CHAPTER FOUR RESLUTS AND DISCUSSION
In this chapter, the results of the data analysis are reported and discussed. I will first report the results of the pre-test and post-test composition scores. Then, I will report the analyses of the participants’ responses to the pre-test and post-test questionnaires. The report of analyses will be followed by discussion.
Results
In the following section, I will first present the results of the comparison of the participants’ writings scores in the pre-test and the post-test. The total scores and the scores for five components (content, organization, grammar, vocabulary and
mechanics) are compared between the two groups. Then I will present the results of comparing participants’ responses to the pre-test questionnaire and the post-test questionnaire.
Results of Participants’ Writing Scores
Presented in this section is the effect of the writing instruction and practice of the study on writing quality for all participants, including the multi-draft group (the experimental group) and the single-draft group (the control group).
To examine the writing ability of the two groups at the beginning of the study, the researcher used t-test to compare the group mean scores obtained in the pre-test.
The result is shown in Table 3.
Table 3
Comparison of the Writing Scores in the Pre-test Between the Two Groups
Mean (SD)
Multi-draft (n = 42) Single-draft (n = 35)
t-value
Overall 9.52 (2.24) 9.53 (2.58) -.02 n.s.
Content 2.24 (.74) 1.99 (.75) 1.53 n.s.
Organization 2.15 (.76) 1.90 (.75) 1.43 n.s.
Grammar 1.58 (.55) 1.56 (.64) .16 n.s.
Vocabulary 1.80 (.60) 1.89 (.70) -.58 n.s.
Mechanics 1.74 (.31) 1.70 (.39) .53 n.s.
Note. 1.The scores were measured by the average of the sums of the two raters’ scores.
2. The full score is 20 and the discrete scores for the five components are as
follows—content: 5 points; organization: 5 points; grammar: 4 points; vocabulary: 4 points; mechanics: 2 points.
n.s. = not significant at 95% level of probability
* p < .05. **p < .01.
Based on Table 3, the average overall scores of the two groups were almost identical (9.52 and 9.53). Also, the average scores gained on the five components indicated no significant differences. It is evident that these two groups demonstrated very similar writing ability in the pre-test.
A second t-test was conducted to compare the composition grades the
participants had received from their composition class in the previous semester. The result is shown in Table 4.
Table 4
Comparison of the Participants’ Composition Scores in the Previous Semester Between the Two Groups
N Mean SD t-value
Multi-draft group
Single-draft group
42 35
2.62 2.78
.91 .90
-.77 n.s.
Note. The total score is 5.
*p= < .05
The mean scores for the multi-draft group and the single-draft group were 2.62 and 2.78 respectively (t = .77, p > .05). This result again indicates that there was no significant difference between the writing ability of the two groups. This provides extra evidence that the multi-draft group and the single-draft group were equal in their writing proficiency at the beginning of the study.
To examine the effect of the writing instruction and practice on the participants’
writing abilities, the scores gained by each group in the post-test were computed and compared with those gained in the pre-test. A paired sample t-test was used and the results are displayed in Table 5 and Table 6.
Table 5
Comparison of Writing by Multi-draft Group Between Pre-test and Post-test
Mean
Pre-test Post-test
Mean Gain (SD) t-value
Overall 9.52 12.15 2.64 (2.62) 6.52**
Content 2.24 2.83 .59 (.84) 4.56**
Organization 2.15 2.89 .74 (.94) 5.52**
Grammar 1.58 2.13 .55 (.68) 5.24**
Vocabulary 1.80 2.54 .74 (.59) 8.08**
Mechanics 1.74 1.76 .0179 (40) .29n.s.
Note. 1.The scores were measured by the average of the sums of the two raters’ scores.
2. The full score is 20 and the discrete scores for the five components are as
follows—content: 5 points; organization: 5 points; grammar: 4 points; vocabulary: 4 points; mechanics: 2 points.
n.s. = not significant at 95% level of probability
* p < .05. **p < .01.
As indicated in Table 5, students in the multi-draft group made a significant progress in the writing post-test not only in the overall performance but also in almost all discrete components of writing: content, organization, grammar and vocabulary. A comparison of their improvement in these components shows that they made a more significant improvement in organization and vocabulary.
