The Effects of Anonymous and Non-Anonymous Online Peer Review on Feedback Types and Students’ Revisions in Second-Language Writing

Abstract

A promising tool for writing improvement is the use of online peer review, which has become available since the introduction of computer mediated technology in classrooms. One of the possibilities offered by online peer review is the use of anonymity, which has been found to reduce the perception of peer pressure in educational settings. This research examines the effect of anonymous and non-anonymous peer review in second-language writing education. More specifically, it investigates the effect of anonymity and non-anonymity on different types of peer feedback (higher-order and lower-order concerns / directive and non-directive), revisions and writing performance.
The research was conducted in the spring of 2018 with 114 second-year university students, age 18 to 24. These students were enrolled in an English writing course as part of the regular programme. Ten groups of students were randomly assigned to two experimental conditions: five anonymous and five non-anonymous groups (no level groups). In this quasi-experimental field study, 53 students received online peer review anonymously, and 47 students non-anonymously (114 students acted as assessor, 100 as assessee since some essays were excluded from the research as the first drafts were too short). Students uploaded their first draft of a five-paragraph essay in English, before they offered online peer feedback to two fellow students in class.
All peer review comments were collected, segmented and coded. Building on previous research, the revision-oriented comments were categorised as directive or non-directive and higher-order or lower-order concerns. Directive feedback included specific suggestions whereas non-directive feedback involved nonspecific observations without solutions, leaving it to the students to self-correct the text. Higher-order concerns included development of ideas, audience and purpose, style, organisation and argumentation while lower-order concerns involved spelling, grammar, wording and punctuation. These categories were combined in four types of feedback: directive feedback on higher- and lower order concerns and non-directive feedback on higher- and lower order concerns. Subsequently, students’ first drafts were compared to their final drafts to check whether the feedback items were processed, partly processed or not processed. Finally, analyses were run on all data including the final grades for the writing module.
Results showed that students in the anonymous condition provided significantly more feedback on higher-order concerns and offered significantly different types of feedback. As for revision, although assessees in the anonymous condition did not process more feedback than their non-identified peers (adoption rate), there was a difference in the revision of feedback between the two conditions; while directive, lower-order feedback was processed by students in both the anonymous and non-anonymous condition, directive, higher-order feedback was processed more by assessees in the anonymous condition. On top of that, students in the anonymous condition scored significantly higher final grades than students in the non-anonymous group. The relationship between anonymous and non-anonymous condition and final grades was significantly moderated by both revision and trust in the assessor, although these both accounted for only a small part of the variance.
In this study, anonymity affected the types of feedback students gave, it affected revision, and it was also significantly related to higher final grades. These results provide some insights on the impact of anonymity on the online peer review process. It seems to suggest that anonymity could be used to optimise the success of online peer review in the writing process for second-language learners.

Date of Award	31 Aug 2018
Original language	English
Supervisor	Esther Tan (Supervisor)