000074045 THE EFECTS OF TEACHER'S DIRECT AND INDIRECT CODED FEEDBACK ON STUDENTS' WRITING. A QUASI EXPERIMENT AT LE QUY DON-DONG DA HIGH SCHOOL ẢNH HƯỞNG CỦA PHẢN HỒI MÃ HÓA TRỰC TIẾP VÀ GIÁN TIẾP CỦA GIÁO VIÊN ĐẾN VIỆC VIẾT CỦA HỌC SINH. MỘT THÍ NGHIỆM B

000074045 THE EFECTS OF TEACHER'S DIRECT AND INDIRECT CODED FEEDBACK ON STUDENTS' WRITING. A QUASI EXPERIMENT AT LE QUY DON-DONG DA HIGH SCHOOL ẢNH HƯỞNG CỦA PHẢN HỒI MÃ HÓA TRỰC TIẾP VÀ GIÁN TIẾP CỦA GIÁO VIÊN ĐẾN VIỆC VIẾT CỦA HỌC SINH. MỘT THÍ NGHIỆM BÁN THỰC TẠI TRƯỜNG THPT LÊ QUÝ ĐÔN - ĐÔNG ĐA

MINISTRY OF EDUCATION AND TRAINING

HANOI UNIVERSITY

i e i c k i e i e i f k ' i c k i f k - k ' k i e

TRAN THI HUYEN

THE EFECTS OF TEACHER’S DIRECT AND INDIRECT CODED FEEDBACK ON STUDENTS’ WRITING A QUASI EXPERIMENT AT

LE QUY DON-DONG DA HIGH SCHOOL

A C K N O W L E D G E M E N T S

First, I would like to express my sincere thanks and my deepest gratitude to my supervisor, Dr

Nguyen Van Trao for his constant encouragement, thorough instructions, invaluable advice and detailed comments throughout my research Without his help, this thesis could not have 3een completed

n addition, I am also thankful to my teachers at the Department o f Post- Graduate Studies at lanoi University for their professional teaching and devotion that inspire me with the love o f loing research

am also grateful to my colleagues and students at Le Quy Don- Dong Da High School They elped me to collect data for this study and provided valuable information to improve my icsis My special thanks also go to the Management Board at my school who gave me great upport and create all favorable conditions for me during my research

inally, my deep gratitude goes to my parents, my husband and my two daughters I could not live gone this far without their love, support, and faith in me

A B ST RACT

The role o f corrective feedback in second language acquisition and whether it helps improve accuracy and overall writing have been a topic o f much controversy for many years While some researchers advocate the usefulness o f corrective feedback, Truscott (1996) claimed that all error correction is unnecessary, ineffective, and even counterproductive Up to present, resuks from prior research have shown to be inconclusive because o f inadequate methodology with the main problem that m ost studies did not include a proper control group Furthermore, bette. designed and m ore longitudinal research is necessary before any conclusions can be drawn about the (in)effectiveness in improving students’ writing over time

The current study aimed at investigating the impacts o f direct and indirect coded feedback on students’ writing accuracy and exploring whether they resulted in different outcomes in terms

o f tht improvement in stud en ts’ writing accuracy and their writing marks It also aimed to explo-e the students’ attitudes towards different types o f corrective feedback Three intact classes consisting o f 105 students (74 males and 31 females), were involved in the study and divided into three classes: the control class (receiving general comments) and twoexperimental classes (with one receiving direct feedback and the other receiving indirect coded corrective feedback)

The stidy was conducted in one academ ic term which lasted 18 weeks Data were collected from tiree sources: three tests (pre-test, mid-term test and post-test), the questionnaires and the stuients' reflections

The results o f the study show ed that the two types o f corrective feedback did improve studems’ writing accuracy and their writing marks, thus improve their writing quality The results -evealed that direct feedback is more helpful with lower proficiency students and more

effective with “unbeatable” errors Indirect coded feedback is more helpful with students o f higher proficiency and more effective in long-term compared with direct feedback and general feedback Overall study yielded results that there is no absolute answer to what type

o f feedback is the most beneficial as a specific type o f feedback may be effective to a specific group o f students o f a certain proficient level at a certain period o f time It is suggested that a wide range o f flexible ways o f providing corrective feedback should be developed in writing classes according to students’ needs at a certain proficient level at a certain period o f time