Table 6
Comparison of Writing by Single-draft Group Between Pre-test and Post-test
Mean
Pre-test Post-test
Mean Gain (SD) t-value
Overall 9.53 12.82 3.38 (1.92) 10.41**
Content 1.99 2.87 .91 (.68) 8.01**
Organization 1.90 3.01 1.13 (.73) 9.13**
Grammar 1.56 2.38 .84 (.66) 7.48**
Vocabulary 1.89 2.74 .82 (.62) 7.84**
Mechanics 1.70 1.78 .0829 (.41) 1.34n.s.
Note. 1.The scores were measured by the average of the sums of the two raters’ scores.
2. The full score is 20 and the discrete scores for the five components are as
follows—content: 5 points; organization: 5 points; grammar: 4 points; vocabulary: 4 points; mechanics: 2 points.
n.s. = not significant at 95% level of probability
* p < .05. **p < .01.
As indicated in Table 6, students in the single-draft group made a significant progress in their overall performance as well as all the components except mechanics.
Their improvement in organization is the most remarkable among the five components.
To explore the effect of the two different writing procedures on students’ writing quality, the scores gained by the two groups in the post-test were compared. The results are illustrated in Table 7.
Table 7
Comparison of the Writing Scores in the Post-test Between the Two Groups
Mean (SD)
Multi-draft (n = 42) Single-draft (n = 35)
t-value
Overall 12.15 (2.66) 12.82 (2.21) -1.19 n.s.
Content 2.83 (.878) 2.87 (.78) -.18 n.s.
Organization 2.89 (.85) 3.01 (.73) -.70 n.s.
Grammar 2.13 (.67) 2.38 (.69) -1.64 n.s.
Vocabulary 2.54 (.60) 2.73 (.45) -1.55 n.s.
Mechanics 1.76 (.28) 1.78 (.34) -.33 n.s.
Note.
n.s. = not significant at 95% level of probability
* p < .05. **p < .01.
Table 7 shows the mean score obtained by the multi-draft group on the overall performance is slightly lower than that of the single-draft group, but there is no significant difference. Likewise, the mean scores obtained by the multi-draft group on the five components are all lower than those of the single-draft group, but again, no significant differences are observed.
Results of Participants’ Responses to the Attitude Questionnaires
In the following sections, I will first analyze the comparison of the participants’
responses to the questionnaires between the two groups. Then, I will analyze the
participants’ responses between the pre-test questionnaire and the post-test questionnaire for each group.
Comparison of the Participants’ Responses to the Pre-test Attitude Questionnaire Between the Two Groups
In order to see if the participants held similar attitudes toward revision and writing before the treatment, the participants’ responses to the pre-test questionnaire between the two groups were compared. The pre-test questionnaire was composed of three Likert-scale items and one open-ended question and will be discussed in such order.
The Likert-scale Items
A Chi-square test was used to examine if participants from the two groups differed in their attitudes toward revision and writing at the beginning of the
experiment. The results are shown in Table 8. Table 8 also shows a comparison of the percentages of the participants’ positive responses (including “agree” and “strongly agree”) and negative responses (including “disagree” and “strongly disagree” ) to each item.
Table 8
Comparison of Participants' Responses to the Pre-test Questionnaire Between the Two Groups
Multi-draft ---
Single-draft ---
Item χ2
P (%) N (%)
P (%)
N (%)
1.Writing English compositions makes me
anxious. .66 n.s. 78.6 19.0 82.8 14.3
2. Writing English compositions is a
delightful thing. 4.19 n.s. 26.2 54.8 40.0 37.1
3. I prefer revising an old composition to
writing on a new topic. 1.47 n.s. 54.7 33.3 60.0 25.7 Note. P = positive (“agree” and “strongly agree”); N = negative (“disagree” and “strongly disagree)
n.s. = not significant at 95% level of probability
* p < .05 **p < .01
As indicated in Table 8, there was no significant difference in the participants’
responses to any of the three Likert-scale items in the pre-test questionnaire between the two groups.
The comparison of the positive and negative responses reveals two major findings. First, the majority of the students in both groups felt anxious when writing English compositions (78.6% and 82.8%). Also, more than half of the students in both groups preferred revising an old composition to writing on a new topic (54.7% and 60.0%).
