The impact of peer-assessment on self-regulated learning : pairs vs. groups

Abstract

The modern demands of life have created increased demands for not only the ability to self-regulate but also the aptitude to successfully negotiate with peers. Peer-assessment can be one such pedagogical approach to this ever-growing societal need. With this in mind, this study investigated the effects of peer-assessment constellation size on the outcomes of self-regulation. Furthermore, it looked to understand the causal factors behind any similarities or differences between the two chosen peer-assessment constellations of pairs and groups of three. While exploring the above points, this study also hoped to provide additional evidence regarding peer-assessment’s effectiveness at improving student self-regulation, especially compared to self-assessment. A quasi-experimental model was used in this research. Five classes of Hong Kong tertiary students in their second year of study, were selected for this research; a total of 113 students. The classes were assigned into three categories: Pairs, Groups, and Comparison (self-assessment). A three-week pre-test, intervention, and post-test model was followed. Students were filmed during the intervention, with the aim of allowing insights into the varying dynamics of each constellation. Lastly, students were invited for short follow-up interviews after the three week procedure. This study found that both peer-assessment treatments were more effective at enhancing self-regulation compared to self-assessment. The two different peer-assessment constellation sizes were found to only produce a minor practical variation in outcome between them. It was also observed that students in both peer-assessment orientations displayed similar levels of communication and largely held similar views of the peer-assessment process. However, the interaction pattern of the two constellations was markedly different, with students in groups demonstrating a much more positive dynamic. The results indicate that peer-assessment in either of the tested forms can significantly increase student self-regulation. It is also suggested that while there are many similarities between the pairs and groups there are key differences in the interaction patterns of students in each constellation. While the immediate outcome of the two peer-assessment constellations appears similar, the observed difference in interaction pattern warrants further investigation, particularly for it long term effects on attainment, motivation and other potentially important factors.