A Longitudinal Study On Lexical Richness In College English Writing

Over the past few decades, an important component of second language writing research has been the analysis of vocabulary in written texts and there have been quite a few fruitful research findings on this part. Lexical richness (LR) is used to measure the breadth of lexical use by a speaker or writer. Nation (1995) argues that a well-written composition, among other things, makes effective use of vocabulary. Engber (1995) also confirms that there is certain correlation between lexical diversity and writing quality either including or excluding lexical errors.A longitudinal study is conducted on the development of lexical richness in their writing by taking 30 Chinese non-English majors as the subjects. According to Read (2000), lexical richness is classified into four dimensions, i.e. lexical variation, lexical sophistication, lexical density and a low number of errors.359 compositions in total have been collected over one year. This research analyzes the development of LR in the writings of these participants and the relationships of LR to writing quality by means of the combination of qualitative and quantitative methods and various softwares such as RANGE, AntConc, Gotagger and SPSS.The findings indicate that as the participants learn, there has been a steady increase in lexical variation, lexical sophistication and lexical density. At the same time, they are making fewer and fewer lexical errors as they make progress in English proficiency, but article errors and spelling errors have always been the most serious. With regard to the relationship between LR and writing quality, lexical variation, lexical sophistication and lexical density are found to be positively correlated with the writing quality at a significant level. There is a significantly negative correlation between lexica error and the writing quality.Finally, this thesis analyzes these phenomena and provides some suggestions based on the above study.