T A B L E O F C O N T E N T S Page

A C K N O W L D G E M E T N S I

A B S T R A C T ii

TABLE OF C O N T E N T S Iv LIST OF F IG U R E S vii

LIST OF A B B R IV IA T IO N S ix

C H A P T E R I: I N T R O D U C T I O N 1

1.1 Rationale 1

1.2 Aim o f the study and research questions 3

1.3 Scope o f the stu d y 4

1.4 The outline o f the s tu d y 4

C H A P T E R II: L I T E R A T U R E R E V I E W 6

2.1 Overview o f writing skills 6

2.1.1 Definitions o f writing sk ills 6

2.1.2 Learner autonom y and teacher’s role in writing skills 7

2 2 Overview o f corrective feedback 9

2.2.1 Definition o f corrective feed b ack 9

2.2.2 Perspective on corrective feedback 9

2.2.3 Focus o f feedback on student writing 11

2.3 Controversies around corrective feedback 13

2.3.1 Teacher’ corrective feedback and accuracy im provem ent 13

2.3.2 Types o f corrective feedback 15

2.3.2.1 Focused and unfocused corrective feedback 15

2.3.2.2 Direct and indirect corrective feedback 16

2.3.2.3 Indirect coded and non-coded corrective feedback 18

2.3.3 Feedback, self-correction and language acquisition 21

2.3.4 Feedback on different types o f errors 25

2.3.5 Students’ perceptions and preferences for teachers' corrective 28

feedback 30

2.3.6 Previous stu d y

CHA’T E R III: M E T H O D O L O G Y 34

3 1 Research Q uestions 34

3.2 Descriptions o f variables 35

3.2 Independent variables 35

3.2.! Dependent variables 36

3.3 Subjects o f the stu d y 36

3.3 The students 36

3.3.! The r e s e a r c h e r 38

3.4 Data collection instrum ents 38

3.4.1 The tests 39

31.1.1 The purposes o f the tests 39

3 4.1.2 The details o f the tests 40

3.42 Questionnaire 40

3.43 The students’ reflections 41

3.5 Procedure o f the study 42

3.6 Data analysis 47

3.7 Sum m ary 51

CHAPTER IV: R E S U L T S 52

4 1 Results 52

4.11 Results o f the pre-test 52

4.12 Results o f the pre-test, mid-test and post-test com pared 54

4.2 The effects o f direct and indirect coded feedback on five error categories 63

4.3 Students’ views o f their error problems and their preferences for teachers’ 65 corrective f e e d b a c k 65

4.; 1 Students’ views o f their own errors 65

4 3 1 Students’ preferences for teachers’ corrective feedback 66

4.3.5 Students’ reflections 70

CHAPTER V: R E C O M M E N D A T IO N S AND C O N C L U S IO N S 73

5.1 ■Nummary o f major findings 73

5.2 lecom m endations 77

5.3 .imitations and suggestions for further study 80

5.4 Conclusions 81

R E I E R E N C E S 83

A P I P E N D I C E S

Appeidix 1 : Questionnaire (English version)-for indirect coded feedback class 90

Appeidix 2: Questionnaire (Vietnamese version) 94

Appeidix 3: Questionnaire (English version)-for direct feedback class 97

Appeidix 4: Questionnaire (Vietnamese version) 100

Appeidix 5: Numbers o f errors/words in the control class 102

Appeidix 6: Numbers o f errors/words in the experimental class (direct feedback class) 103

■ Appeidix 7: Numbers o f errors/words in the experimental class (Indirect coded feedback class) 104

Appeidix 8: Results o f second semester ex am s 105

Appeidix 9: Results o f the control class 106

Appeidix 10: Results o f the experimental class (direct feedback class) 107

Apepmdix 11: Results o f the experimental class (indirect coded feedback c l a s s ) 108

Appeidix 12: Scoring rubrics 109

Appeidix 13: Writing pre-test, mid-test and post test 110

Appeidix 14: Checklist for providing feedback I l l Appeidix 15: Table o f F-statistics P= 0.05 114

L IS T O F F IG U R E S AND T A B L E S Page

Figure 1 : Developing a Process Composition in classroom 8

Figure 2: The process o f learning implicit knowledge 24

Figure 3: Means o f errors in the pre-test (1), mid-test (2) and post-test (3) 59

