Abstract
This study investigated the impact of revision and teacher indirect feedback on the acquisition of the present unreal conditional in a Chinese EFL college classroom context. Four research questions were addressed in the current study: 1) Does revision produce a positive effect on the formal accuracy of the target form when Chinese EFL university students receive teacher feedback? 2) Do Chinese EFL university students who are asked to revise their writings outperform those who are NOT in the formal accuracy of the target form when their errors are underlined? 3) Does teacher feedback generate a positive effect in the formal accuracy of the target form when Chinese EFL university students are engaged in the process of revision? 4) Do Chinese EFL university students who receive teacher feedback outperform those who receive NO teacher feedback in the formal accuracy of the target form when they are required to revise their subsequent writings? Ninety-three college students participated in this study. They were divided into three groups: 1) the Revision Group; 2) the Feedback and Revision Group; and 3) the Feedback Group. Results of the study suggest a very positive role of revision combined with teacher indirect feedback in the Chinese EFL college context. Further, they clearly show that receiving teacher feedback without the engagement of revision tasks does not improve accuracy in such a classroom.

Key words: teacher feedback; revision; formal accuracy

1. Introduction
There has been a heated debate on the role of teacher feedback in the field of second language writing research. There are people who believe in giving corrective feedback to students to improve their written accuracy and those who do not (Gue´nette, 2007). Ferris (1999, 2004) criticized the strong claim held by Truscott (1996, 2004) against grammar feedback and maintained that “it would certainly be premature to formulate any conclusions about this topic” (Ferris, 2004, p. 49) as existing research does not adequately address the issue whether grammar feedback is beneficial to L2 student writers or not. If we consider the role of teacher feedback from the perspective of learners, their views tend to be unanimously for the pedagogical practice of teacher feedback. According to Leki (1991), grammar feedback is viewed as helpful by college level ESL students. Hyland expressed a similar view that grammar feedback can serve as guidance for eventual writing development as far as students are concerned (2003). Acknowledging the uncertainly from the existing research data, it is legitimate to further address this role of teacher feedback in L2 writing.  

Some research findings have suggested a positive role of indirect teacher feedback, when it incorporated with student self-revision1, in facilitating accuracy in L2 wiring. Ferris and Roberts (2001) examined the differential effect of teacher feedback among university ESL student writers in terms of the percentage of errors they could revise when they self-edited their texts across three feedback conditions: (1) errors marked with codes; (2) errors underlined with no codes and (3) no feedback at all. Results showed large beneficial effects for feedback groups versus the non-feedback group. However, differences in terms of the proficiency level between the feedback group and the non-feedback group might be a confounding factor in their study (Ferris & Roberts, 2001).

Lee (1997) obtained similar results by investigating the performance of ESL college students in Hong Kong: Students corrected more errors when the errors were underlined than errors were slighted indicated or with no indication at all. One of our recent studies (Li & Sun submitted) revealed a gain in formal accuracy in the use of the mixed type of unreal conditional when Chinese college students received indirect teacher feedback. Our study suggested that teacher feedback relevant to a linguistic form focused in revision tasks could be useful in bridging the gap between interlanguage and the target language. Further, indirect feedback in the form of underlining linguistic errors together with classroom instruction seemed to be more desirable than indirect feedback alone. However, it remained unclear whether this effect would last.
Chandler (2003) examined whether teacher feedback in the form of underling errors could help East Asian college students improve their writing accuracy and whether the effects would last over one semester. The results of this study demonstrated that formal accuracy of student writing improved significantly if the participants were required to correct their errors than if they were not. Moreover, results also exhibited a gain in accuracy which was not accompanied by a decline in fluency over the semester. Findings of the study signified the importance of student correction or revision, and it might be possible that “if students did not revise their writing based on feedback about errors, having teachers mark errors was equivalent to giving no error feedback” (Chandler, 2003, p. 280).

However, in many Chinese college English classrooms, student self-revision has not been integrated with the provision of teacher feedback for some practical reasons2. So, we do not know whether teacher feedback would be beneficial to college student writers if the chances of performing revision tasks are rare. This study is intended to investigate the role of revision and teacher indirect feedback on the increase of formal accuracy of a target form in a Chinese EFL college classroom context. In particular, it is designed to determine the effect of teacher indirect feedback when college student writers are not required to revise their writings.