The Open-ended Question
The open-ended question (I □ like / □ dislike to revise an old composition
because_____) was aimed to elicit more detailed information regarding the reasons why students liked or disliked to revise. Students’ positive and negative responses to the question were calculated and compared (See Table 9).
Table 9
Participants’ Attitudes Toward Revision in the Pre-test Questionnaire
Multi-draft f (%)
Single-draft f (%) Positive toward revision
Negative toward revision
21 (50%) 21 (50%)
20 (57%) 15 (43%) Note. f = Frequency.
Table 9 indicates that students in the two groups held similar attitudes toward revision at the beginning of the study: in each of the two groups, about half of the students liked revision (50% and 57%) and half of the students did not (50% and 43%). The result is consistent with the result obtained in the Likert-scale items (See Table 8).
The reasons why the students liked to revise were generally the same between the two groups. Their reasons included that revision could improve one’s writing skills as well as language ability and that revising an old composition was easier and more relaxing than writing on a new topic. Below are some examples:
1. "I really want others to understand my ideas so I am happy to make revision so as to make my text more readable." (M-40)
2. "Revision makes me more aware of my writing problems." (S-32) 3. "I can remember my problems better if I revise." (S-30)
On the other hand, the reasons why the students disliked revision included that revision was frustrating, that it was not helpful to revise and that it was boring. Below are some examples:
1. "It’s not necessary to revise. Reading teacher’s feedback alone can help us write better". (M-29)
2. "It’s a waste of time to work on an old stuff. What we need to do is remember the problems in our compositions and try a new topic, which is more
challenging." (S-27)
Comparison of the Participants’ Responses to the Post-test Questionnaire Between the Two Groups
The participants’ responses to the post-test questionnaire items were compared between the two groups to see if students in the two groups differed in their attitudes toward revision and writing after the treatment.
The Likert-scale Items
According to the classification in Appendix D, the 11 items are classified into four categories. Items 1, 2 and 7 explore the participants’ writing apprehension; Item 3 is concerned with the participants’ attitudes toward revision; Items 4-6 describe the participants’ attitudes toward feedback and sharing; and Items 8-11 are concerning the participants’ attitudes toward the influence of the writing course. A Chi-square test was used to examine if participants from the two groups differed in their attitudes toward the Liker-scale items after the treatment. The results are shown in Table 10.
Table 10 also displays the comparison of the percentages of the participants’ positive responses and negative responses.
Table 10
Comparison of Participants' Responses to the Post-test Questionnaire Between the Two Groups
Multi-draft Single-draft
Item χ2
P (%)
N (%)
P (%)
N (%)
1.Writing English compositions makes me anxious. 12.97* 47.7 45.2 54.3 20.0
2. Writing English compositions is a delightful thing. 3.54 47.6 21.5 37.1 62.9
3. I prefer revising an old composition to writing on a
new topic. 11.67* 69.0 14.3 34.3 28.6
4. The activity of “composition sharing and analysis” is
very helpful to my writing .61 73.8 4.8 71.4 2.9
5. The activity of “peer feedback” is very helpful to my
writing 1.11 64.2 14.3 68.6 8.6
6. The teacher’s feedback is very helpful to my writing. 3.89 95.2 2.4 82.8 5.7
7. Due to the writing practice in this semester, now I feel more confident when given a new topic to write on.
1.78. 71.5 16.7 62.9 20.0
8. Due to the practice in this semester, now I pay more
attention to content when writing. 3.15 90.5 4.8 85.7 5.7
9. Due to the practice in this semester, now I pay more
attention to organization when writing. 2.24 88.1 7.1 80.0 8.6
10. Due to the practice in this semester, now I pay more
attention to grammar when writing. 5.65 81.0 2.4 71.4 17.1
11. Due to the practice in this semester, now I pay more
attention to vocabulary when writing. 5.23 78.5 2.4 77.2 14.3
Note. The items underlined are those that have appeared in the pre-test questionnaire.
P = positive; N = negative
* p < .05. **p < .01.
As indicated in Table 10, significant differences were found between the two groups’ responses to Item 1 that investigate students’ writing apprehension and Item 3 that investigated students’ attitudes toward revision.
