Corpus-Based Study On Collocations In EFL Learners' Writings

This thesis is based on an analysis of three corpora, comparing the language data from Chinese Learner English Corpus (CLEC) and those from native speakers' corpora Translational English Corpus (TEC) and British National Corpus (BNC). It is observed that collocates chosen by the Chinese learners differ greatly from those chosen by native speakers, and that some typical collocates used by the Chinese learners are significantly subject to the corresponding collocation patterns of the equivalent word in their mother tongue Chinese.Vocabulary is the basic element of language. As for EFL learners, the center of language learning is the learning of vocabulary. Collocation is one important aspect of vocabulary knowledge. Recently, more and more scholars pay special attention to the study of collocations. It is usually seen that the learners misuse a collocation while they know the meaning of each word well. By means of corpus concordancing, this paper aims at finding out the differences between native speakers and non-native speakers in word usage and analyzing the possible causes.With the help of corpora and concordancing program, it is possible to find out the differences in the use of some frequent words between native and non-native speakers. Cases of overuse, underuse and inappropriate use in EFL learners' corpus are identified. Pedagogical implications for language teaching are thus drawn and suggestions made for future research. The author also suggests that corpus be made full use of in both language learning and teaching.