Optimising the Effectiveness of Peer Feedback in Higher Education
Priyanka Pereira is a PhD student in the Department ELAN Teacher Development. (Co)Promotors are prof.dr. K. Schildkamp, prof.dr.ir. B.P. Veldkamp and dr. M.C. Heitink from the Faculty of Behavioural, Management and Social Sciences.
Higher education institutions have increasingly been replacing or complementing teacher feedback with peer feedback to involve students more actively in feedback provision and receipt. One of the reasons for this change may be that the use of peer feedback has been shown to have several benefits for students. Since peer feedback seems to result in the improvement of students’ academic as well as non-academic skills, it may be an effective alternative to teacher feedback. Another reason for the change may be that there has been an increasing number of students enrolled in higher education institutions, but these institutions have been unable to proportionately increase the number of teachers, resulting in an increase in the student to staff ratio. The increased ratio has made it difficult for teachers to dedicate sufficient time to providing each student with individual feedback, and this has necessitated finding alternatives to teacher feedback. Since peer feedback crowdsources the feedback process, it may be an efficient alternative to teacher feedback.
With the increased use of peer feedback in higher education, we should ensure that implementation of peer feedback activities is beneficial to students’ academic achievement and should determine how the benefits of these activities can be optimized. Unfortunately, the existing literature does not provide us with a definite answer to whether peer feedback improves students’ academic achievement, and if it does, which factors influence this improvement. Therefore, in the doctoral research, we aimed to answer these questions.
Similar to any educational intervention, the factors affecting the effectiveness of a peer feedback intervention could include the characteristics of the intervention, the teacher, the students, and the social and contextual environment. In the dissertation, we focused on the intervention characteristics and the student characteristics, because there is potential for teachers to manipulate these characteristics at the grassroots level, as a first step towards optimization of the effect of peer feedback. Manipulation of teacher characteristics and environment characteristics might require measures at a higher level, for example, in the higher education institution, which might be more appropriate and effective once the basics are in place.
This led to the research questions “What is the effect of peer feedback on higher education students’ academic achievement?” and “How can intervention characteristics and student characteristics influence the effect of peer feedback on higher education students’ academic achievement?”, which formed the basis of the dissertation.
First, we aimed to determine the effect of a peer feedback intervention on the academic achievement of students in the STEM fields of higher education. Based on a meta-analysis of a large number of studies, we found that peer feedback significantly improves this achievement. This finding makes a valuable contribution to the knowledge base on whether peer feedback is beneficial for academic achievement in the STEM fields of higher education. It means that policymakers and practitioners can expect that asking their students to participate in peer feedback activities will improve their students’ academic achievement, and therefore, they can confidently initiate or continue the implementation of peer feedback in their programs.
Next, we aimed to find how peer feedback intervention characteristics moderate the effect of a peer feedback intervention on the academic achievement of students in the STEM fields of higher education. Moderator analyses did not provide evidence of any peer feedback intervention characteristics that influenced this effect. This means that recommendations for the optimization of the effectiveness of a peer feedback activity could not be made on the basis of the design of the activity. Although the analyses suggested that the effect of peer feedback remains consistent even with different intervention characteristics, it does not rule out the possibility that (at least some of) these characteristics can make a difference (Borenstein et al., 2009; Harrer et al., 2021). A true lack of a moderating effect is possible; however, the insignificant result from a moderator analysis is also possible due to the low power of the test to detect the moderating effect. Further research can contribute to the knowledge base on how the use of peer feedback in higher education can be made more effective, thus providing researchers, policymakers, and practitioners with the information they need to design and implement peer feedback interventions for maximal effectiveness.
Then, we aimed to discover how student characteristics, specifically motivation and academic ability, influence the effectiveness of a peer feedback activity. To do this, we started by using methodological pluralism to capture the complex concept of students’ motivation in a peer feedback context, and we developed the M-PerFeCt questionnaire to allow for a comprehensive assessment of this motivation. The questionnaire provides theoretical insight into the underlying structure of this motivation, including a motivation to learn scale and a motivation to participate scale. Aside from being of theoretical relevance, the questionnaire is also of practical value. This is the first questionnaire that allows for the reliable and valid assessment of students’ motivation in a peer feedback context. It can provide teachers with an easy way to assess their students’ motivation before implementing a peer feedback intervention, so that teachers can take the necessary measures to improve the specific aspects of motivation that are low, before implementing the intervention, in order to maximize students’ use of the intervention. It can also enable researchers to assess students’ motivation before and after participating in a peer feedback activity, such that the data can be used not only by itself to answer research questions regarding students’ motivation in the context of peer feedback, but also in combination with other data to answer research questions regarding the relationship between students’ motivation and other variables.
Once the M-PerFeCt questionnaire was available, we looked into how students’ motivation in a peer feedback context and their academic ability are related to their perception and use of the peer feedback they receive. We found that students’ motivation in a peer feedback context predicts their perception and use of the received peer feedback, and their academic ability does not predict their perception and use of the received peer feedback. These findings contribute to the theoretical knowledge base on the effectiveness of peer feedback activities. The research provides a comprehensive understanding of the relationship between each aspect of students’ motivation in a peer feedback context and their perception and use of the received peer feedback. It highlights the aspects of motivation that are the most important to the perception and use of the received peer feedback, and because both the pre-activity and post-activity motivation are considered, it also provides insight into the aspects of motivation that gain or lose importance from before to after participation in the activity. In practice, if teachers find that their students’ perception and/or use of received peer feedback is low, it is advantageous for them to enhance their students’ perception and use of received peer feedback. Our research shows that the measures could include improving students’ motivation, using strategies that target the specific aspects that are found to be most important.
To start developing the strategies to improve students’ motivation in a peer feedback context, we set out to first identify the motivational profiles of higher education students in a peer feedback context, and then develop strategies teachers can use to improve the motivation of higher education students belonging to each of the identified (and possibly other) profiles. Our research based on responses to the M-PerFeCt questionnaire revealed five motivational profiles of higher education students in a peer feedback context. The profiles contribute to the understanding of students’ motivation in a peer feedback context, namely that students can be differentiated based on their overall level of this motivation.
Further, during focus groups of higher education teachers who had used peer feedback in their courses and during a consultation with a peer feedback expert, there was an overwhelming consensus on the benefit of optimizing students’ motivation prior to them participating in a peer feedback activity. The teachers and the expert suggested a number of strategies for improving students’ motivation, which focused on expectancy and value constructs, in line with most strategies to improve student motivation that are found in the literature. The suggested strategies can function as a practical guide for higher education teachers who are planning to implement a peer feedback activity in their course and who want to maximize their students’ motivation to participate in the activity.
Our research showed that students’ motivation influenced their perception and use of received peer feedback. However, the teachers and the peer feedback expert unanimously recommended that the same strategies be used to improve the motivation of all students, irrespective of their motivational profiles. Given the discrepancy between the results and the small scale of the studies, further investigation into the need and/or possibility of using different strategies for students with different motivational profiles is warranted.