In terms of students’ writing apprehension, for Item 1, the percentage of students in the multi-draft group who disagreed was two times higher than the percentage in the single-draft group (45.2% and 20.0%). In other words, twice as many students in the multi-draft group thought writing did not make them anxious. Moreover, for Item 2 of the same category, though the difference was not significant, it was important and worth noticing. The percentage of the students in the multi-draft group who regarded writing English compositions as NOT delightful was only one-third of the percentage of the students in the single-draft group (21.5% and 62.9%). It is, therefore, evident that after a semester’s treatment, students in the multi-draft group had a lower apprehension than students in the single-draft group.
As for students’ attitudes toward revision, for Item 3, significantly more students in the multi-draft group preferred revising an old composition to writing on a new topic than students in the single-draft group (69.0% and 34.3%). It is evident that after a semester’s treatment, students in the multi-draft group obviously gained a more positive attitude toward revision.
On the other hand, no significant difference was found in the students’ attitudes toward feedback and sharing between the two groups. For Items 4-6, students in both groups affirmed the helpfulness of the peer feedback activity, the teacher’s feedback as well as the composition analysis instruction.
Likewise, no significant difference was found in the students’ attitudes toward the influence of the writing course between the two groups. For Items 8-11, a great majority of students in both groups affirmed the function of the writing practice in
helping them pay more attention to content, organization, grammar and vocabulary.
The Multiple Response Item
The statement in the multiple response item (Item 12) was “Revising is
□ delightful □ painful □ troublesome □ helpful to my writing
others:______ （You may choose more than one answer）.” Cross-tabulation was used to compare the two groups’ responses to the statement. The results are shown in Table 11.
Table 11
Comparison of Participants’ Responses to the Multiple Response Item Multi-draft
(N = 42)
Single-draft (N = 35) f Pct of Cases f Pct of Cases
Helpful to writing 38 (90.5%) 25 (71.4%)
Delightful 14 (33.3%) 4 (11.4%)
Bothersome 14 (33.3%) 21 (60.0%)
Painful 2 (4.8%) 4 (11.4%)
Others 4 (0.1%) 2 (0.06%)
Note. Pct of Cases = percent of cases = number of students who checked the particular adjective / total number of students in a group; f = frequent.
As shown in Table 11, as high as 90.5% of students in the multi-draft group thought revision was helpful to writing and one-third of the students in this group thought it was delightful (33.3%). By comparison, although most students in the single-draft group thought it was helpful to writing (71.4%), only 11.4% of the
students saw it as delightful to revise an old composition, which was only one-third of the percentage in the multi-draft group.
On the other hand, there were twice as many students in the single-draft group who regarded making revision as bothersome (60.0%) or painful (11.4%) as students in the multi-draft group (33.3% and 4.8% respectively).
Other adjectives students in the multi-draft group used included “confusing” and
“time-consuming,” and those used by students in the single-draft group included
“boring,” and “time-consuming.”
The Open-ended Questions
There were three open-ended questions asked in the post-test questionnaire (Items 13-15), among which Item 13 was also asked in the pre-test questionnaire.
Item 13: “I □ like □ dislike to revise an old composition because_____.”
Students’ positive and negative responses to Item 13 were calculated and compared. The result is shown in Table 12.
Table 12
Participants’ Attitudes Toward Revision in the Post-test Questionnaire
Multi-draft f (%)
Single-draft f (%) Positive toward revision
Negative toward revision
35 (84%) 7 (16%)
16 (46%) 19 (54%) Note. f = Frequency.
As indicated in Table 12, at the end of the study, as high as 84% of the students in the multi-draft group liked to revise. By contrast, in the single-draft group, about half of the students liked to revise and half of them did not.
The reasons that the single-draft group liked revision can be categorized into three aspects: that revision could improve the text in form and meaning, that revision
could improve one’s writing skills as well as language ability and that revising an old composition was easier and more relaxing than writing on a new topic.
Students in the multi-draft group provided more reasons. Besides those
mentioned by students in the single-draft group, their responses included that it gave them a sense of achievement when the text read better, that it enabled them to devote their attention to one kind of writing problem a time, such as content or vocabulary and that it helped them become more aware of problems across sentences and that it provided an opportunity to explore one’s limit and try different styles. Below are some examples:
1. "The first draft was usually rough and hurried without much examination.
