Review M c l e on "The Effectiveness of Teacher's Written Feedback on L2 Writing"

Eunjung Park
(Seoul National University)

Park, Eunfung 2006. Review Article on "The Effectiveness of Teacher's Written feedback on L2 Writing." SNU Working Pnpers in Englisll
Lingrristics nnd Lanpage 5, 61-73. This paper reviews previous studies on the effectiveness of three types of teacher's written feedback: form-focusd, content-based, and integrated feedback. More specifically, the study pursues the following issues: Why certain types of feedback are put forth, how they are performed in students' writing and what the main issues are in this field. Through the extensive review on the issues suggested, rm this study attempts to help LZ teachers choose f o different types of written feedback considering the advantages and disadvantages of each feedback type to more effectively improve their students' writing skills. (Seoul National University)
Keywords: written feedback, integrated feedback

form-focused,

content(meaning)-focused,

1. Introduction
As it becomes widely accepted that four skills including reading, listening, speaking and writing should be well-balanced for learners' actual language development, writing education has been of a great concern among L2 researchers these days. From the same perspective, language tests like TOEFL (Test of English as a Foreign Language) and TOElC (Test of English for International Communication) decided to include writing skills as one of evaluation i terns for learners' English abilities and learners became more interested in developing their writing skills. Simultaneously, teachers and researchers in this field have begun an attempt at finding a more effective teaching method to improve their students' writing performance. In a L2 writing course, written feedback on student writing has

been considered as one of the essential parts and, thus, even though it requires a great deal of efforts and time to give written feedback, teachers seem to think that it is a kind of duty as a writing instructor. According to some research (Song 1998, Ferris & Roberts 2001, Ferris 2004), students prefer this because they regard teachers' written feedback as crucial to their improvement as writers. Among the different types of teacher's written feedback, form-focused feedback (or grammar correction) is one of the most commonly used methods from the past. This feedback mainly focuses on student's grammatical knowledge and teachers give corrections in only grammatical features. It is believed that, through the process of feedback, L2 students come to be aware of what kind of grammatical errors they often make and they come to acquire grammar rules, which leads them not to make the same errors in subsequent writings. On the other hand, some other researchers have questioned the effectiveness of form-focused feedback. For instance, Cohen and Robbins (1976) reported that three ESL advanced students receiving written grammar feedback showed no significant improvement on errors later on, so that they argued that grammar correction didn't affect the improvement of learners' writing skills at all. Zamel (1985), who disagreed with the practice of grammar correction, also criticized teachers using the method. In result, different types of written feedback were suggested. Among them, many researchers began to use feedback not focusing on surface features like grammar but more paying attention to logical fallacy or content quality of student writing and comment on this respect. This method is usually called content or meaning-based feedback. Through experiments or by theoretical grounds, some researchers offered evidence that this method is much more effective than form-focused feedback. Researchers not only tried on other types of written feedback in L2 writing class, but also would like to reveal the relationship between types of feedback and the effectiveness. First of all, form-focused feedback and content-based feedback have been compared in many studies. However, most of them ended up yielding conflicting results depending on research. Therefore,

despite continuous attempts to find out the most effective type of feedback, there still remain lots of controversy concerning this matter even now. For this reason, this paper attempted to do extensive reviews of previous literature concerned with the usefulness of different types of teacher's written feedback, and further sought for the answer about which type of written feedback can enhance L2 students' writing skills most. Even though a variety of feedbacks have been proposed up to date, this paper mainly dealt with three kinds of feedback because they are most commonly used in L2 writing class. They are form-focused feedback, content or meaning-based feedback and integrated feedback combined grammar correction with content-based feedback Additionally, several other forms of teachers' written feedback were briefly introduced in the latter part of this paper.

