According to Xie’s study (2020), in the 5-likert scale, the higher the score was, the higher the students’ level of peer feedback perception was. In other words, if they got higher scores in the scale, they had a more positive attitude towards peer feedback. In general, the average score between 4 and 5 points is high level, 3 to 4 points is medium level, and below 3 points is low level.
In this study, the overall peer feedback perception level (M=3.97) was medium, indicating the postgraduate students generally had a positive attitude and believed that peer feedback could help them in writing but still had some doubts. The result was similar to findings in several studies such as Hu (2005) and Zhao (2010), and was contrary to Wang’s (2014) and Kaufman’s (2011) findings that students resisted adopting peer feedback to modify their drafts.
Chapter Six Conclusion
6.1 Major Findings
The intention of this study is to investigate into the current situation of postgraduate students’ peer feedback perception, and reveals the impact of peer feedback perception on student behavioral engagement in providing and implementing feedback. Through collecting feedback and writing data, questionnaire for measuring peer feedback perception and an interview protocol for a detailed analysis, this paper analyzed and discussed the level of students’ peer feedback perception and their behavioral engagement. Furthermore, this paper investigated the relationship between peer feedback perception and student behavioral engagement in providing and implementing feedback. In general, the research results are synthesized as three primary conclusions:
(1) On students’ peer feedback perception, the average score of the participants was 3.97, the medium level between the score 3 and 4. The overall perception level indicated that the students’ attitude towards peer review were not negative. Of all the six sub-scales, the level of students’ perception of review assurance (M=4.31), review outcome (M=4.06) and feedback mode (M=4.06) were high. For peer comment (M=3.98), review reliability (M=3.69) and grouping mode (M=3.84), the level was medium. The perception level of review assurance mechanism that scored the highest indicated that students thought it necessary to receive systemic training before peer review and recognized the mechanism of the system; while the perception level of peer review reliability that scored the lowest presented students had some doubts about the reliability of peer review, such as worrying about the fairness and effectiveness of peer review. When analyzing the specific items contained in each dimension, the study found that the reason for students’ relatively low perception of peer review reliability could be attributed to students' belief that peers had contingency and uncertainty when giving comments, and the comments given by the same reviewer for the same text at different times were also different.
(2) On students’ behavioral engagement, it involves providing feedback (M=97.89, SD=37.55) and implementing feedback (M=31.63, SD=4.59), and the level of students’ behavioral engagement in both is usually good and consistent. In terms of providing, students provided 46.95% revision-oriented criticism, 7.47% well-founded praise and 45.58% only praise. It demonstrated that nearly half of the students engaged in the reviewing process by pointing out deficiencies from a critical perspective. They also focused on identifying the strengths of their peers. However, very few of the praises explained the reasons, which indicated that students spared least behavioral engagement by offering well-founded praise. From the perspective of implementation, students implemented 73.6% of the revision-oriented critical comments, including 45.8% full implementation and 27.8% partial implementation. With the support of cooperative learning theory, the data assumed that students tended to implement comments that co