Figure 4: Means o f scores in the pre-test (1), mid-test (2), post-test (3) o f three classes 62

Table 1 : Demographic details o f the subjects 37

Table 2: Students’ results o f second semester exam s 38

Table 3: Description o f error categories used for feedback (Ferris & Robert, 2001) 43

Table 4: Description o f error and error codes used for feedback 46

Table 5: Description o f the interpretation o f K appa 48

Table 6: The description o f the interpretation o f F-test statistics, p-value and significance in one-way A n o v a 49

Table 7: One-way Anova descriptive statistics o f the pre-test’s means o f scores o f the control class, the direct feedback class and indirect-coded feedback class 52

Table 8: One-way Anova F-test o f the pre-test’s means o f scores o f the control class, direct class and indirect-coded feedback class 53

Table 9: One-way A NOVA descriptive statistics o f the pre-test's means o f errors per ten words in three classes 53

Table 10: One-way ANOVA F-test o f the pre-test’s means o f errors per ten words in three classes 54

Table 11 : Error means o f three classes in the pre-test, mid-test and post-test 55

Table 12: Means of scores o f three classes in the pre-test, mid-test and post-test 55

Table 13: Anova repeated measure statistics o f error means o f three classes in the pre- 56 test, mid-test and post-test 56

Table 14: Anova repeated measure statistics o f score means o f three classes in the pre- 56 test, mid-test and post-test 56

Table 15: One-way Anova error means statistics o f the mid-test and post-test 57 Table 16: Poc-hoc pair wise error means comparisons o f the mid test and post test o f

three classes 58

Table 17: one-way Anova score means statistics o f the mid-test and post-test o f three classes 60

Table 18: Poc hoc pair wise score means comparisons o f the mid-test and post-test o f three classes 61

Table 19: Frequency o f errors in five categories in three classes in the pre-test 63

Table 20: Frequency o f errors in the pre-test, mid-test and pos- test in three classes 66

Table 21: Results o f the students’ views o f their own e r r o r s 67

Table22: Results o f students’ preferences for teachers’ corrective feedback 68

LIST OF A B BRIVIAT IO NS

CC: Control Class

CF: Corrective feedback

CG: Coded Group

EC: Experimental Class

E F L : English as a Foreign Language

NCG: Non Coded Group

SLA: Second Language Acquisition

SL: Second Language

TESOL: Teachers o f English to Speakers o f Other Languages WCF: Written Corrective Feedback

Thte main role o f w riting teachers is to help their students improve their writing projficiency according to the students’ needs and course objectives How to best achieve thus has been concern o f m any E S L /EFL writing teachers and researchers Providing feed b ack is viewed-both by teachers and students- as an important part o f writing instruction One type o f feedback that writing teacher provide is corrective feedback It is peirhaps the most widely used method o f responding to students’ writing Yet, there are cointradicted attitudes tow ards corrective feedback Truscott (1996) triggered the debate by' claiming that corrective feedback on second language learners’ written output is not omly unnecessary and ineffective, but also counterproductive He based this claimed on foiur types o f arguments First, he indicated several theoretical problems o f error correction He argued that language teachers, when providing corrective feedback, adlopted a simplistic view o f learning as essentially the transfer o f information from tetacher to student w ithout realizing that interlanguage development is a complex and grradual process Regarding the problem s involving the order o f language acquisition he cllaimed “when students are corrected on a point for which they are not yet ready, the

correction is not likely to have much value” (p.l 1) Furthermore, on the basis o f practical considerations, he doubted teachers’ ability to provide feedback comprehensibly and consistently.

In response to Truscott’s arguments, Ferris (1999; 2002; 2004) showed her strong support for the use o f corrective feedback In her opinion, corrective feedback give learners information they need to notice errors She suggested that students need distinct and additional intervention from their writing teachers to make up their deficits and develop strategies for finding, correcting and avoiding errors According to her, Truscott’s conclusions are premature She reasoned that the results from the previous research have shown to be inconclusive due to inadequate methodology and the fact that most research does not include control group She suggested further research be needed before any conclusions can be drawn about the (in)effectiveness o f error correction in improving students’ future writing

