A Corpus Analysis Of The Uses Of SO In Chinese College Students' English Compositions

English has long been one of the most important courses for Chinese students. It has never been easy to Chinese students, not only for its sophisticated words, but also for its frequently-used words covering multi-word classes. The crux of the learning hurdle lies in the overuse or underuse of certain word classes of those words. This thesis aims at the most frequently-used word SO as a point of breakthrough, to discover the overuse, underuse or misuse of SO in Chinese college students'compositions.In the present research, the contrast between native English speakers and Chinese college students on the uses of SO is based on Corpus of Contemporary American English (COCA, over 335 million words with the spoken genre deducted) and three sub-corpora (ST3, ST4, ST6, over 640,000 words in all) from Chinese Learner English Corpus (CLEC). And computer software AntConc3.2.1 and SPSS are applied to explore the differences in using SO from such perspectives as word class, collocation, and position.The results show that Chinese English learners do not have a good command of SO. In comparison with native English speakers, Chinese English learners are inclined to overuse or underuse SO as certain word classes. A great disparity also exists in collocations and positions of SO in Chinese college students'compositions. In addition, the application of error analysis theory reveals that Chinese English learners make SO errors as a result of interference of mother tongue, overgeneralization, incomplete application of rules, and ignorance of conditions for rules to apply. On the basis of the above study, some suggestions for pedagogical practice and further research are provided.