| Writing is,among the four basic skills of language learning(i.e.,listening,speaking,reading and writing),the most difficult one but can best reflect learners’ language literacy.Student writings in English across China’s senior high schools are however satisfactory,in particular with sharp contrasts between those of outstanding students and underachieved ones.The current thesis undertakes a corpus-based study of lexical and syntactic errors in these two groups of students at a high school by following Granger’s(1998)CIA(Contrastive Interlanguage Corpus Analysis).The error tagging system of CLEC(Chinese English Learner Corpus,2003)and BFSU Qualitative Coder 1.1 were deployed to mark up a self-compiled learner corpus of student writings and Ant Conc 3.5.8 to make retrievals.Results show that these two groups of students resemble each other mainly in the following two aspects:(1)The top three types of errors remained the same between them at the lexical and syntactic levels.At the lexical levels,the top three categories are substitution errors,spelling errors and part of speech errors,while at syntactic levels,the top three are structural deficiency,sentence fragment and run-on sentence.(2)At the lexical level,a significant decrease of spelling errors can be observed along with the ongoing process of student reviewing while the substitution and part of speech errors did not improve significantly.On the other hand,these two groups of students differed from each other mainly in errors of spelling and substitution: Much bigger intra-group variations can be found among underachieved students in comparison with their outstanding counterparts.Furthermore,the underachieved students were liable to more diversified but largely simpler syntactic errors.This study suggests that negative language transfer,style of teaching and environment of learning turn to be the common causes,but for underachieved students,poor management of learning emotions and improper learning strategies may develop into a major factor resulting in their poor performance.In addition to suggestions on enhancing teaching methods and integrating data-driven learning,learner autonomy and learning emotions of underachieved students are also discussed in the context of using learner corpora of student writings to facilitate data-driven pedagogy. |