While debate on the effectiveness o f error correction may never be fully resolved, providing written corrective feedback is indispensable because it plays an important role

in guiding, motivating, and encouraging students to improve their accuracy in second/foreign language writing From this point o f view, it is clear that investigating the effects o f different types o f teachers’ corrective feedback on students’ writing and students’ attitudes and preferences regarding corrective feedback are important

In spite o f a wide range o f research which has been conducted to investigate writing feedback practice in different EFL/ESL contexts, to the best knowledge o f the researcher, very few studies have attempted to explore the effects o f teachers’ direct feedback and indirect coded feedback on students’ writing, especially these effects o f five different error categories in Vietnamese contexts Therefore, the researcher decided to choose the topic “The effects o f teacher’s direct feedback and indirect coded feedback on students’

writiing A quasi experiment at Le Quy Don-Dong Da” to investigate what types o f corrective feedback is the most effective in helping students improve their writing accuiracy and then improve their overall writing quality It is hoped that the study will be

o f s o m e use to English teachers in general and English teachers at Le Quy Don Dong Da

H ig h School in particular

1.2 Aims o f the study and research questions

Thiss study is aimed at investigating the effects o f different types o f teachers’ written convective feedback including direct feedback, indirect coded feedback on students’ writting at Le Quy Don- Dong Da High School

T herefore, it aims to achieve following objectives:

1 To examine whether direct feedback and indirect coded feedback can help improve students’ writing accuracy and thus their writing marks

2 To compare the outcomes o f the direct and indirect coded feedback in improving students’ writing accuracy over time

3 To explore the effects o f direct feedback and indirect coded feedback on five error categories in students’ writing

4 To examine students’ attitudes towards and their preferences for teachers’ direct feedback and indirect coded feedback

In order to achieve the above-mentioned objectives, the following research questions we;re asked in the study:

1 Do direct fee d b a ck an d indirect-coded feed b a ck affect stu d en ts' accuracy in

writing and thus improve their writing marks?

2 I f so, which o f the two techniques o f giving feed b a ck is more effective in improving

students ' accuracy in writing and thus their writing marks over time?

o f errors in students ’ writings over time?

preferences towards te a c h e r’s corrective feedback?

1.3 Scope o f the study

Thiss study is limited to explore the effects o f direct and indirect-coded feedback feedback

on 11th form students’ accuracy in writing at Le Quy Don- Dong Da High School It exaimines three intact classes o f students who are voluntarily participated in this research

It is> hoped that the findings would help teachers to develop various theory-laden ways o f giviing feedback to help students reduce the number o f errors in their writing and thus impirove their grammatical accuracy in writing It is also hoped that the study will provide the insight into how students view their problems o f writing errors and how they response

to tteachers’ feedback in order to minimize the discrepancy between teachers’ feedback- giv'ing practices and students’ preferences

1.41 The outline o f the study

Thiis study consists o f five main chapters:

Clhapter I: Introduction

Thiis chapter clarifies the rationale for the study, the aims o f the study, the research quiestions, the scope o f the study as well as draws out its outline

Chapter II: Literature Review

The chapter reviews necessary literature related to direct and indirect coded feedback and the relationship between feedback and accuracy improvement in writing, corrective feedback, self correction

Chapter III: M ethodology

The chapter describes the subjects o f the study and site, the data collection instruments and procedure to collect data

Chapter IV: Results

The chapter presents the results o f the study about the effects o f direct and indirect coded corrective feedback on improving students’ accuracy in writing

Chapter V: Recom mendations and conclusions

The chapter reaches the conclusion o f the study, the limitations o f the study and some recommendations for further studies also make up in this chapter

C H A PTER II: LITERATURE REVIEW

This chapter reviews the literature related to the study’s issue It is expected to provide readers with background knowledge for better understanding o f the rest o f the paper

2.1 Overview o f writing skills

2.1.1 Definitions o f writing skills

Writing seems to have taken on different definitions for different groups o f people in order to suit their different needs and purposes for writing Even for professionals involved in the field o f English Language Teaching, not one definition o f writing could satisfy everyone

According to Klein (1985 as cited in Tan, 2009), writing is the ability to put pen and paper to express ideas through symbols This way, representations on the paper will have

m eaning and content that could be communicated to other people by the writer From another view o f writing, Zamel (1987) defined writing as “a cognitive behavior and a nonlinear process o f discovery” (p.699)