2. Three types of teachers' written feedback

2.1 Form-focused feedback
First of all, let me look through the research on the first type of teachers' feedback, form-focused feedback or grammar correction. As stated above, even though most of I teachers have been using this 2 type of L2 writing instruction for such a long time, the effectiveness has remained one of most controversial issues among L2 scholars. Burt (1975) was one of those who first cast doubt on grammar correction, claiming that no current standards seem to exist on whether, when, which, or how learners errors should be corrected or who should correct them. Hendrickson also turned to the problem of teacher's error correction in grammar, approaching to this matter with more theoretical grounds. In a work published in 1978, he reviewed available previous research and concluded that little was known about the efficacy of grammar correction. Posing pessimistic attitude toward the practice of teachefs correction on oral and written errors, he pointed out that the practice lacks theoretical grounds and is rather speculative. He added that even if form-focused feedback may be beneficial to students in some cases, it is not necessarily an effective instructional strategy for every

student or in all language classrooms as some empirical studies indicated. He claimed that, accordingly, continued research is required to substantiate the effectiveness of form-focused feedback in L2 writing classes. Among research on the matter, in particular, Tmscott's (1996) review article ignited arguments among teachers and scholars in this field. After reviewing a large amount of previous research, Tmscott made a rather radical conclusion that grammar correction by L2 teachers is ineffective and even harmful. Therefore, it should be abandoned right away. He presented three reasons to support the argument. First, previous studies that he reviewed didn't offer any valid grounds for grammar correction and, though there are some studies showing the positive effect of grammar correction, they are mostly due to learner's tendency to avoid using grammatical features they are poor at. Additionally, based on morpheme studies of Dulay and Burt (1973, 1974), which implies that L2 learners should reach appropriate level to acquire linguistic howledge including grammatical rules, he claimed grammar correction performed in current writing class ignores this natural learning process. The last reason for his argument was found in L1 studies by Knoblauch and Brannon (1981),which proved the futility of L1 grammar correction. Tmscott noted that since the L2 situation is the same as that of L1, the evidence showing the invalidity of teachers' grammar correction in L1 can apply to L 2 learning as well. After Tmscott's article against form-focused feedback was published, a great deal of discussion and controversy was followed as to what the better approach is to the issues of accuracy and error correction in L2 composition. There were several scholars who exchanged open debates with Truscott. Especially, Ferris (19%, 2001) actively responded to Tmscott in Journal of Second Lnnguage Writing, refuting that Truscott's argument is premature and overly strong and discussed areas requiring more in-depth studies. According to him, Truscott's claim is based on several flaws including the use of a vague definition of 'grammar correction', and Truscott tended to overstate research findings that support the claim against grammar correction and dismiss the studies which contradicts him. However, Ferris partially admitted some aspects that Tmscott had pointed out

and addressed that Tmscott's work in 1996 contributed to L2 writing education, in respect that it had L2 teachers and scholars rethink the matter of the effectiveness of grammar correction in student composition. In 1998, Fems with Harvey and Nuttall presented empirical evidence supporting his earlier argument in favor of grammar correction in a research article. In the research, it was reported that twelve MA students given grammar correction in a tutorial training program for 10 weeks, showed meaningful improvement of grammatical knowledge such as identifying and correcting student errors. There were several other studies proving the benefits of grammar correction except that of Fenis. Fathman and Whalley (1990) found that all the students (n=36) who had their errors corrected gained higher grammar scores than students without getting feedback ( ~ 1 4 in the next writings. Sheppard (1992) produced the similar ) result as well. Two types of feedback (grammatical error correction and general request of clarification) were compared in terms of the effectiveness. When it comes to the gain on the percentage of correct verb forms or on the ratio of subordinations to the total number of sentences, there was no significant difference between the two, but the ,pup with error correction made significantly more growth in percentage of correct sentence boundaty marker than the other group with general request of clarification. In line with the result above, Chandler's (2003) recent research exhibited some positive results of form-focused feedback in writing assignments. In the research, the former helped students produce better writing performance in comparison of correction plus revision with no correction. More specifically described, two experiments in this study showed that teacher error correction and even underlining errors resulted in significant improvement on both accuracy and fluency in following writing of the same type over the semester. However, an interesting fact is, in spite of an overall growth in accuracy and fluency, there was no significant positive change in terms of holistic rating of overall writing quality after 10 weeks. For this, Chandler concluded that it might be due to the fact writing quality is slow to show measurable effects. So far, previous studies dealing with the effectiveness of teachers' grammar correction were briefly reviewed. As mentioned above,

despite a large amount of works on this matter, it is interesting that the definite conclusion of the efficacy of form-focused feedback has not been made yet.