When revising, I could reread it and devote my attention to refining it." (M-1) 2. “I could see my scores improve from the first draft to the final draft, which
pleased me and gave me expectation and confidence.” (M-13)
3. "When I am revising, I not only check sentences and words but also change the paragraphing when necessary." (M-25)
4. "Thanks to the practice in the semester, now I understand small-scale modification may not really improve a text; sometimes an overall change is necessary." (M-22)
The reasons why the students disliked revision did not differ between the two groups, except that the reasons given by the students in the multi-draft group were more specific. The reasons included that revision was frustrating and boring and that it was not helpful to revise. Below are some examples from the multi-draft group.
1. "I wanted to revise my text but my grammar was not good enough for me to correct the errors.” (M-3)
2. "Revision gave me frustration. I felt bad about myself when required to revise again and again.” (M-17)
3. "Revising old works occupied the time for writing on new topics. We needed to practice different topics to meet the challenge of the coming big exam."
(S-10)
Item 14: "I think teacher feedback (including symbols and word comments) is __________. "
Students in both groups showed their gratitude for the teacher's efforts and affirmed the usefulness of the feedback. The responses they gave can be classified into three respects: the feedback helped them revise; they liked the teacher's generous praises as well as direct and precise criticisms with a sense of humor; the symbols were helpful in making them think and revise. Below are some examples:
1. "The teacher marked our errors with symbols instead of correcting our errors directly. I could figure out how to revise on my own and I have benefited a lot from such experience." (S-11)
2. "The teacher’s encouragement inspired me a lot. When she gave us criticism, sometimes she would draw some cute cartoon faces; this made me feel more comfortable when reading the criticism." (M-14)
3. "I loved to read the teacher’s comments. I wanted to know how the teacher felt about my compositions." (S-10)
No obvious difference was found in the responses between the two groups except the response given by one student in the multi-draft group, who particularly
mentioned her appreciation of multi-daft revising: "The teacher commented on our content and organization in our first draft and then on our grammar, vocabulary and mechanics in our second draft. This way I wouldn’t feel overloaded and too frustrated to revise" (M-22). Another student in the multi-draft group stated that he would read the teacher’s feedback more carefully since he was required to revise: "I will ponder on my teacher’s suggestions and comments instead of just giving them a quick look"
(M-4).
Item 15: "I think peer feedback is _____________."
Students had both positive and negative comments, but, again, no obvious difference was found in the responses between the two groups. Students' positive
responses were generally concerned with the following respects. First, peer review was helpful because classmates could help each other clarifying their ideas. Second, through peer review, the students' reading skills and writing skills were improved, as one student stated, "It’s helpful. When I found some interesting sentences or examples in my classmates’ compositions, I would remember them," (S-19) and another
mentioned, "The activity could develop our abilities at proofreading and reading comprehension." (S-30)
At the same time, some negative effects were also pointed out. A majority of the negative responses were concerned with student's incapability of giving reliable comments. Some students admitted that they could hardly understand their classmates' compositions due to their own poor comprehension ability; some mentioned that even though they could understand others' ideas, their English ability was inferior to their partners' and thus felt unqualified to give suggestions; still some commented that they did not really trust their classmates' suggestions because few of them were good enough to give reliable suggestions. Two observant students responded, "Whether the activity was helpful really depended on whom you discussed with" (M-19) and "We can only give suggestions or comments on the content; we are not able to comment on other aspects, like grammar or organization" (S-2). Moreover, a student in the
single-daft group said, "It’s more helpful to read compositions written by the multi-draft group than those written by the control group because the multi-draft group made revision." (S-5)
Another problem of peer review that were widely mentioned was concerning the students' unwillingness to share. Some students didn't want to share when they
thought they did not write well; some felt their writing was too awful to be read; still some just wanted to chat while they were supposed to discuss their writing.
Discussion
In this section, the results of the data analysis are discussed. I will first discuss the performance of the two groups in their writing pre-test and post-test in response to Research Questions One and Two, and then I will discuss the results of the
participants’ responses to the two questionnaires in response to Research Questions Three and Four.