Grabe and K aplan (1996 as cited in Tan) explored the meaning o f writing in terms o f the rhetorical triangle in writing, and such triangle consists o f the reader, the recipient o f the final product o f writing process; the writer, the originator o f the message; and the subject matter and the text itself Both the writer and the reader have to consider all these aspects

w hen writing and reading respectively, for each one plays a significant role in the journey towards meaning

Moreover, process theorists (Flower & Hayes, 1981) viewed writing as “the result o f long, laborious, and intensely personal process in which writer addresses several

questions ranging from What do I write about? to Who is m y audience? to How do I

structure m y essa yl and to What short o f language and voice should I u s e T \ In language

teachers’ views, writing is “a language skill which is difficult to acquire” (Tribble, 1996) and “normally requires some form o f instruction” and that “ it is not a skill that is readily picked up by exposure” , (p.l 1)

2.1.2 Learner autonom y and teacher’s roles in writing skills

In the traditional writing approach the teachers are more concerned with the written product o f writing, whereby in class students are assigned a task, and then the teacher collects, marks and grades the paper According to Emig (1971) the traditional approach

to writing has not considered the composing process involved in writing He further adds that teachers o f composition tend to “underconceptualize” and “oversimplify” the process

o f composition Gage (1986) asserted that writing is not simply a “skill” to be

m astered but the ongoing reflection o f students developing understanding o f ideas

To develop learner autonomy in writing, teachers can use the process approach in writing where the focus is not on the end product but on the process o f writing Flower (1981) asserted that “writing process” includes activities, which take place inside the writers’ head Murray (1980) referred to this recent process in writing as an act o f discovery Gage (1986) argued that writing class must be a place where students encounter ideas and where they are free to respond to them honestly and critically Flower (1981) suggested that in a writing class the teacher should assist students to concentrate on learning to write; the final product should be the last step in the complicated process o f writing The process approach to writing provides students with important “how to” scaffolding that allows students to discover writing strategies and practices that can lead to successful

products Students are given sufficient time to write and rewrite, to discover what they want to say, and to consider intervening feedback from teachers and peer as they attempt

to bring closer and closer intention in successive drafts Revision becomes central and the teacher intervenes throughout the composing process, rather on reacting only to the final product (Siti Hamin, 1998)

The teacher in a process-approach classroom becomes the facilitator In such classroom, writing is essentially learnt, not taught Providing input or stimulus for learners is perceived as unimportant, since the teacher’s task is only to facilitate the exercise o f writing skills and draw out the learners’ potential The teacher acts as a facilitator because he or she has multiple roles to play such as being a reader and/or an advisor The teacher provides feedback throughout their learning process from the initial stage o f writing to the final product o f the composition Process writing is a recursive process in which the writer begins to write, then makes changes and re-write until the final composition is produced The process is illustrated as in Figure 1

Figure 1: Developing a process composition in classroom

(Norlida &Puteri Rohani (2002), p.303)

Little (1991) claimed that all learning (and learning writing is no exception) is ultimately autonomous learning in the sense that it depends on the efforts o f the learners themselves

In fact, the students who take control o f their own learning are better learners Therefore, teachers should promote and develop autonomous learning in the classroom by facilitating, helping, counseling, guiding and fostering interaction

2.2 Overview o f teachers’ corrective feedback

2.2.1 Definition of corrective feedback.

L'ghtbrow n and Spada (1999) defined corrective feedback as any indication to the learners that their use o f the target language is incorrect This includes various responses that learners receive W hen a language learner says, “He go to school every day”, corrective can be explicit, for example, “no, you should say goes” or implicit “yes he goes to school every day”, and may or may not include metalinguistic information, for example, “D o n ’t forget to make the verb agree with the subject.” (pp 171-172)

A ccoid'ng to Swain(1995), during the process o f second language acquisition, learners will develop several learning strategies, which are noticing the gap between their o ¡ginal output(interlanguage) and native-like language, hypothesis testing and internalizing metalinguistic information One way to promote these strategies is through providing corrective feedback which refers to the responses to a learner’s nontargetlike L2 production

