Research Of Web-Corpus-Based Self-Correction On College Students' English Composition

Error Analysis is a teaching method which has been widely resorted to in foreign language teaching. As far as the traditional foreign language teaching is concerned, much importance is attached to self-correction whether from the aspect of human cognition or face saving, which may be classified into two forms both at home and abroad:peer-correction and teacher-triggered self-correction. Peer-correction is generally realized within the group, wherein language errors are judged and corrected through group discussion. Such a form of judgment and correction depends on students'language competence and has limitations, though beneficial to students' mutual correction. Teacher-triggered self-correction is convenient and efficient, while canceling the process of "error-identification". Both peer-correction and Teacher-triggered self-correction, in the eye of the author of this thesis, are not self-correction in a strict sense. Self-correction should be based on the students' own recognition and correction of language errors. The so-called self-correction in traditional English teaching is only confined to peer-correction and teacher-triggered self-correction, which are not conducive to help students judge and correct errors. In order to make a breakthrough into students' self-correction strategy, it is necessary that the Internet nowadays be availed of its huge corpus to achieve the goal of students' self-correction, thus improving their independent learning ability.Based on the related theories of Corpus Linguistics and Error Analysis, this study is aimed to evaluate the effect of self-correction through the educational experiment, with non-English major students using web corpus to identify and self-correct their errors as the subjects. The subjects are 60 freshmen (majoring in Process Design) from Arts Department, College of Information Science, Guilin University of Electronic Technology. Class One (31 students) and Class Two (29 students) serve as experimental class and control class respectively. The conditions for the two classes such as curriculums, teachers, textbooks and class hours are just the same. Their common teaching goal is to improve composition writing. The Experimental Class takes the students' self-correction via web corpus as the teaching formula, while the Control Class adopts the traditional composition teaching formula, i.e., the teacher is in charge of error-correction. By means of pre-testing and post-testing on the two classes, and a comparison between their composition results, we tentatively draw conclusions on the effect of college students' self-correction on their English compositions by using web corpus.The conclusions of the present study are as follows:1) Corpus-based self-correction in English composition teaching is a new trend in college English teaching; 2) College teachers can help students to improve their English writing by corpus-based self-correction on language errors; 3) The corpus-based self-correction strategy is more effective to those high-scored students than those low-scored student; 4) Students with low proficiency need more help and guidance in applying this new method; 5) Self-correction strategy in English writing can greatly promote students' independence in learning.In summary, this study is only an evaluation of the effect of self-correction through web corpus in English writing, focusing on the role of web corpus in college students'self-correction from one aspect; more comprehensive research is necessary if a complete understanding of the effect of college students' self-correction of all-typed errors is to be reached.