Use Of DMs By Chinese College English Majors And Oral Performance

Discourse markers (hereafter DMs) play a very important role in producing coherent and well-organized discourse. Scholars at home and abroad have done researches to examine the relationship between DMs and oral performance. However, there is a lack of essential statistical work and sufficient cases concerning the use of DMs and its correlation to oral performance in L2 learners' spoken discourse. The present corpus-based research is an attempt to investigate the use of DMs and its correlation to oral performance in Chinese College English majors' spoken discourse. This study does not solely rely on statistical evidence, but combines the quantitative information with a qualitative, interpretive analysis to answer the following research questions:(1) What are the characteristics of Chinese College English majors' use of DMs in contrast to English native speakers?(2) Do students of different oral proficiency level behave differently in the use of DMs?(3) Is there any correlation between oral performance and the use of DMs with regard to the frequency and typology in Chinese College English majors' Spoken English?The study is carried out on the basis of SECCL (Spoken English Corpus of Chinese Learners). The native spoken corpora BNC World is adopted as a reference corpus. We combine Fraser's classification of DMs with Wan & Zhu's, deciding 4 types of DMs (Logic Markers, Modal Markers, Vocal Hiccups and Fillers) totaling 98 specific lexical items for retrieval. Frequency of DMs in SECCL and BNC is calculated with the help of the lexical analysis tool AntConc3.2. DMs used in high-scoring group (HG) and low-scoring group (LG) are also counted for Independent Sample T-tests with the aid of SPSS 13.0. And finally Pearson Correlation Tests are conducted between DMs and oral performance. The thesis comes up with the major findings as follows: (1) Use of DMs by Chinese students is not equal to that of the native speakers: Chinese students tend to overuse several DMs and on the other hand rarely use those that appear very frequently in native corpora.(2) Students of different language proficiency behave differently in using DMs: HG students employ more Logic Markers to achieve discourse coherence than LG; both students in HG and LG have a tendency in repeating one Modal Markers I think when they come to express their personal views, comments and attitudes; and HG students are more competent in using strategic fillers like you know and I mean instead of Vocal Hiccups like um and uh to maintain free flow of speech production.(3) The picture is rather complex as to the correlation between the use of DMs and oral performance: the use of Logic Markers is positively correlated with oral performance and the correlation between them is perfectly significant; the use of Modal Markers and Vocal Hiccups are negatively correlated with oral performance, but the correlation in both situations is not significant; the use of Fillers is positively correlated with oral performance, and the correlation between them is significant.The present research would be of great help for the teachers to identify some specific factors that influence students' oral performance and raise their focus on DMs in oral English teaching. It also sheds some new lights on the evaluation of oral tests. If automatic identification of DMs by computers is made possible, the correlation results will provide theoretical reference to the improvement of computer-aided evaluation systems.