2. The study
2.1 Research questions
This study attempts to answer four research questions: 1) Does revision produce a positive effect on the formal accuracy of the target form when Chinese EFL university students receive teacher feedback? 2) Do Chinese EFL university students who are asked to revise their writing outperform those who are NOT in the formal accuracy of the target form when their errors are underlined? 3) Does teacher feedback generate a positive effect in the formal accuracy of the target form when Chinese EFL university students are engaged in the process of revision? 4) Do Chinese EFL university students who receive teacher feedback outperform those who receive NO teacher feedback in the formal accuracy of the target form when they are required to revise their subsequent writings?
Revision in this study refers to any changes made with respect to the target form (i.e. the present unreal conditional)3 that student writers make in their writings. Besides, teacher feedback examined in this study involves underlining linguistic errors specific to the target form by the class teacher. 

2.2 Method
2.2.1 Participants
Ninety-three participants were sampled from 150 sophomores, who came from four-year degree programmes in the College of Bioengineering and the College of Material Science and Engineering in Chongqing University, PR China. The English course was a required one and it was designed to develop their general English proficiency with a communicative orientation. The students had three English lessons each week. It needs to note that writing was taught integrated with other language skills. Ages of the participants ranged from 18 to 20. They all learnt English in formal classrooms and none of them had any study abroad experience at the time of the study.  All the participants were at the same English proficiency level with similar educational background.

2.2.2 Procedures
Before the study, participants were carefully sampled by a Grammar Test. In the Grammar Test, the students were required to complete 30 multiple-choice questions with 5 questions focused on the target form, 5 questions on the future unreal conditional and the rest as distracters. For instance: 
If I were in the movie, then it _____ about time that I buried my head in my hands for a cry.
A. would be      B. is       C. will be      D. was
Only the scores of the target form were analyzed, and the maximum score was 5 points. Anyone with a score of 0 or 5 was excluded, leaving a total of 93 participants for the study. They were randomly assigned to three treatment groups (for Revision Group n = 31; for Feedback and Revision Group n=32; for Feedback Group n=30). There was no significant difference between the mean scores of the three groups (F=1.210; p =.301). 
In the study, the participants went through two phases (see Table 1). Phase 1 was designed to encourage all the participants to produce and notice the target form freely without teacher feedback in an output and input treatment. They were first asked to write a composition in 30 minutes using at least five sentences with the target form on topic 1 (What Would Happen If There Were No Water?). After the writings were collected, a model composition was provided. Phase 2 was designed so that each group received a different treatment. The Revision Group revised their writings on topic 1 without receiving any teacher feedback, while the Feedback and Revision Group revised their writings with their errors of the target form underlined by their class teacher. The Feedback Group received their writing on the same topic with errors underlined. Three weeks after the treatments were administered, the three groups were asked to write a composition using the target form on a new topic (What Would You Do If You Didn’t Go to College?)

Table 1
Design of the study

Task 1: Grammar Test (Week 1: 15 min.)
Task 2 (Topic 1): Writing on topic 1 (Week 3: 30 min.)
Task 3: Reading a model composition (Week 4: 15 min.)
Revision Group Task 4: Revising writings on topic 1 (Week 4: 30 min.)	Feedback and Revision Group Task 4: Revising writings on topic 1 with errors underlined (Week 4: 30 min.)	Feedback Group Task 4: Reading writings on topic 1 with errors underlined (Week 4)
Task 5 (Topic 2): Writing on topic 2 (Week 7: 30 min.)

2.2.3 Scoring method
Similar to one of our recent studies (Li & Sun submitted), the scoring system was modified from Izumi, Bigelow, Fujiwara & Fearnow (1999). Although the writing direction required the participants to produce at least five sentences using the target form, the actual number of sentences written using the target form varied from one participant to another. So, accuracy of the form was scored as follows: the percentage of correctly formulated target items of the present unreal conditional divided by the total number of the target items in all the target-like expressions attempted.

For the correctly used target items, four component features of the target form were considered: in the if-clause, 1) the past tense; and in the main clause, 2) a modal, 3) the past tense and 4) the original form of verb. For each target form item, 1 point was given for its presence and 0 point for its absence4. If a participant wrote five sentences using the target form and each sentence was correctly written, a total score of 20 (5 sentences×4 points per sentence) was given. Then the score for this writing would be 100% (20÷20=100%).

2.3 Analyses and results
Descriptive data from the three groups collected in Phase 1 and 2 were presented in Table 2. To determine the effect of revision versus no revision on the gain in formal accuracy, a two-way ANOVA with repeated measures was conducted based on the mean scores obtained by the Feedback and Revision Group and Feedback Group with treatment (feedback and revision vs. feedback) and writing tasks (writing on topic 1 vs. writing on topic 2) as variables. The analysis showed a main effect of treatment (F=4.556, p=.037<.05), and writing tasks (F=12.915, p=.001<.05), but the interaction was not significant (F=3.433, p=.069>.05). Further, results of a paired-samples t-test on the mean scores of the Feedback Group obtained in their writings on topic 1 and topic 2 did not reveal a significant difference (t=1.037, p=.308>.05). So, the main effect of writing tasks was contributed mainly by the Feedback and Revision Group.
Table 2

