on higher‐order thinking: A meta‐analysis

Abstract Online peer assessment (OPA) has been increasingly adopted to develop students' higher‐order thinking (HOT). However, there has not been a synthesis of research findings on its effects. To fill this gap, 17 papers (published from 2000 to 2022) that reported either a comparison between a group using OPA ( n = 7; k = 22) and a control group or a pre–post comparison ( n = 10; k = 17) were reviewed in this meta‐analysis. The overall effect of OPA on HOT was significant ( g = 0.76). Furthermore, OPA exerted more significant effects on convergent HOT (eg, critical thinking, reasoning and reflective thinking; g = 0.97) than on divergent HOT (eg, creativity and problem‐solving; g = 0.38). Reciprocal roles and anonymity were found to positively moderate the impacts of OPA on HOT, although their moderating effects were not statistically significant because of small sample size of studies in the analysis. The results of the meta‐analysis reinforce the arguments for regarding OPA as a powerful learning tool to facilitate students' HOT development and reveal important factors that should be considered when adopting OPA to enhance students' HOT. Practitioner notes What is already known about this topic Online peer assessment (OPA) has significant positive impacts on learning achievement. OPA has been regarded as a potential approach to cultivating students' higher‐order thinking (HOT) but has not been proved by meta‐analysis. OPA should be carefully designed to maximise its effectiveness on learning. What this paper adds OPA has been proved to significantly positively influence students' HOT via meta‐analysis. OPA exerted more significant effects on convergent HOT than on divergent HOT. The potential of reciprocal roles and anonymity for moderating the impacts of OPA on HOT should not be underestimated. Implications for practice and/or policy OPA could be a wise choice for practitioners when they help students to achieve a balanced development of HOT dispositions and skills. Students' divergent HOT can be encouraged in their uptake of peer feedback and by allowing them autonomy in deciding assessment criteria. OPA with design elements of reciprocal roles and anonymity has great potential to promote students' HOT.