The present research aimed to investigate the effects of two types of the teacher response (TR) on students’ revision and subsequent writing qualitatively and quantitatively.
To access students’ writing performance in terms of grammar and content, participants were separated into two groups, one of which received TR in the form of a question whereas the other group, direct correction as TR. Unlike previous studies which targeted college learners, this study examined EFL students at a senior high school. With three rounds of writing, students wrote an essay, then revised this essay after one week based on the TR they received, and finally wrote a new essay three weeks later. They were also asked to make a retrospective protocol. After finishing this writing cycle, they filled out a questionnaire in which items were designed to access their attitude towards these three writing experiences.
Participants
This study was experimented in a public senior high school in Hsinchu City. The participants are 40 male 10th graders, who had already learned the usages of tense and subjunctive mood when this study was undertaken. They had also been taught the need of using a thesis statement when writing an English essay. In this study, they were randomly categorized into two groups, 20 receiving TR in the form of a question and 20 receiving direct correction as TR. The English proficiency of the two groups was roughly equal based on the average English grade in the high school entrance exam, 58 and 57 respectively1, and their English academic performances of the semester this study proceeded. Moreover, except
1 The total score of the English exam in the entrance examination is 60.
for personal few writing experiences in junior high school, all participants had no other experience to write an English essay in class; that is, English writing was new to them.
Types of Teacher Response in the Study
As discussed in the previous chapter, L2 writers, unlike L1 writers, are prone to focusing on grammar concepts. Because of L2 students’ shortage of vocabulary and lack of thinking in target language, L2 students’ essays—take Taiwanese students for example—are often rife with Chinglish and insipid description. In the present study, although the participants have studied English at least for six to nine years, they just start to learn to write short essays. As a result, the written TR adopted in this study focused on both local and global issues.
Generally, participants in the experimental group were treated with TR in the form of a question and those in the control group received direct correction as TR. The former were given questions focusing on tense usage (local issue) and thesis statement (global issue) (see Appendix A). In contrast, the latter were directly given answers to their problems with tense usage and thesis statement. Then, the performances of the participants in two groups were compared.
In detail, for students in the experimental group, errors related to tense were indicated in questions and students were asked to revise their first draft based on these questions. Since there might be other grammatical errors, the teacher had not marked them until students turned in their revision, so that students would not think their first essay was error-free. The same criterion was applied to the control group: errors in relation to tense were corrected directly; other grammatical errors would be untreated in the first draft but they were corrected in the revised draft.
As for the TR centering on the thesis statement, students’ essays in the experimental group without or with an inappropriate thesis statement were given a question which asked
them where the thesis statement was. In contrast, those in the control group were directly presented a thesis statement if it was missing in their essays (see Table 1).
Table 1 Research Design: Two Types of TR in the Two Groups
Focus of TR Group of Question-based TR Group of Direct-correction TR Grammar-focused TR Ask a question based on the
error type related to tense.
Directly correct errors related to tense usage.
Content-focused TR Ask the question “Where is your thesis statement?”
Directly offer a thesis statement.
More specifically, grammar-focused TR was used when students made mistakes in the following forms: (1) a wrong tense was used and (2) a correct tense was used but written in the wrong form. For instance, in the experimental group, a question was prompted as students were found using simple present tense in a subjunctive clause in which simple past tense should be used, or as they used the wrong verb form (see Table 2).
Table 2 TR for Errors in Tense Usage
Error Types in Tense Example Question-as-response 1. A wrong tense was used. If I am rich, I would travel
around the world.
(am Æ were)
How do you make a hypothesis for something that does not really happen in the present?
2. A tense was written in the wrong form.
I have been study English for 10 years.
(study Æ studying)
What verb form should be used when you use present perfect progressive tense?
Procedure
In the first week of the study, participants were asked to write an English essay of 90 to 110 words in class. During this 50-minute span, they were allowed to use dictionaries for spelling-check but not allowed to consult other reference books such as English textbook or grammar manuals. Additionally, before they started to write, the teacher spent several
minutes reviewing how to construct a short essay, indicating specifically that an English essay should contain a clear “thesis statement.” The first essay was entitled “A Day of Bad Weather.” This title was adopted because it may elicit the use of various tenses. For example, while recalling one personal experience in a rainy day, students must use past tense.
In the second week, all students received their own drafts with written TRs on them; in class, they were asked to revise the essay according to the responses they received. Again, they were not allowed to consult grammar books. In addition, while they were making their revision, they were also asked to make a retrospective protocol (see Appendix B).
The protocol contains two parts: part A mainly focuses on students’ immediate response to the TRs on their first draft as soon as they read them. In detail, students were asked to find out sentences with errors of tense usage; meanwhile, they wrote down their reaction to the TR on their first draft in response to the question “How and what do you think about the teacher’s response?” Besides, students were also asked in the part A to recall why they used the wrong tense(s) in the context(s) of the essay2. In part B, students’ retrospective thinking on how they organized their first draft is focused. In detail, they were asked to recall whether they organized their essay on the basis of their thesis statement3.