2.2.2 Perspectives on corrective feedback

The main goal in L2 writing before the 1970s was error correction and grammar instruction as mirrored by theories o f structural and behavioral psychology At that time the focus o f language teaching was on how to best teach grammatical forms In teaching

and learning writing skills, the common type o f writing practice was single- draft production and teachers prioritized giving feedback on grammar usage and they usually provided correct forms directly Towards the end o f the 1970s, L2 researchers became highly concerned with ways in which L2 can be acquired naturally without conscious learning process During 1980s, Communicative Language Teaching(CLT) emerged as

an influential teaching theory, which devoted to communicative meaning o f language production With the advent o f communicative approaches, language pedagogy moved beyond a mere focus on the knowledge o f rules and structures towards emphasizing communicative ability in real-life encounter Not only did teachers start to think o f grammar instruction as old-fashioned and uninteresting (Nassaji & Fotos 2011 as cited in Mohammad), but also a number o f researchers argued that grammar instruction had no effect on second language development and thus had to abandoned (Krashen, 1981) Therefore, in language classroom, language teachers prefer or are advised to give feedback on the content or meaning o f the whole massage rather than on grammatical features Consequently, people involve in communication are rather sympathetic towards poorly grammatical massage provided that they understand the meaning that their interlocutors try to get through (Ha, 2010)

Since the beginning o f 1990s, this view o f grammar as an unnecessary component in language pedagogy has been questioned Studies in the context o f French immersion classrooms in Canada, for instance, have found that learners failed to acquire a native­like grammatical competence despite the continuous exposure and practice opportunity the immersion context provided for Research have revitalized form-focused approach and overt attention to linguistic form Focus on form acquired special emphasis with Schmidt’s Noticing Hypothesis which claims that learners must consciously notice forms from input for acquisition/intake to occur With regards to noticing, Swain (1995) argued that output promotes noticing When students notice the “gap or hole” they become aware

o f their linguistic problems

Noticing can also occur when learners allocate attentional resources to specific features in the input (Schmidt, 1990), or when forms are made salient through interactional feedback (Leeman, 2003 as cited in Adam, 2003) In both o f these situations, learners are able to compare their own interlanguage forms with the targerlike forms supplied in the input and determine w'here the discrepancies lie (Adam, 2003) Swain (1995) argued that it is important to draw on second language learners’ productive skills since producing output not only promotes noticing o f linguistic features but combined with feedback also pushes learners’ awareness towards the gaps and problems in their interlanguage.

2.2.3 Focus o f feedback on L2 writing

The role o f corrective feedback in L2 writing has been a constant source o f interest and debate among teachers and researchers Research studies examining this issue have made

a distinction between fe e d b a c k on content/m eaning, which often consists o f comments to

help learners develop and organize their ideas, and feed b a ck on form , which addresses

grammatical errors and punctuation

In SLA, whether meaning -fo cused instruction is sufficient for acquiring a second language has been questioned by researchers Krashen (1981) advocated the notion that the provision o f comprehensible input supported by a low affective filter is sufficient for acquiring a second language, and he further claimed that meaning-based instruction can provide such conditions Prabbu (1987), on the other hand, argued that drawing the learners’ attention to grammatical forms in meaning-based instruction is not necessary and totally unhelpful Furthermore, results from other studies revealed that learners who learn a second language through communicative classes do not gain high level o f language proficiency (Higgs and Clifford, 1982 as cited in Sasan, 2010) or although the majority o f these students achieve native-like comprehension skills, their productive

skills remain far from native-like norms This suggests that meaning- focused instruction results in fossilization (Lotherington, 2004 as cited in Sasan).

In an experimental study, Fathman and Whalley(1990) applied four kinds o f treatment on four groups, namely focus on form feedback, focus on content feedback, a combination

o f both and no feedback, indicated that only the group which was treated with both kinds

o f feedback showed gains in accuracy a a statistically significant level This would suggest that students can improve their writing institutions where content and form feedback are given simultaneously In Fazio’s (2001) study, three kinds o f treatment were adopted: focus on form, focus on content and combination o f both A surprising finding was found: the participants made more mistakes in grammatical spelling in subsequent writing tasks regardless what kinds o f feedback they received He accounted this result for students’ lack o f attention to their own mistake and teacher’s feedback

Up to now, there have not been any exclusive studies which thoroughly compare the impact o f focus on form feedback and focus on content feedback on enhancing students’ grammar accuracy However, when taking feedback on meaning into deeper consideration, it could be realized that this kind o f feedback result in several drawbacks First, the fact that teacher accept every interlanguage form by the learner without correction would result in fossilized errors As Lightbrown and Spada (1990) pointed out that this type o f error may not cause breakdown in communication, but the problem is that learners are often unaware o f the existence o f errors Second, teachers may misinterpret students’ ideas, hence providing accurate but inappropriate expressions or

“put words into students’ mouth” (Ellis, 2010, as cited in Ha, p i 8) Meanwhile, form- focused feedback, which focuses on linguistic aspects o f language keeps informing students about the types o f grammar mistakes they often make, thus helping them to avoid them gradually Accordingly, form-focused feedback seems to be more

advantageous in helping improve students’ writing accuracy, compared with the content- focused feedback.

