| This dissertation presents results of a field study that investigates the effects of teacher feedback, modes of teacher feedback to five grammar categories and student perceptions to teacher feedback in the context of two Chinese writing classes targeting CET 4 and CET 6. 87 non-English majors from the two classes participated in the study over the winter semester in 2009.In light of related previous studies and the researcher's pilot study, the researcher devises the following research questions in her action research:(1) Does teacher feedback to overall grammar errors or a particular error category in Chinese learners'essays result in improved accuracy in the use of the grammar in new essays?(2) Does teacher feedback to overall grammar errors or a particular error category in Chinese learners'essays result in improved accuracy in the use of the grammar in short-term or long-term revisions?(3) Do explicit feedback and implicit feedback equally function in correcting grammar errors committed by Chinese learners with different language proficiency?(4) What are the student writers'attitudes to teachers'corrective feedback in general and a specific mode of correction employed by teachers? What are the rationales behind their attitudes?To answer the first and second research questions, the researcher uses experimental data to probe whether the teacher's correction of grammatical errors every other week between seven assignments lessens 87 subjects'overall grammar error rate or progress made toward five selected grammar categories: Sentence Structure, Verb, Noun, Preposition and Article. Conclusions and implications are made through a comparison of error rates between the first and the last writing assignment and more specifically, the process of rising and falling of error rates from the first writing assignment to the last one. The two analyses target both the overall error rate figure and the error rate for the five grammar categories.As for the third research question, the study moves on to examine how teacher feedback should be conducted. It is designed to probe the efficacy of explicit feedback and implicit feedback on 4 groups of students with different language proficiency. Two classes of 87 subjects are divided into four groups receiving two modes of error treatment respectively.The treatment modes are listed as follows:Treatment 1= Underlining +Correction (explicit feedback);Treatment 2= Underlining +Codes (implicit feedback).87 subjects are divided into 4 groups:Group 1 (randomly chosen 24 students from Band 4 class) : Treatment 1Group 2 (the rest 23 students from Band 4 class) : Treatment 2Group 3(randomly chosen 20 students from Band 6 class) : Treatment 1Group 4 (the rest 20 students from Band 6 class) : Treatment 2The researcher examines the effects of two different combinations of feedback types measured by error rate and accuracy rate after the students in four groups receive two modes of teacher treatment, make revisions and compose new essays.Apart from the two quantitative studies, a questionnaire is also administered to investigate student perceptions and opinions of the value of teacher feedback at the end of the research. The questionnaire consists of 16 close-ended questions on how students view teacher feedback, the efficacy of teacher feedback, effects of two modes of treatment, effect of revision, suggestions for teachers'future work, etc.Findings of the study indicate that the effectiveness of the practice of overall error correction is not significant since the mean scores displayed show that subjects in the last writing assignment scored not significantly lower in overall error rate than did subjects in the first one. This result is in complete agreement with Truscott's (1996) argument that the effectiveness of error correction is not significant. Therefore, the value of the practice of overall error correction is doubtful in Chinese writing classes. The instructors'energy and time could otherwise be dedicated to other more promising writing activities.However, with regard to five selected error categories, the p value of .011 reveals that error correction for the category of Verb is very encouraging. Besides, according to the curve in the figure of Verb error rate, we can safely predict that the longer we practice Verb error correction, the more progress the student writers will achieve, and that it is worth continuing. However, the effectiveness of error correction for the categories of Sentence Structure, Noun, Article, and Preposition is insignificant. Therefore, the results can assure English teachers'continuation of Verb error correction and urge the teachers to think seriously about not using those ineffective ones.As for the second research question, the drastic improvement of grammar accuracy of overall error correction as well as particular error categories in revisions is noticeable. This result is similar to that of the aforementioned study conducted by Fathman and Whaley that short-term and long-term grammar feedback result in improved writing accuracy on revisions. Further, revision made immediately after the first draft enjoys lesser error rates compared with revision made after a longer period. Student writers'improvement in different linguistic categories also varies: they effectively rectified most Sentence Structure, Verb and Noun errors, while doing a bad job in rectifying Preposition and Article errors. They did a worse job for the two categories in long-term revisions.It can be concluded that receiving grammar feedback is not sufficient for students to improve their grammar accuracy. What seems to be a crucial factor, as is shown by the second research question, is having the students do something with the error correction. In other words, students should not simply rely on receiving feedback. When students incorporate the feedback in revisions, feedback on writing is a way to draw attention to form without influencing their original communicative intent. Since making revisions or correcting errors after receiving feedback facilitates noticing their problematic aspects, it leads learners to develop their interlanguage. This finding seems to refute the assertion that having students correct errors is ineffective. This finding is also supported by the answers given in the Questionnaire, where most students took a positive attitude towards learning to revise.Concerning the third research question, it is concluded that two types of correction methods do result in disparate performance of four groups of students in both the revision and the new essay. Group 1 students perform better than Group 2 in revisions as well as in new essays, so did Group 4 than Group 3. This result conforms to that of Carroll & Swan's study that correction and proficiency may interact in the learning process. The research finds that explicit and implicit forms of feedback do function differently in correcting SLA learners'writing errors with explicit feedback functioning better for beginners and implicit feedback more suited to writers with better language proficiency in both revisions as well as new essays.To verify the conclusion drawn from the previous studies, the research analyses the data from 76 questionnaire samples (19 samples from each group). Data from the Questionnaire indicate that the majority of subjects think there is a need for teachers to provide feedback and teacher feedback is helpful to improve their writing. Students believe they benefit most in Sentence Structure and Verb use from teacher feedback, while they make little progress in Article and Preposition, which echoes the results of the previous study. Obviously, students have a clear picture of where they are and what they need. Band 6 students have sharper observations to the issue compared with their Band 4 counterparts. Therefore, learners with higher language proficiency may not only be able to notice more about the linguistic feature of their own output as they compose, but they may also be better equipped to notice the gap between their writing and the feedback or the revisions given by the teachers. However, students misinterpret their mastery of Noun. Previous studies indicate that Noun feedback are detrimental to their future writing performance, however, as many as 15 students think they make the maximum progress in noun use.As for their views of learning to revise, the majority of students consider it helpful to improve their writing, though some proficient writers from Band 6 class expressed their wishes to spare more efforts to new writing assignments. However, in reply to the question how long they spent revising their draft, more than half of the respondents chose the answer less than 20 minutes. This finding is very frustrating in that in contrast to teachers'vigorous efforts in error correction, students don't pay due attention to their hard work.In regard to students'views to two modes of feedback, it is noticeable that students do welcome teachers'varied feedback catering to their language proficiency to enhance their ability of making revisions as well as composing new essays.Concerning their views to writing improvement, the result indicates students'positive attitude to error correction and accordingly their progress due to feedback, contradictory to the harsh reality that they don't achieve significant progress in most of error categories and overall error rate. Therefore, students'judgment of their progress and their actual progress are two different things.All in all, on the basis of the implications and conclusions drawn from the case study, the findings from both quantitative and qualitative data on learner errors and error treatment were analyzed in a detailed manner, and following this, various comparisons were made to find out the relationship between those patterns of error treatment and the learner factors, thus providing a highly accessible and principled approach to both the theory and practice of error treatment that can guide pedagogical decision-making.
|
PDF Full Text Request