The Effects Of Different Teacher Feedback On EFL Students' Writing Accuracy

Many earlier researches (Ferris & Roberts, 2001; Ashwell, 2000; Fathman & Whalley, 1990; etc.) have demonstrated that teacher feedback has positive effects on English learners'writing accuracy improvement; and surveys of student opinion about teacher feedback have consistently revealed that L2 students hold a positive view on grammar feedback from their teachers, and further, both teachers and students have strong preference to direct grammar feedback (Leki, 1991; Duff & Li, 2004; etc.). However, the results of many experimental studies seem to show that indirect feedback facilitates learners'grammatical accuracy to a greater extent (Lalande, 1982; Ferris, 1995a, 1995b; Ferris & Hedgcock, 1998; Lee, 1997; Robb, Ross, & Shortreed, 1986). Meanwhile, previous studies have mainly focused on comparing the effects on students'overall writing accuracy between giving feedback or not, or between direct and different indirect teacher feedback types, few studies have checked the effects of other feedback options in comparison with the currently frequently proved indirect feedback types, to see whether there is one kind of teacher feedback which not only caters to students'preferences but also generates positive effect. Additionally, many researchers (Frantzen & Rissell, 1987; Ferris, 1995a; Bitchener, Young & Cameron, 2005) agree that different linguistic categories represent separate domains of knowledge that could not be acquired at the same time. So the present study examines another feedback type (i.e. direct feedback plus teacher-student individual conference feedback), comparing its effect on the accuracy of two target error types in lower intermediate EFL students'writings with that of indirect feedback and traditional direct feedback. Two research questions are addressed: (1) Are there differences in the effects of three feedback types on students'writing accuracy of the target error types? If yes, which kind of feedback exerts a greater influence on the improvement? (2) Are there differences between error types in students'writing accuracy improvement under the three feedback conditions? If yes, what are the differences?This research was designed grounded on relevant studies on teacher feedback home and abroad especially on the study of Bitchener et al. (2005). 72 participants, selected from three intact classes of non-English major freshmen at the lower intermediate proficiency level, participated in the study. The selected participants were divided into three groups to complete the writing tasks on three given topics. First, they were asked to write on topic 1 (Writing Task 1) according to the given directions, the class teacher then provided three feedback types (direct feedback only; direct feedback plus teacher-student individual conference feedback in class; indirect feedback by underlining the errors) to the collected first writings across the three groups. Immediately after feedback, the second writing task (Writing Task 2) was assigned to detect the immediate effects of these feedback patterns. Two weeks later, all the participants were required to conduct Writing Task 3 in an attempt to check the delayed effects.The results indicated that all the three feedback groups performed significantly better in Writing Task 2 than in Writing Task 1, and direct feedback plus teacher-student individual conference feedback exerted a marginally greater effect on the accuracy of the two target linguistic forms, but no significant difference between the direct feedback group and the indirect feedback group was observed. From Writing Task 2 to Writing Task 3, although all groups reduced their accuracy mean scores, the scores of the direct feedback plus teacher-student individual conference feedback group was significantly higher than the other two groups. Similarly there was no significant difference between the direct group and the indirect group. Another finding was that there were no significant differences in the effects of teacher feedback between the two error types targeted in this study. Therefore, this study, to some degree, has proved that direct feedback plus teacher-student individual conference feedback has a desirable effect on the accuracy of specific error type for students at the lower intermediate English proficiency level. Of course, the effects of more feedback types on more linguistic features for varying levels of students deserve more future experimental researches.