Improving the effectiveness of peer feedback for learning

Gielen, S., Peeters, E., Dochy, F., Onghena, P., & Struyven, K. (2010). Improving the effectiveness of peer feedback for learning. Learning and Instruction, 20(4), 304-315.

Background

The authors highlighted that there had been strong interest in research on formative assessment, especially on peer assessment as a tool for learning. However, to enhance the potential impact of peer assessment on learning, they emphasized the importance of understanding which mechanisms affected learning, and how these mechanisms could be supported.

Aims

The study examined the effectiveness of (a) peer feedback for learning, and (b) a particular instructional intervention to support the use of feedback for writing.

Sample

The participants were 43 seventh graders from two classes of the same school, taught by the same teacher.

Method

The data comprised two types of feedback conditions – the peer feedback with reply (PEERFB-REPLY) condition and peer feedback without reply (PEERFB) condition, and the students’ writing assignments were collected as pre- and post-tests. The participants were randomly assigned to the conditions. In the PEERFB-REPLY condition, students were asked to report to the teacher which comments they considered, how they dealt with them, what they learned from providing peer feedback, and to reflect on their own accomplishments. The PEERFB condition was the purely peer-feedback condition. For both conditions, the feedback forms’ structure, guiding prompts and the rationale for peer feedback were explained. The teacher then modelled the peer assessment process and formulated the assessment criteria together with the students. The extent to which the feedback was constructive was conceptualized through ‘quality criteria’. The quality criteria were appropriateness to the assessment criteria, specificity, presence of justifications and suggestions for improvement, and clear formulation. The feedback that the students were expected to provide corresponded to the scoring rubric for the pre- and post-tests.

Findings

The results indicated that the PEERFB-REPLY condition had no significant effect on writing performance as compared to the PEERFB condition, when controlling for pre-test performance. However, the presence of justifications significantly enhanced writing performance. The effect of ‘justifications’ demonstrated the informative value of feedback. Other quality criteria, such as appropriateness, specificity, clear formulation, and presence of suggestions, also did not have a significant impact on performance.

Implications

The authors highlighted that ‘justifications’ that produced the greatest impact on writing performance were, however, among the most difficult to teach. The prompts that were provided might not have been sufficient to generate justification, as reflected in in its low occurrence in the study. This implied a need for peer assessment training, guidance and quality control so that students could provide justifications more frequently.

Original article retrieved from Taylor & Francis.