In both groups, the revised drafts were returned to the students the next week with all errors directly corrected by the teacher. This was done because most participants in this study were inexperienced in English writing and errors should be pointed out for these novice writers, or they were very likely to make the same grammatical mistakes next time. Besides grammatical correction, a proper thesis statement was offered in the revision in the hope of showing participants how to write a thesis statement for an essay.
Three weeks after they made their revision, a new essay was assigned in class. The second essay was entitled “An Experience that Affects Me a Lot,” which was anticipated to
2 Students were asked “Why did you write it with this tense?”
3 Students were asked “How did you organize your essay?” and “Did you follow an organizational pattern?”
elicit past experiences of the students. Also, students may describe in the essay how their personal experience has influenced them. Moreover, when writing “A Day of Bad Weather” in the first essay, students were told that one possibility to develop their essay was the use of hypothetical statements such as “If yesterday had been a sunny holiday, I ….” In this new essay, they were also suggested to write their reflection as a kind of wish statement such as
“If I were able to reset my life, I ….”
Finally, once students received their new essay with TRs, a survey (see Appendix C) was conducted using a questionnaire to access their attitudes towards this writing cycle. Some students were interviewed individually if the information written in the earlier retrospective protocol was insufficient or unclear.
Data Analysis
The data of the present study include students’ three writing drafts, their retrospective protocols, the questionnaires and the interviews. From these data, we hoped to extract (1) group performances of the three drafts, (2) students’ writing process and revising process, and (3) students’ attitudes towards teacher response.
In general, group performances were accessed by examining students’ three drafts as well as their error profile4 quantitatively. The error profile was made based on students’ three drafts (see Appendix D). The qualitative analysis of the error profile provided plenty of raw data which helped scrutinize the writing process of the participants. Students’ revising process was accessed via retrospective protocols which were analyzed qualitatively to support or further explain not only the statistics obtained from the analyses of the three drafts but also students’ perception of the TR they received. Finally, students’ attitudes towards TR and this writing procedure were gained by analyzing the questionnaire quantitatively and
4 The error profile was made to record every student’s errors in tense usage and whether the error was corrected with the help of grammar-focused TR. Besides, it also recorded the drafts with or without a proper thesis statement. In other words, the error profile serves as a database for further qualitative analyses, if necessary.
qualitatively. The interview was conducted, depending on whether the information gathered from above data was sufficient or not.
In detail, we gathered qualitative information from students’ three drafts of two essays and their error profile. The error reduction as a result of grammar-focused TRs over the three drafts was first analyzed; therefore, every student’s ratio of the amount of errors in tense usage to the amount of total clauses in the first essay, in the revised draft, and in the new essay was calculated respectively. The error ratio served as an indicator of the effectiveness of the grammar-focused TR; a lower ratio suggested greater effectiveness of the grammar-focused TR. In addition, the number of drafts containing a thesis statement over the three drafts was recorded. The increasing number in the revised draft would mean the effect of the content-focused TR. The more essays were found to have a thesis statement in the revised drafts, the more effective the type of the TR was.
With above statistics resulting from grammar-focused TR and content-focused TR, we then compared students’ performances between first draft and the revised draft so that we could obtain the immediate (short-term) effect of the TR with the different focuses in quantitative terms. Therefore, in each group, the fluctuation in students’ error ratios between the first draft and the revision because of grammar-focused TR helped answer the research question “Does grammar-focused TR help reduce students’ grammatical errors in revision?”
In the same vein, within each group, the change in the number of essays containing a thesis statement between the first draft and the revised drafts answered the question “Does content-focused TR lead to students’ writing progress in text organization?”
Second, to see the immediate effect of two types of TR—TR in the form of a question versus direct-correction TR—we compared two groups in terms of (1) group decreasing error ratios in the revised draft and (2) group increasing numbers of the draft with a thesis statement in the revision. These two comparisons reveals the effect, from a quantitative perspective, of the two types of TR on students’ revision, thus answering the research
question “Do different types of TR make differences in students’ revision?”
After the analysis of immediate effect of two types of TR with two different focuses, we next probed into whether this short-term effect could sustain in the new essay. Therefore, we examined the participants’ new essay, hoping to find that both types of TR would show their carry-over effects of the grammar-focused TR and of the content-focused TR. The same analytic procedure was applied in the new essay: errors in tense usage as well as the existence of a thesis statement in the new essay were calculated and checked. The results were compared with those from the analyses of the two drafts of the first essay; the findings in the new essay were used to answer the research question, “Does the effect of TR on students’
revision sustain?”