The Participants’ Writing Improvement Between the Pre-test and the Post-test
The comparison of the writing scores gained between the pre-test and the post-test showed significant progress for both groups in their overall writing
performance and content, organization, grammar and vocabulary. This result indicated that both the multi-draft group and the single-draft group made significant
improvement in their writing quality even though the two groups experienced
different writing procedures. Thus, Hypothesis 1 was supported and the answer to the first research question is that both multi-draft writing procedure and single-draft writing procedure could improve students’ writing quality. In other words, the result of the study gives support to the findings reported in Chen’s (1997) study that
suggested there was a positive relationship between the frequency of writing practice and writing quality.
Further examination of the participants’ score gains in the five components revealed that both groups improved most in organization. Such improvement had brought the students a great sense of achievement. As a student responded in the questionnaire, "During the study, what gave me the most sense of achievement was my progress in my ability to organize my thoughts into a coherent and cohesive text "
(S-12). According to Chiang’s (1999) research, students’ organization can be improved through instruction. The current study supports the finding and further asserts that organization is easier to learn and apply to new topics through instruction and frequent practice than content, grammar or vocabulary.
Comparison of Participants’ Writing Improvement Between the Two Groups
The comparison of the score improvements made by the two groups between the pre-test and post-test showed that the single-draft group made slightly more progress than the multi-draft group in all the five components and the overall performance.
However, the t-tests of the comparison of the scores gained in the post-test between the two groups showed that the difference was not statistically significant. Based on this, Hypothesis 2 was rejected and the answer to the second research question is that multi-draft writing and single-draft writing were equally effective in improving students’ writing quality.
The results of the study tended to contradict some previous findings that the multi-draft writing procedure was more helpful than the single-draft writing procedure to improve students' writing ability (Chen, 1997; Ferris, 1995; Paulus, 1999).
The possible reasons why multi-draft writing was not more effective than
single-draft writing and even slightly less effective, though not significantly, are listed as follows: First, the students in the two groups were required to practice writing for equal amount of time. When students in the multi-draft writing were revising their old compositions, students in the single-daft were writing too—on a new topic. Second, students in the single-draft group had received a greater amount of feedback by the end of the study because the students in the multi-draft group wrote on three topics
while students in the single-draft group wrote on nine topics. Although the students all received feedback on content, organization, grammar, vocabulary and mechanics, students in the multi-daft group received them in separate drafts, while students in the single-draft group received them in every draft.
Comparison of Participants’ Attitudes Toward Writing Between the Two Groups
The comparison of the participants’ responses to the pre-test questionnaire
revealed that there was no significant difference in their attitudes toward writing at the beginning of the study. However, a significant difference was found in the
participants' responses to Items 1 in the post-test questionnaire between the two groups. Students in the multi-draft group were less anxious about writing English compositions than students in the single-draft group. Therefore, Hypothesis 3 was supported and the answer to the third research question is that the multi-draft writing procedure has reduced students’ apprehension about English writing while single-draft writing procedure has not.
This was probably because writing on new topics required students to come up with new materials and vocabulary and to organize all these new materials into a whole. Many students responded in the open-ended questions that this was more challenging and stressful than revising an old composition especially when not much time was allowed. By comparison, when revising an old composition, students had old materials to base on and they had teacher feedback to follow. What they needed to do was to figure out what had gone wrong and how to elaborate or correct their ideas.
Comparison of Participants’ Attitudes Toward Revision Between the Two Groups
The participants’ responses to the Likert-scale items, the multiple response item, and the open-ended questions in the post-test questionnaire all pointed out that the two groups held substantially different attitudes toward revision at the end of the study.
Students in the single-draft group had no special preference for writing on a new topic or revising an old composition, yet students in the multi-draft group strongly preferred revision to writing on new topics. Also, more students in the multi-draft group thought revision was delightful or helpful than students in the single-draft group. Less students in the multi-draft group thought revision was bothersome or painful than students in the single-draft group.
The students also indicated in their responses to the open-ended question, Item 13, that, when revising, they could reorganize their thoughts, try different styles and further explore the topic. Besides, the multi-draft writing procedure had taught them that revision required more than rewording and that sometimes changes beyond sentences or a new paragraphing was needed to improve a text.
