The Effect Of Students' Language Proficiency On Negotiation Moves And Modified Output

In the earlier study of SLA, the main focus was on whether pair or group interaction could provide each member with Comprehensible Input which has been claimed to be a crucial element in SLA. Results of research have shown that interaction has a positive effect on language acquisition, one area of acquisition in particular: listening comprehension. However, only comprehensible input is not sufficient to language acquisition. Swain (1985) investigated French immersion program in Canada. At school, students whose mother tongue is English were instructed in French. As a result, they received a rich source of comprehensible input, although they were advancing in listening and reading comprehension, their language in speaking and writing were still off-target. On the basis of investigation, Swain (1985) concluded that Comprehensible Input is important to language acquisition, but it is not able to facilitate learners to develop their second language. If learners want to speak fluently and accurately, they need not only comprehensible input, but also Comprehensible Output. Producing output can not only make learners speak fluently, but also improve learners' language accuracy. Based on Swain's Comprehensible Output, the focus of the research of SLA shifted to study the relationship between negotiated interaction and modified output. Researchers have conducted experiments which covered such factors affecting modified output as task types, background knowledge, gender and familiarity. But few studies have been concerned with the relationship between learners' language proficiency and modified output.In light of precious discussion, this present study aimed to investigate whether students' language proficiency had an effect on modified output. A total of 58 students from Applied Chemistry Department participated in this study, but only 30 students' transcripts were available. Before testing, we conducted tests of oral, writing and listening comprehension and evaluated their performance in these three tests and other two tests: Entrance Exam of English and College English Placement Test. We divided them into three groups according to two different language proficiency levels: High-High dyads, High-Low dyads, and Low-Low dyads. Each group had five dyads, ten students in all. Experiment was carried out in the setting of a 'real classroom' as possible as we could. Communicative tasks chosen in the study could provide an opportunity not only to produce the target language, but also, through negotiationmoves (e.g. clarification requests and confirmation checks) to manipulate and modify it. One was the required information exchange task; the other was the optional information exchange task. The data was drawn from transcripts of audio-recordings of 30 students. We analyzed 30 subjects' language production, incidence of negotiation and modified output. In order to ensure reliability, the performance in transcripts were identified by two other raters. The results have shown that students when teamed up with another with high language proficiency could provide the most opportunities to modify output and most amount of production of modified output; moreover, interaction in mixed group could facilitate low proficiency students to improve their quantity and quality of output than those in Low-Low dyads.