Abstract
Due to the growing popularity of Assessment for Learning in higher education, self- feedback and peer feedback are often highlighted for their role in improving writing performance. In order to provide appropriate support for students' effective implementation of the feedback, teachers must comprehend the differential characteristics of self- and peer feedback. However, empirical research comparing the two remains scarce, particularly when they are used in combination. In this study, 116 Hong Kong undergraduate students participated in an abstract writing task and engaged in self- and peer feedback processes. The amount, types, and implementation of self- and peer feedback and their effects on writing improvement were analyzed and compared. Hierarchical regression analyses indicated that about 25% of the variance in the students' writing improvement was collectively accounted for by the two feedback processes. One form of feedback contributed about 15% of the variance while the other form explained 10%. Feedback types and the amount of implemented feedback were found to be positive predictors of writing improvement, whereas the overall feedback amount negatively affected the improvement. Moreover, the implementation of peer feedback was found to have a greater effect on the improvement than those of self-feedback. Several pedagogical implications of these findings are addressed.