2.2 Content-based feedback

Now let's turn to the second type of feedback, which is known as content or meaning-based feedback. Unlike form-focused feedback, content-based feedback focuses more on content quality and organizational features in students' composition and teachers provide overall comments on where it doesn't make sense in terms of content or give some comments on logical fallacies in writing without pointing out specific grammatical errors. Since this type was put forth from the dubiousness of grammar correction, many researchers conducted their research to reveal the relative superiority comparing to form-focused feedback. Semke (19%) examined four groups' performance receiving different types of feedback. His findings indicated that the group of students treated with only comments on content was better than any other groups (group of comment on errors, group of comment on both content and errors, group of self-comection) on both accuracy test and grammar test. This result is often mentioned as evidence showing the effectiveness of content-based feedback in later studies. Zamel (1985) was also interested in the effects of teacher's written feedback. According to her study, when a teacher gave two types of feedback together, for example grammar feedback and content feedback, learners corrected only local grammatical errors and did not pay attention to overall content features or logical fallacies which can be underlying problems in writing. In result, learner's writing ability didn't show any positive changes later on. Based on her findings, she claimed that pointing out grammatical errors and commenting on general content and organization together can cause learners to be confused which type of response deserves higher priority and obstruct their development of actual writing competence. Therefore, she recommended that when revising student writing teachers should consider meaning-level issues first because they can help to develop student

underlying writing competence. Kepner (1991), who strongly believed in the efficacy of content-based teacher feedback, showed superiority of the method by comparing form-focused feedback with content-based comments in terms of level of grammatical accuracy and level of thinking expressed in contents of student writing. In an experiment conducted by him, two groups of students learning Spanish as L2 received different types of written feedback and were measured their degree of development during one semester. According to his conclusion, grammatical error correction is not likely to help to improve the level of accuracy nor enhance the ability of thinking in L2 writing. One interesting fact in this study is when content-based comments are given to students at periodic discourse-level, it can promote students' grammar accuracy, as well as ideational quality. Despite arguments and empirical results in favor of content-based feedback above, Lee (1997) showed a different result that correcting surface errors yields a better result than meaning-based correction in student writing performance. For this result, she analyzed that it is because it is more difficult to fix correcting meaning and logical errors of writing than surface errors. In other words, correcting meaning and logical errors of writing is a cognitive demanding work particularly for students with low language proficiency. In this respect, the study gives an implication that when deciding an appropriate feedback for learners, other variables such as learner' proficiency should be considered as well. However, most of recent studies have shown content-based feedback is more effective on student writing than form-focused feedback and researchers are likely to agree with it. From learners' point of view, however, content-based feedback is not likely to be as favorable as that of teachers and researchers in the field. As shown in some research (Chandler 2003, Ferris 2004)' when many students received only meaning-related feedback, they tend to feel their teachers don't pay much attention to their writing or even regard that teachers lack sincerity. Therefore, this data means that the efficacy of teacher written feedback is one thing and student's need is another. That is, regardless of results of studies, students have a strong desire for their teachers to supply more direct error feedback on their writing, which can not be dismissed

or ignored by teachers so easily.

2.3 Integrated feedback

To make an attempt to solve the problems of using only one type of feedback, some other teachers and researchers came up with the third form of teacher written feedback by combining grammar correction with content-related feedback, which is usually called integrated feedback. On the part of teachers, in fact, they want to believe that the direct insttuction such as correcting grammatical errors helps their students improve the accuracy of writing. Besides, they are concerned that if student's desire of teacher feedback is ignored, it might interfere with student's motivation and confidence in the writing class. For this reason, hypothesizing that integrating the two types of feedbacks can be more beneficial to learners, researchers wanted to verify the effectiveness of this form of feedback. First, Song (1998) aimed to research on which gives more positive effect on student's writing ability between meaning-related feedback and integrated feedback. This study exhibited that students with integrated feedback gained higher scores in holistic aspect as well as two analytical aspects such as content/organization and mechanics. However, there was no meaningful difference between content-based feedback and integrated feedback in aspects of vocabulary and style. Though students were not superior in writing style and word bowledge after receiving content-based feedback here, it was shown that integrated feedback is more effective and advantageous to improve L2 student's general writing skills in this study. The similar result was seen in an experiment by Ashwell (2000). He made a comparison of four cases; form feedback only, content-based feedback only, feedback combined with two types in a different order and two types of feedback simultaneously. He noted that the result didn't show a significant differentiation between which order they received form or content feedback when two types of feedback are treated separately. Interestingly enough,

