Effectiveness Of Monitor Training In Helping L2 Learners Self-Correct Errors In Writing

Based on the theories of Error Analysis and Krashen's (1985) Monitor Hypothesis, this study carried out an exploratory quasi-experiment in writing teaching method. It was mainly aimed at testing the effectiveness of a Monitor Training (MT) in helping L2 learners self-correct errors in English writing. It also attempted to explore L2 learners' Monitoring Consciousness (MC) in writing and their Attitude Towards the Pointed-out Errors (ATPE). Besides, the author also tried to find out if there was any difference between the upper- and lower-level students in their MC, ATPE and Error-Detection Ability (EDA).A total of 105 freshmen specializing in Information Technology and Social Security from Nanjing Agricultural University participated in this study from March to June, 2006. Instruments employed in the study included the Pre-Questionnaire (Pre-Q), the Error-Detection Test (EDT), the Post-Questionnaire (Post-Q) and the Writing Test in the final English exam. The data collected from the experiment were analyzed through SPSS 14.0.The study has major findings as follows:1) The results of the Pre-Q showed that most of the college students did not have a strong MC in writing. They could not manage to use the grammatical rules they have learnt to monitor their written output so as to improve their writing. At the same time, they did not treat the pointed-out errors seriously, let alone learned from their own errors.2) The results of the Pre-Q also revealed that the MC of the Upper-level Group (UG) was significantly stronger than that of the Lower-level Group (LG), but their difference in ATPE was not significant. The results of the EDT showed that the UG's EDA was also significantly higher than that of the LG Students in the UG were much more skillful than those in the LG at detecting errors related to article, subject-verb disagreement, verb phrase, sentence logic and passive voice. The higher the English proficiency was, the stronger the MC and the higher the EDA would be.3) The MT was proved to be effective in enhancing learners' MC and reducing grammatical errors in their compositions. The MC of the Experimental Group (EG) was significantly improved through the MT compared with their MC in the Pre-Q. In the Writing Test, the total writing error rate of the EG's compositions was significantly lower than that of the Control Group (CG) in the error types involved in the MT. It indicated that the MT was effective in reducing learners' careless errors.4) The present study of a short-term MT suggests that errors deserve our attention and students can be trained to effectively self-correct writing errors within a relatively short period of time.5) Error correction was proved to be effective in the elimination of these ten types of errors: spelling, punctuation, subject-verb disagreement, part of speech, number, redundancy, tense, comparison, pronoun and article. It was particularly effective in the elimination of errors related to "article". Those ten types of errors have not yet been fossilized in students' IL. It is possible for the learners to eliminate those errors by raising their MC in writing.6) Error treatment can promote SLA. This study is an empirical support for Ellis' (1994: 586) statement that by raising learners' consciousness, it is possible for error treatment to contribute to language acquisition in the long run.This research proves that the model of MT is effective in improving learners' MC and their ability to self-correct errors in writing. Consequently, the grammatical errors in their compositions are greatly reduced.