| The studies of collocation in English language are increasingly coming into people's notice, focusing on the lexical co-occurrences in both written and spoken discourses. In recent years, the advanced computational technology has made it possible to conduct study of collocations on the bases of large corpuses and in broader areas. Such studies in L2 acquisition indicate that L2 learners' lack of idiomaticity is largely due to their inadequate collocation competence. Then, how about non-English majors' collocation competence in their writings? This thesis was intended just to answer the question. The research included in the thesis was conducted quantitatively with the help of corpus softwares and statistic tools, for the purpose of identifying and analyzing the collocation distribution in 90 non-English major writings so as to find ways to improve their writing.The subjects involved in this study were 90 non-English majors from Yangzhou University. They were required to finish a writing of at least 120 words on a given topic and outline within 30 minutes. All the writings were fed into a computer and the collocations in the writings were identified and extracted with the software CLAWS. Human intelligence was employed to do the further identification job. At the same time, all the 90 writings were printed in a random order and graded independently by three English teachers who have received special training in grading. According to the final scores, the writings were divided into two groups: high-score group (HG) andlow-score group (LG)The major findings of the study are as follows:1. There does exist positive significant correlative relationship between the non-English majors' writing ability and their collocation frequency. That is, with other conditions being equal, the more frequently a writer use different significant collocations, the higher quality the writing will be of. It follows that, in writing, frequent use of significant collocations helps to convey meaning more efficiently, which in turn improves the quality of writing. Conversely, the non-English major writers who use less significant collocations in their writings have to turn to the on-line word combination more frequently in order to express their ideas, which is responsible for the relatively poor quality of their writings.2. HG writers use significantly more collocations than their LG counterparts; The writers of both groups tend to use less noun collocations; The non-English major writers differ from one another mainly in their ability to use verb collocations.3. Both groups have much in common in terms of using or producing collocations. For example, they tend to produce word combinations according to their own language habit, mostly, literal translation from Chinese; Writers in LG tend to overuse single common words, when compared with their high-score counterparts; LG writers also tend to use simple collocations or their familiar word combinations, which, in most cases, are merely literal translation; In addition to the shared weaknesses in both groups, LG non-English major writers commit more collocation mistakes because of literal translation; In contrast, negativetransfer, lack of adequate common usage of words and incorrect combination of collocations are mainly responsible for incorrect collocations in HG writings. HG writers use relatively more complex collocations in terms of average collocation length and Z-score.The above mentioned findings are suggestive for English learning and teaching, especially for non-English majors. First, using collocation as input unit for English teaching and learning can greatly reduce learners' error in the acquisition of English words, especially in terms of meanings and common usages. Widdowson (1990: 43) argues "a good deal of communication is achieved not so much by the on-line assembly of analysed items but by the adaptation of formulaic phrases". The "formulaic phrases mentioned by Widdowson here is none other than a matter of collocation. Second, learners should be...
|
PDF Full Text Request