Descriptive data of the three groups obtained in tasks of Phase 1 & 2

Groups	Tasks	Mean (%)	Std. Deviation	Minimum (%)	Maximum (%)
Revision	Writing on topic 1	78.12	15.77	38.89	100.00
	Revision task	81.87	15.70	44.44	100.00
	Writing on topic 2	79.54	15.74	41.18	100.00
Feedback and Revision	Writing on topic 1	74.38	15.15	47.62	100.00
	Revision task	89.52	14.12	47.83	100.00
	Writing on topic 2	88.96	13.57	57.14	100.00
Feedback	Writing on topic 1	73.42	17.49	45.00	100.00
	Writing on topic 2	78.08	14.34	47.62	100.00

Results of the above analyses show that when revision was required, participants in the Feedback and Revision Group demonstrated a gain in accuracy of the target form from writings on topic 1 to topic 2 (74.38% vs. 88.96%) when their errors relative to the target form were underlined. However, participants in the Feedback Group failed to exhibit a significant increase when revision was not required (73.42% vs. 78.08%). Revision generated a better effect on writing accuracy of the target form as the Feedback and Revision Group outperformed the Feedback Group.

To examine the effect of feedback versus no feedback on the gain in formal accuracy, a two-way ANOVA with repeated measures was conducted based on the mean scores obtained by the Revision Group and Feedback and Revision Group with treatment (revision vs. feedback and revision) and tasks (writing on topic 1, revision task vs. writing on topic 2) as variables. The analysis did not show a main effect of treatment (F=2.259, p=.138>.05), but it showed a main effect of tasks (F=12.589, p=.001<.05). Results of pairwise comparisons showed a very significant difference between the mean scores of writing on topic 1 and revision task obtained by the two groups (p=.000<.05). Moreover, results revealed a significant difference between the mean scores of writing topic 1 and 2 obtained by the two groups (p=.001<.05), but the difference between the mean scores of revision task and writing on topic 2 did not reach the level of significance (p=.54>.05). Finally, the interaction between the two main effects was significant (F=2.259, p=.003<.05).

Results presented in the previous paragraph means that the Revision Group improved the rate accuracy from writing on topic 1 to the revision task (78.12% vs 81.87%) and this is also true to the Feedback and Revision Group (74.38% vs. 89.52%). Those two groups also increased their rate of accuracy from 78.12% and 74.38% respectively to 79.54% and 88.96 respectively from their writings on topic 1 to writings on topic 2. Further, the Feedback and Revision Group outperformed the Revision Group as the former group demonstrated a higher rate of accuracy in both the revision task (89.52% vs. 81.87%) and writing task on topic 2 (88.96% vs. 79.54%).

2.4 Summary and discussion
In light of the results presented above, four major findings emerged from the study:

1. Revision improved accuracy of the target form when the participants received indirect teacher feedback in the form of underlining errors relative to the target form.
2. Revision generated a better effect than no revision treatment condition on writing accuracy of the target form as the Feedback and Revision Group outperformed the Feedback Group when the participants were asked to produce the target form in a new writing topic. Feedback alone did not produce a significant effect on the accuracy of the target form.
3. Teacher feedback generated a positive effect on the accuracy of the target form when the participants were asked to revise their writings. The participants demonstrated a gain in accuracy from their writings on topic 1 to the revision task as well as from their writings on topic 1 to a new topic.
4. Participants in the feedback treatment condition showed an advantage over those with no feedback when they were engaged in the revision task as the former group outperformed those in the latter group in both the revision task and writing task on a new topic. 

The first two findings, summarized above, suggest a positive role of revision when grammar errors specific to the target form were indicated by teacher indirect feedback in a Chinese college English classroom. What is more, they demonstrate that receiving teacher feedback without the engagement of revision tasks is not effective in such a classroom. The evidence observed in this study lends support to the current view in SLA which posits that output promotes learners to notice the gap between their interlanguage and the target language, drawing learners to construct a hypothesis about the target form, which will be presented in their follow-up language production, and tested in their following output (Swain & Lapkin 1995). To the students who were required to revise their writings on topic 1, producing a revised output engaged them in looking for solutions to their linguistic problems in ways that were appropriate to the context of the revision task, making the provision of teacher feedback a meaningful pedagogical activity. In the process of revision, student writers might attend to the linguistic errors which were underlined by the teacher and subsequently search for possible ways to solve their problems. So, the act of revision makes learners more aware of the mismatches between their own language and the target language, hence leads to a careful review of their interlanguage capabilities, which further leads to a gain in accuracy in subsequent writings.
The beneficial effect of revision is consistent with the results reported in the first study of Chandler (2003), which also shows an advantage of student correction over no correction in the improvement of accuracy and college student who did no error correction between writing assignments did not increase in accuracy. It is also important to note that the advantage of correction shown in Chandler (2003) did not result in a decline in fluency over the semester.