These results revealed that revision through the multi-draft writing procedure had led students to understand what true revision was and to appreciate the benefits of revision. Hypothesis 4, thus, was supported and the answer to the fourth research question is that practice of revision through multi-draft writing procedure has resulted in deeper understanding of revision and more positive attitudes toward revision.
The above results were in consistent with Chen’s (1997) study which pointed out that students who had experienced multi-draft writing would find revision helpful and prefer revision to writing on a new topic. The results also showed that students in the
multi-draft group had understood writing as “a discovery of meanings” (Zamel, 1983).
They also had gained a new belief similar to that of Sommers’ (1980) experienced writers that writing is rewriting and were thus willing to engage in major
reconstruction during the course of their writing.
The reason for the more positive attitude held by the multi-draft group toward revision could be attributed to the more relaxing, step-by-step revision process they went through. Students in the single-draft group, however, lacked the experience and thus kept the negative ideas they used to hold about revision.
Participants’ Attitudes Toward Teacher Feedback
With regard to the teacher feedback, a majority of students pointed out the way the teacher gave feedback (marking errors with symbols and giving word comments on both form and meaning) was inspiring and helpful to them in improving their writing skills and language skills, a finding in accordance with Ferris’ (1995) study.
Students' appreciation of the teacher's efforts in giving such extensive
comments could be illustrated by some of the students' responses. One student from the single-draft group stated, "Every time I handed in my composition, I started looking forward to getting it back. I couldn't wait to know how the teacher felt about my compositions. I was always delighted to see the teacher's comments and
suggestions" (S-32). This result was in consistent with Conrad & Goldstein’s (1999) finding that students liked comments, especially those that “explain specific problems and make specific suggestions” (p. 148). Another student, from the multi-draft group, stated, "I like the way the teacher commented on our compositions. By focusing on different kinds of problems in different drafts I could handle revision better and I wouldn't get frustrated or confused." The student’s reaction supports some
researchers’ suggestion that teachers should keep their comments on form and on content separate to avoid confusing students (Sommers, 1982; Zamel, 1985). Besides, many students in both groups even stated they liked the teacher's style of giving praises generously and criticisms precisely with a sense of humor. This finding gives further support to Cardelle and Carno’s suggestion (1981) that giving a combination of criticism and praise was more effective than giving criticism alone.
Participants’ Attitudes Toward Peer Feedback
As for peer review, most students held a positive attitude toward the activity.
They generally felt it was rewarding to share and discuss their compositions. The benefits they mentioned about peer review corroborated with previous findings: it enhanced their audience awareness(Min, 2003, Mangelsdorf, 1992; Mittan, 1989;
Paulus, 1999; Tsui & Ng, 2000; Zamel, 1982), developed their reading comprehension ability (Paulus, 1999) and exposed them to different perspectives and a variety of examples and ideas (Mangelsdorf, 1992; Min, 2003; Mittan, 1989; Tsui & Ng, 2000;
Paulus, 1999; Zamel, 1982).
On the other hand, the students’ major concern about the peer review session was the person with whom they discussed their compositions. As Mangelsdorf (1992) observed, students could learn better if they shared and discussed with students with a higher level of English proficiency. Another problem of peer feedback was that, as previous students mentioned (Nelson & Carson, 1998; Paulus, 1999), some of the students were passive, indifferent, chatty, or unwilling to share, which could make the activity unproductive.
In brief, although many students insisted that they preferred teacher feedback because it was more reliable, helpful and important to them than peer feedback, a
claim that was consistent with previous studies (Mangelsdorf, 1992; Nelson & Carson, 1998; Paulus1999; Tsui & Ng, 2000; Zhang, 1995), the peer feedback was still
beneficial in many respects and students expected to share and discuss. Therefore, the current study supports Paulus (1999)’s suggestion to make peer and teacher feedback an integral part of the multi-draft writing classroom but asserts that peer feedback be conducted only to complement teacher feedback.
Summary of the Results and Discussion
In this chapter, I have presented the results of the participants’ writing
performance in the writing pre-test and post-test. I have also presented the results of their responses to the pre-test and post-test questionnaires. Then, in the second part of the chapter, I have discussed the participants’ writing improvement. I have also discussed the participants’ attitudes toward writing and revision, teacher feedback and peer feedback.
