| The present study depending on Error Analysis will employ corpora as the subject in order to find out the characteristics of the misuse of articles committed by non-English-major students. Besides, based on the results of the corpus study, a semi-structured interview will be involved in the study to make an in-depth analysis of the factors behind the article misuse. Then implications of English article teaching and learning will be produced from the results and discussion of the present study.Chinese learner corpora and a native corpus have been selected for the present study. The sub-categories FS4and JS6of Chinese Learner English Corpus (CLEC) which have been error-tagged are adopted as the researching subjects. And the controlled native corpus for comparison is LOCNESS, belonging to the International Corpus of Learner English (ICLE). Moreover, LOCNESS has been adapted in the present study as sub-LOCNESS to make the comparison more reliable in terms of total tokens in the corpus, age and settings of the writings of FS4and JS6. First, we use the corpus retrieval software AntConc3.2.1w to get the frequencies of articles in the three corpora with the purpose of finding out the article usage on a whole by non-English-major college students in comparison with the native speakers’ article usage. In order to find out if the results of the comparison are significant, a log-likelihood ratio calculator is employed. With the aim of exploring the main distribution and characteristics of the article errors in FS4and JS6, we make use of AntConc3.2.1w to retrieve the article-error-tags related to article misuses. After that, the retrieved results will be categorized, annotated and analyzed manually.As a result of analysis, it is found out that the characteristics of misusage of English articles committed by Chinese non-English-major college students are as follows:1) They tend to underuse articles in general in comparison with the native corpus, sub-LOCNESS;2) With regard to article misuses in the writings of FS4and JS6, omission of articles occurs most frequently, redundancy of articles and article confusion comes second and third respectively.3) The article errors in FS4and JS6could be attributed to the first language transfer, the intralingual factors within the target language and complicated article semantics. It is also necessary to point out that the causes work interactively. That is, a kind of article misuses could be resulted from more than two causes.4) There is no significant correlation between learners’ language proficiency and article acquisition.In order to further investigate the causes behind article errors, the present study also adopts a semi-structured interview based on the results of the previous corpus-based analysis. By collecting the writing samples of two non-English-major seniors and interviewing with them, some possible factors are discussed from the perspectives of the first language transfer, intralingual factors and article semantics.Taking all the results and analyses of the present study, a few suggestions for English instructions are given in the present thesis as well. |
PDF Full Text Request