The effect of teachers' feedback on the students' ability to self-edit in L2 writing classes

النوع : Mémoire

Several studies on error correction in second language (L2) writing have provide d evidence that students who receive feedback from teachers improve in accuracy over time. Yet, one issue has not been thoroughly examined which is how "explici t" the error feedback should be in order to help students to self- edit their texts. The purp ose of this study was to examine whether or not error feedback helps ESL student s to improve the accuracy and quality of their writing. In this experimental cla ssroom study, 51 high- school ESL students' differing abilities to self- edit th eir texts across three feedback conditions were investigated: (1) errors marked with codes from five different error categories; (2) errors in the same five categories un derlined but not marked or labeled; (3) no feedback at all. The study also exami ned the relationships between students' self- assessments of their own writing problems, their prefere nces about error correction and their prior knowledge about specific grammar pro blems and their written forms (essays). Statistical procedures were used to anal yze all the data including descriptive statistics (frequencies, percentages, and standard deviations), ANOV A, MANOVA, Post- Hoc Scheffe, and Pearson- product correlations. First, a MANOVA test revealed statistically significant differences between feedback groups. Po st- hoc Scheffe test reported that the significant difference was between the co ded and the no feedback groups in their ability to self- edit errors in all five categories. However, no significant differences were reported between the "code s" and the "no codes" groups. Second, an ANOVA test showed no significant differ ence in the students' ability to self- edit across error categories. Finally, Pe arson- product computations revealed no statistically significant relationship b etween the students' formal knowledge of grammar and their ability to self- edit their errors. The results were then discussed in the light of previous research findings, and recommendations for further research were made.