A Study On Teachers’ Corrective Feedback And Learners’ Uptake In English Classroom Interaction

In the area of second language classroom research, classroom interaction betweenteacher and students has drawn much attention of foreigners as well as researchers andteachers at home. Teachers’ corrective feedback, which is regarded as the medium ofclassroom interaction, is at the core of studies of the classroom interaction. It is suggested thatthrough the interaction between the teacher and students, the latter can be exposed to thetarget language, pick up useful feedback from the teacher and make more language output tofacilitate their interlanguage development. With a view to exploring how teachers’ correctivefeedback lead to students’ uptake in classroom interaction, an empirical study including twoEnglish teachers,126students in two regular classes (RC) and121students in two key classes(KC) from Pian Guan High School is conducted. Data were collected from video-recording ofclassroom observation together with a questionnaire for students and interview with the twoteachers in four classes. The major findings of the present study are as follows:Students in both RC and KC make errors in classroom interaction. The distributions oferror types they make are different, however. The error type students in RC make the most isgrammatical errors. And the followed are phonological errors and lexical errors. Thoughgrammatical errors are also most frequently made by students in KC, the lexical errors anddiscourse errors occupy a large part of all the errors as well.Teachers do not over-react to students’ errors in both RC and KC. The choice ofcorrective feedback from teachers is affected by types of errors and students’ proficiency levels. In RC, elicitation is more likely to be employed to treat grammatical errors and lexicalerrors, and explicit correction is often used in response to phonological errors and discourseerrors. But in KC, teachers tend to use recast to treat students’ grammatical errors whileelicitation for lexical errors and clarification request for discourse errors. Teachers in KC treatstudents’ discourse errors quite more frequently than in RC. This phenomenon is on accountof the relatively higher frequency of discourse errors from students’ speech in KC.With regard to students’ different L2proficiency levels, the choice of feedback has arole to play in the types of uptake that follows. In KC, prompts are likely to lead to students’self-repair. By contrast, in RC, more needs-repairs but less self-generated repairs were elicitedby prompts. For students in RC, explicit correction is more suitable.Much less frequently is metalinguistic feedback provided than other types of correctivefeedback by teachers both in RC and KC in the present study, though the fact that it is a muchpowerful way in resulting in students’ repair.Students with different proficiency levels have different preferences to teachers’corrective feedback. Most of the students in RC and KC want their teachers to treat theirerrors. Students of RC prefer explicit correction the most while students of KC like elicitationand metalinguistic the best. Students of KC prefer the corrective strategies which providehints and allow self-correction.Teachers are more likely to provide explicit corrective feedback for phonological errors,which meets students’ preference in both RC and KC. In regard to grammatical errors,teachers tend to provide recast and implicit feedback in KC, while students seem to preferteachers to give them a little more explicit correction.Teachers’ choice of corrective techniques is also affected by the teaching contents andteaching process. When the teaching contents are large enough and the teaching time islimited, the corrective feedback is more likely to be explicit, so teachers tend to employ morerecasts and explicit corrections.