The nature of feedback: How peer feedback features affect students’ implementation rate and quality of revisions.

Abstract

Although feedback is often seen as a critical component of the learning process, many open questions about how specific feedback features contribute to the effectiveness of feedback remain-especially in regards to peer feedback of writing. Nelson and Schunn (2009) identified several important features of peer feedback in their nature of feedback model. In the current study, we test an updated theoretical model that includes a broader set of features and considers not only students' likelihood of implementing a comment but also the quality of their revisions. To empirically test the updated theoretical model, we analyze over 7,500 comments from 351 reviewers to 189 authors. Each comment was coded for the presence of praise, a problem description, a suggested solution, localization, focus (i.e., low prose, high prose, substance), implementation, and revision quality. To account for the cross-classified nesting of data, we used a 2-level, cross-classified, hierarchical logistic regression model. Only 2 feedback features increased students' likelihood of implementation (i.e., overall praise and localization), while several feedback features reduced students' likelihood of implementation (i.e., mitigating praise, solutions, and high-prose comments). Overlapping feedback features affected students' ability to revise and in opposing directions from their effect on likelihood of implementation: Revisions were less likely to improve the quality of their paper when implementing comments that included a specific location in the text, but they were more likely to improve the quality of their paper when implementing comments that focused on high-prose and substance issues. Implications of these findings are discussed.

Strengthening dialogic peer feedback aiming for deep learning in SPOCs

“…Features such as explanations about why an answer may (not) be right and suggestions for improvement are in line with previous studies about good quality feedback (e.g. Gielen et al, 2010;Patchan, Schunn, & Correnti, 2016), but add the specific aim of deep learning. To the best of our knowledge, the feature of adding one's own perspective, such as the feedback provider telling the feedback receiver a personal opinion on the assignment submitted, has not been found before with regard to deep learning.…”

Toward Enhancing the Validity in Multi-Peer Assessment: An Approach Based on Students’ Speed-Response-Question Performance

“…In fact, certain studies pointed out that practitioners having a positive attitude towards peer feedback are willing to provide revision exceeding the expected standards [10]. The engagement level will be increasing and the carelessness level decreasing when practitioners hold a positive attitude towards it [14]. Interactive peer assessment, while reducing the number of burdens in favor of teachers, helps learners to understand the learning process of other peers as well as the identification of their own strengths or weaknesses.…”

Development and calibration of the student opportunities for deeper learning instrument

“…So, it may be prudent for the SODLI to also include items that pertain directly to the specificity of teacher feedback about student work, the regularity with which students receive feedback, and students' chances to revise their work in response to that feedback. Moreover, peer (e.g., Patchan, Schunn, & Correnti, ) or parental (e.g., Vandermaas‐Peeler, Massey, & Kendall, ) feedback on student work has also been shown to support deeper learning, and therefore was deemed relevant to the SODLI. In this way, student opportunity to receive feedback , is here considered a component of deeper learning.…”