A comparison of peer, teacher and self-feedback on the reduction of language errors in student essays

This quasi-experimental study examined the effect of different sources of corrective feedback (teacher, peer, self) on reducing pronoun agreement and lexical errors in student essays. Three Lebanese university student groups sat for a pre-test, received instruction on the errors under study, and practiced correcting them in response to their respective source of feedback. Next, students sat for an immediate and a delayed post-test. Comparison of students' performance on pronoun agreement errors revealed that all groups increased the percentage of this error on the immediate post-test but decreased it on the delayed post-test, with no significant difference between the groups. However, comparison of students' performance on lexical errors at the immediate post-test revealed that the self-feedback group significantly decreased this error percentage compared to the peer feedback group. No significant difference appeared between teacher and peer feedback groups, nor between teacher and self-feedback groups on that test. On the delayed post-test, the self-feedback group significantly reduced the percentage of lexical errors compared to the teacher feedback group, but no significant differences appeared between peer and self-feedback groups, nor between teacher and peer feedback groups. Results show that language development is mediated by source of CF, type of error, and affective factors. (C) 2016 Elsevier Ltd. All rights reserved.