Considering the positive role of revision observed in this study and Chandler’s study, it is legitimate to stress the incorporation of revision in writing tasks in Chinese college English classrooms.

The effect of feedback versus no feedback revealed by the last two findings of the study suggests that indirect teacher feedback, in the form of underlining grammatical errors, facilitates accuracy in L2 wiring if it is integrated with student self-revision tasks. Such effect extends the results reported in Li & Sun (submitted), as the effect revealed in the current study lasted over a period of six weeks. Findings imply that teacher feedback relevant to a linguistic form focused in revision tasks plays a positive role in promoting the accuracy of the form for Chinese EFL university student writers. By indicating linguistic problems, teacher feedback is useful as it may lead to a more form focused learning process when the general intent is still kept on expressing meanings in writing. Further, findings also support the provision of teacher feedback in Chinese EFL teaching at college level as an integral component of classroom practice. 

3. Conclusion
So, to round up the results revealed in this study, this study shows a beneficial role of revision and teacher feedback, relevant to the linguistic form of the present unreal conditional, in promoting the formal accuracy for Chinese EFL university student writers. The study clearly shows that receiving teacher feedback without the engagement of revision tasks is not effective in facilitating accuracy in such a classroom. As teacher feedback is expected to be an important component of English instruction, favored by many Chinese-speaking English learners in a similar learning context, results of this study indicate the value of teacher feedback when, and only when it involves revision tasks of those learners. To conclude, we believe that it is useful to design more classroom activities to involve student writers in the process of self correction or revision as a natural part of writing tasks in the Chinese EFL setting.
Notes

1. Revision has been termed as a goal-oriented thinking process, which a writer goes through in order to make possible changes to what is written. It is broader than editing for errors (see Williams, 2004).
2. In many college English courses, designed for non-English majors in China, writing is taught integrated with other language skills. Normally, revision is not incorporated with classroom writing tasks as it is too time-consuming to consider for classroom practice.
3. According to the class teacher, the target form (e.g. If she knew French, she would not ask me for help) was relatively hard for the students mainly due to two factors: 1) a lack of use and 2) a subtle difference between this form and the future unreal conditional, e.g. If he were to/should call me tomorrow, I would let him know.
4. For example: 1) If there were no electricity, the world would be different. Score for this sentence is: 1+(1+1)+1=4.

References
Chandler, J. (2003). The efficacy of various error feedback for improvement in the accuracy and fluency of L2 student writing. Journal of Second Language Writing, 12, 267-296.

Ferris, D. R. (1999). The case for grammar correction in L2 writing classes: A response to Truscott. Journal of Second Language Writing, 8, 1-11.

Ferris, D. R. (2004). The ‘‘Grammar Correction’’ Debate in L2 Writing: Where are we, and where do we go from here? (and what do we do in the meantime . . .?). Journal of Second Language Writing, 13, 49-62.

Ferris, D. R., & Roberts, B. (2001). Error feedback in L2 writing classes: How explicit does it need to be? Journal of Second Language Writing, 10, 161–184.

Gue´nette, D. (2007). Is feedback pedagogically correct? Research design issues in studies of feedback on writing. Journal of Second Language Writing, 16, 40-53.

Hyland, F. (2003). Focusing on form: Student engagement with teacher feedback. System, 31, 217-230.

Izumi, S., Bigelow, M., Fujiwara, M., & Fearnow, S. (1999). Testing the output hypothesis: Effects of output on noticing and second language acquisition. Studies in Second Language Acquisition, 21, 421-452.

Lee, I. (1997). ESL learners’ performance in error correction in writing: Some implications for college-level teaching. System, 25, 465-477.

Leki, I. (1991). The preferences of ESL students for error correction in college level writing classes. Foreign Language Annals, 24, 203-218.

Li, H., & Sun, L. (Forthcoming) Teacher feedback in L2 writing output tasks: The case of Chinese EFL writing classrooms. Asian EFL Journal.

Swain, M., & Lapkin, S. (1995). Problems in output and the cognitive processes they generate: A step towards second language learning. Applied Linguistics, 16, 371-391.

Truscott, J. (1996). The case against grammar correction in L2 writing classes. Language Learning, 46, 327-369.

Truscott, J. (2004). Evidence and conjecture on the effects of correction: A response to Chandler. Journal of Second Language Writing, 13, 337-343. 

Williams, J. (2004). Tutoring and revision: Second language writers in the writing center. Journal of Second Language Writing, 13, 173–201.