Font Size: a A A

A Corpus-based Study On The Features Of Chinese EFL Learners’ Spoken English

Posted on:2015-01-17Degree:MasterType:Thesis
Country:ChinaCandidate:Y Y MenFull Text:PDF
GTID:2285330431990537Subject:Foreign Linguistics and Applied Linguistics
Abstract/Summary:PDF Full Text Request
The main function of language is communication and transmission of information. One of the Englishteaching goals is to “train students’ ability to communicate in English”, which highlights the importance ofspoken English. In recent years, spoken English teaching as an important part of foreign language teachinghas received extensive attention, so research on speaking is particularly important. The major objectives ofthis research are to investigate:(1) the features of word cluster used in high achievements in comparisonwith medium and low achievements;(2) the similarities and differences of discourse markers among highachievements, medium and low achievements;(3) the differences of pronunciation among highachievements, medium and low achievements.In this thesis, achievements which are A and A+among COLSEC (College Learners’ Spoken EnglishCorpus), achievements which are B and B+among COLSEC and achievements which are C and C+amongCOLSEC are chosen as the research subjects and these three corpora are: achievements which are A and A+among COLSEC called corpus A(the token is44689), achievements which are B and B+among COLSECcalled corpus B(the token is65618) and achievements which are C and C+among COLSEC called corpusC(the token is28895). This thesis adopts the corpus-based approach, and with the help of cluster anddiscourse marker theories and modern analytical tools, a comprehensive analysis and comparison is madeon the features of the Chinese EFL learners’ spoken English in the three corpora.Through the analyses and comparisons towards the two corpora on the cluster and discourse marker,the research results show that:(1) on the cluster level: this thesis focus on five classifications: phaticexpression, regular responses, discourse signposts, regular expression/semi-fixed phrases and utterancestarters. COLSEC is a corpus of exam, and the rule is clear, so there is no cluster to express phatic purpose;students who get higher scores tend to respond to the others’ questions and express their opinions, so thestandardized frequency of regular responses in corpus A is higher than in corpus B and corpus C; similar asregular responses, students who get higher scores tend to use discourse signposts, because they tend toreveal the logic and semantic relationship of information or the discourse development direction; while forthe classification of regular expressions/semi-fixed phrases, the students who get lower scores use more regular expressions/semi-fixed phrases, the reason may be that they like to use simple and easy expressionand the nominal phrase; because the students who get higher scores can always begin a new talking, thestandardized frequency of discourse signposts in corpus A is higher than in corpus B and corpus C;(2)discourse markers: the similarities of these three corpora: very occurs a lot of times in these three corporaand concluding markers and optional markers are used less, which is due to the reason that COLSEC is acorpus of test; the difference of these three corpora: except such as, because and oh, the standardizedfrequency of all the discourse markers are higher in corpus A than in corpus B and corpus C;(3)pronunciation: the students who get lower scores make more error on pronunciation, maybe this isinfluenced by Chinese phonetic alphabet,[wth-z],[wth-s] and [ww-v] occur a lot of times in these threecorpora; pause: the students who get lower scores pause more than other students, because they need moretime to organize their words or sentences, or to recall words to express their opinions.From the analysis and comparison on the discourse features of the corpus A, corpus B and corpus Cfrom the above three levels, the differences of the discourse features of EFL learners’ spoken English aswell as the similarities between the three different score levels are concluded. In general, high score leveland low score level have some similarities on the cluster, discourse marker and pronunciation levels.However, because they are different score levels, there are indeed several differences between these threecorpora. Through the research of this thesis, not only readers’ understanding of the features of Chinese EFLlearners’ spoken English will be promoted to some extent, but some implications can be made to theresearch of spoken English, moreover, it is of positive influence on EFL teaching, learning and research.
Keywords/Search Tags:corpus, cluster, discourse marker, pronunciation
PDF Full Text Request
Related items