Besides quantitative comparison, students’ writing and revising processes were explored by means of a qualitative analysis in students’ retrospective protocols. The protocol was designed to elicit students’ reflection on (1) how they responded to TR upon reading them, (2) why they used tense incorrectly or improperly in the first essay, and (3) whether they tended to misuse certain tenses in certain contexts. As a result, analyzing students’
retrospective protocol helped penetrate how participants understood the TR they received, and answered the research question “How do they perceive TR (they received)?” Additionally, examining the protocol could see whether students organized their essay with a thesis statement so that we could further understand why students would not place a proper thesis statement after they received the content-focused TR. In brief, the qualitative analyses of the retrospective protocols revealed students’ cognitive activities when they were revising. It is anticipated that from protocol, we could probably find the reason for which or the context in which students used wrong or improper tense, or were unable to provide a proper thesis statement. Besides protocol, students’ error profile was also referred to timely so as to provide useful detail for additional explanations of students’ writing and revising process.
Finally, the questionnaire served an access to students’ attitudes toward this writing
cycle. The questionnaire was analyzed quantitatively and qualitatively to understand (1) students’ attitudes towards the writing procedure of the three drafts; (2) their perception of the TR and attitudes towards the TR they received; and (3) their self-evaluation of the usefulness of the TR as well as their ability to apply TR to the new essay. The rating of each question in the questionnaire provided the quantitative perspectives on the research question “What attitudes do students have towards TR?” On the other hand, we also analyzed in a more qualitative way; as a result, students’ answers to open-ended questions in the questionnaire, their written information in the protocols, and their responses in the interviews which were arranged with some students in order to fully understand their replies on the protocol and the questionnaire were all adopted to help us understand how students perceived TR and how they were influenced by the TR they received.
Validity
In addition to the researcher herself, a second rater was invited to ensure the inter-rater reliability in the whole experimental procedure. The second rater is an experienced senior high school English teacher who has taught English at least for eight years. Good at English writing and English writing instruction, the second rater helped double-check the TRs implemented in this study. Thus, the inter-rater reliability was confirmed before each of the three drafts was returned to the participants.
The construct validity of the responses used in this study is based on the agreement of the researcher and the second rater. All question-based TR were designed by the researcher and then checked by the second rater. They collectively discussed and checked whether the questions were clear enough and understandable for senior high school students. The content-focused TR used in the direct correction group—thesis statement—was also confirmed based on the discussion between the researcher and the second rater.
As mentioned previously, the corrected items in the present study were limited to errors
resulting from the misuse of tense and the neglect of a thesis statement. Thus, other grammatical errors were not identified in the first essay in order not to distract students from the focused grammatical items5. The second rater also assisted in double-checking the counting jobs. The counting of clauses with errors throughout the three drafts and the tracking of students’ error profile were reconfirmed by the same rater as well.
5 In revising the first draft, participants were reminded that they should double-check their spelling and word choice. They were well informed that even though the teacher did not respond to these errors, other errors might still exist on their paper. These errors would be corrected by the teacher in their revised draft.
CHAPTER FOUR RESULTS
The present study was conducted with forty participants divided into two groups:
twenty in the group of question-based TR (N1-N20) and twenty in the group of direct-correction TR (N21-N40). Students’ performances across their first essay, revised draft, and the new essay were all recorded in their error profile. On the basis of the data gathered from students’ error profile, protocols, questionnaires, and interviews, the findings in this study are presented in the following subsections in the sequence of immediate effects of TR, carry-over effects of TR, and students’ attitudes toward TR. More specifically, quantitative results are shown first, followed by qualitative findings to further support or explain the data from quantitative analyses. These qualitative findings are mainly excerpted from students’
responses in the interviews.
The Influence of TR on Students’ Revision
First, group performances in terms of reduced error ratios are compared. Table 3 shows the error ratios of the two groups in the first draft and its revision.
Table 3 Error Ratios in the First Draft and Its Revision
Group First Draft Revision Decreasing Rate
Group of Question-based TR (QB Group) 10.5% 2.5% - 8%
Group of Direct-correction TR (DC Group) 14.6% 0.5% - 14.1%
Table 3 shows that students in the experimental group had a decrease of 8% in error ratios in the revision (from 10.5% to 2.5%). The lower error ratio in the revision in the
question-based group thus reflects that TR had an immediate impact on students’ writing, facilitating their revision. On the other hand, with a reduction of 14% in tense errors, the error ratio in the direct-correction group drops even more drastically across the two drafts (from 14.6 % to 0.5%), which suggests almost all students in this group, after receiving direct correction, had properly dealt with their errors in tense.
In fact, it is reasonable that the group of direct-correction TR performed better than the group treated with question-based TRs. With teacher’s correction of tense errors, there are better chances that students in this group copied every red mark from their first draft to the revised one. However, students who received questions as TR had to spend some time
In fact, it is reasonable that the group of direct-correction TR performed better than the group treated with question-based TRs. With teacher’s correction of tense errors, there are better chances that students in this group copied every red mark from their first draft to the revised one. However, students who received questions as TR had to spend some time