the most effective result of writing abilities came from when learners received mixed pattern of two types of feedback at the same time. For this result, he commented that his first finding supplies counterevidence to Zamel's (1985) argument that content feedback should be given on earlier draft and form-focused feedback should be conducted on later draft. Moreover, based on another finding, he refuted Fathman and Whalley (1990) and Ferris (1997)'s argument against giving two types of feedback together, noting that giving form and content-based feedback simultaneously does not harmfully affect student writing skill at all and rather it can improve both aspects of writing, content aspect and grammatical accuracy.

2.4 Other types of feedback

Similar but somewhat different from one of three types of written feedback described above, several different approaches as teacher writing response have been suggested. Lalande (1982), though agreeing with the effect of teacher written feedback to students, claimed that the direct correction like grammar correction is not as good as the indirect feedback like just underling student's error. His empirical study offered the data that showed the influence of two treatments of direct and indirect teacher response to student writing. In one, teachers gave direct error correction and in the other, they gave indirect error feedback using correction code which requires students themselves to correct errors. It is revealed that the second method produced fewer errors in student's following writings. According to his analysis, students usually don't care about the reason why they make errors in case that teachers revise them directly, which leads to fail the deep thinking process. In the same line with this, Byrne (1998) in his study argued that minimal marking of one of indirect feedback is more effective than direct teacher's feedback According to him, the advantage of this technique is to make correction neater and less threatening than masses of red ink and help students to find and identify their mistakes. On a basis of belief in effectiveness of content-based feedback,

Hyland. F., and Hyland, K. (2001) focused more on the detailed text analysis to the written feedback such as praise, criticism and suggestion. They found that praise and migration strategies like hedging devices were most frequently employed to soften criticism and suggestions rather than simply responding to good work However, they added that since such indirectness may cause incomprehension and miscommunication between teachers and students, teachers should give a more careful consideration when using these methods. As another alternative, some other researchers use recording remarks on a tape recorder and writing a number on the student paper to indicate what the comment refers to. This new technique can save time and provides listening practice for learners and assists those with an auditory learning style preference. The last one to be introduced is electronic feedback using computers. In this way, teachers can provide comments on electronic submissions by e-mail or by using the comment function, which allows feedback to be displayed in a separate window while reading a word processed text. Feedback on errors can also be linked to texts to show students examples of features they may have problems using correctly. These new ways of written feedback offer greater flexibility in their responding practices, but ultimately effectiveness on student's side and conveniences on teacher's side is likely to be the deciding factor in which are used.

3. Conclusion
Until this part, this paper briefly examined previous literature on three t y p s of teacher written feedback which argued the superiority of a certain kind of teacher written feedback. Howwer, there is also some other research that made a different conclusion from studies described above. Among them is Fazio (2001). Fazio carried out a classroom-based experimental study to investigate the effects of differential feedback (corrections, commentaries, and integ~ated feedback) on the journal writing accuracy for L2 learners. Total of 112 students participated in this study for four months.

Unexpectedly, the outcome indicated that learners didn't experience a significant change in their accuracy in grammatical spelling as a consequence of types of feedback in their journal writing. On a basis of this, he claimed that students are not affected by with which type of feedback teachers instructed them and rather, other variables like students' attentiveness and pedagogical context are likely to play more important role in improving student writing ability. Although a large amount of researcher's effort to find out the relative efficacy among three types of feedback were made as dexribed above, a conclusion is not made yet as to which type of feedback is the better for the development of student writing competence. Maybe, it is because a variety of other factors are involved on this matter along with the types of feedback, which often leads researchers to interpret the same data or research results differently as shown in Truscott (1996) and Fems (1999). In addition, since all studies have flaws to some extent, scholars with different opinions with other scholars attack to that point and make study results invalid. Furthermore, until recently, conflicting results to the same t p of teacher feedback have been yielded. For these ye reasons, it makes us doubt the effectiveness of a specific type of teacher feedback. As mentioned in the earlier part of this paper, since there are many other factors to be considered in deciding the effectiveness on student writing skills, it is not easy to conclude which is the most helpful to our students among methods discussed. However, reviews concerning this matter as in the present paper can provide much valuable information for L2 teachers and researchers to seek for a better method for their students in a writing class. In addition, writing educators in Korea, where the interest in English writing is growing but the studies in this field are still scant, can get much help to perfonn writing instruction in the field.

References

Ashwell, T. 2000. Patterns of teacher response to student writing in a multipledraft composition classroom: Is content feedback followed by form feedback the best method? Jorirnnl of second lanpage writing 9: 227-257. Burt, M. K. 1975. Error analysis in the adult EFL classroom. TESOL Quarterly 9: 53-63. Byrne, D.1988. Teaching writing skills. Harlow, UK: Pearson Education. Chandler, J. 2003. The efficacy of various kinds of error feedback for improvement . in the accuracy and fluency of l2 student writing. Journal of Second Lnnguage Writing 1 2 267-296. Cohen, A. D., and M. Robbins, 1976. Toward assessing interlanguage performance: The relationship between selected errors, learners' characteristics, and learners' explanations. innpage Learning 26: 45-66. Dulay H. C, and M. K Burt, 19B. Should we teach children synfax? h p n h~m i n g 23: 245-258. Dulay H. C., and M. K. Burt, 1974. Natural sequences in child second language acquisition. Lrrnpage Learning 24 :37-53. Fathman., A. K., and E. Whalley, 1990. Teacher response to student writing: Focus on fonn wrsris amfenf. In B. Kroll (eds.), Second language writing Research insights for the classroom. Cambridge: Cambridge University Press. Fazio, L. 2001. The effect of corrections and commentaries on the journal writing accuracy of minority-and majority-language students. Journal of second lnnpnge writing 10: 235-249. Ferris, D. 1999. The case for grammar correction in L2 writing classes: A response to Truscott 19%. Jm~mnlof second Innpage writing 8: 1-11. Ferris, D. 2004. The "grammar correction" debates in L2 writing: Where are we, and where do we go from here? (and what do we do in the meantime...?). Journal of second lanpage writing 13: 49-62. Fems, D., H. Harvey, and G. Nuttall, 1998. Assessing a joint fraining project: Editing strategies for ESL teachers and shrdenfs. Paper presented at the American Association for Applied Linguistic Conference, Seattle, WA. Ferris, D., and B. Robert, 2001. Error feedback in L2 writing classes: How explicity does it need to be? Journal of second Innpiage writing 10: 161-184. Hyland, F., and K Hyland, 2001. Sugaring the pill: Praise and criticism in written feedback. Journal of second Innpiage n~riting10: 185-212. Hendrickson, J. M. 1978. Error correction in Foreign language teaching. Recent theory, research and practice. Modern Lnngliage Jo~rrnnl 378-398. 62: Kepner, C. G. 1991. An experiment in the relationship of types of written feedback to the development to the development of mcmd-language writing skills. Modem Lnnpage Jo~irnnl 305-313. 75:

fioblauch C, and L. Brannon, 1981. Teacher commentary on student writing the state of the art. F r ~ h m a nEnglish News 10: 1-4. Lalande, J. F, Ill982 Reducing composition errors: An experiment. Morlern h ~ u g e . Journal 66: 140-149. Lee, I. 1997. ESL learners' performance in error correction in writing: Some implications for college-level teaching System 25: 465477. Lee, I. 2004. Error correction in L2 Secondary writing classrooms: The case of Hong . Kong. Journal of Second hnguage Writing 13: 285-312. Semke, H D 1984.The effects o the red pen. Fureip h n ~ m n g e .. f Annals 17, 195-202. Sheppard, K.1992. Two feedback types: Do they make a difference? XELC J o z i d 23: 103-110. Song, M. J. 1998. Relationship between types of written feedback and development of English writing skills. English Teaching 53: 135-156. Truscott, J. 1996. The Case Against Grammar Correction in L2 Writing Classes. h n p n g e LRnrning 46: 327-369. Zamel, V. 1985. Responding to student writing. TESOL Q~inrterly 79-101. 19:

Eunjung Park anna1974tZsnu.ac.kr