2.3 Controversies around corrective feedback

2.3.1 T eacher’s corrective feedback and accuracy improvement.

As mentioned in section 2.2.3, if students’ errors are left uncorrected, they will become fossilized and surely have a negative impact on the writing quality Grammar correction was considered the best way to help students avoid repeating errors and improving the accuracy o f their writing, which in turns leads to smoother meaning conveyance (Ashwell, 2000 as d t e d in Ha, 2010) Though a large number o f studies ha\ e investigated the efficacy o f teacher’s feedback on student’s accuracy gains, no conclusive findings have ever been indicated The question o f whether error correction works continues to be

a controversial issue, the strongest case being made by the debate between Truscott and Ferris

Truscott triggered the debate by claiming that feedback posed a threat on language acquisition and development and should be stopped using (1996) Truscott (1996, 1999,2004) made a strong case against grammar correction in L2 writing classes In his study

in 1996, he presented multiple perspectives based on research by others in support o f his argument First, he mentioned that learning a second language would be similar to acquiring the mother tongue which did not require error feedback, thus there should be no reason to provide L2 feedback Second, he looked at several L2 studies that ranged from

1978 to 1993 by different researchers and concluded that correction did not have significant effects on students’ errors and on some cases even causes harm to students Nevertheless, he argued that a variety o f factors could have influenced the negative results on error correction o f previous studies Among these are the difference between foreign language (FL) and second language (SL) settings; the form o f correction used; the

types o f assessments employed; the differences in the type o f instruction practised; learners’ proficiency level or ability; learners’ variables such as age, gender, and educational background and finally, the time spent correcting errors.

However, Ferris (1999), Ferris and Hedgcock (1998), and Ellis (2001) criticized Truscott for being premature in making such a conclusion First, Ferris (1999) claimed that Truscott did not have a firm definition for the error corrective feedback Second, Ferris questioned Truscott’s generalization based on reviews o f previous studies that exhibited crucial variation with regard to groups o f subjects, research and instructional paradigms and type o f feedback These variations are an impediment in making generalizations Ferris (2002) provided some reasons for continuing with corrective feedback: (1) it encouraged short-term development o f students’ accuracy, (2) it made students feel optimistic about the feedback effects, and (3) it trained students to become self- editors

Bitchener and Knoch (2010) supported Ferris’ arguments by claiming that learners who notice the difference between target-like input (be it oral or WCF [written corrective feedback] and their non-target- like output are able to modify it as target like input Additionally, more recent studies reported that learners are able to apply feedback in writings o f new texts, particularly if the feedback is on specific targeted forms (Bitchener & Knoch, 2008; Sheen, 2007)

Despite his call for abandonment o f error correction, Truscott (1999), in his response to Ferris, acknowledged that many interesting questions remain open and that would be premature to claim that research has proven error correction can never be beneficial under any circumstances However, he suggested that researchers and teachers should acknowledge that grammar correction is, in general, a bad idea until future research demonstrates that there are specific cases in which it might not be a totally misguided practice Agreeing with the future research focus proposed by Ferris (1999), he suggested

that attention be paid to investigating which methods, techniques, or approaches to error correction lead to short-term or long-term improvement and whether students make better progress in monitoring for certain types o f errors than others This prompted researchers

to address this need in written CF studies The research studies ranged extensively in the types o f errors addressed, the feedback types and the number o f errors corrected The issue is no longer w hether CF should be done, but rather how it should be done The controversies around CF is also extended to the variety o f strategies o f written CF that a teacher can use

2.3.2 Types of corrective feedback

2.3.2.I Focused and unfocused CF

The focused-unfocused dichotomy refers to the comprehensiveness o f correction methodologies The unfocused (or comprehensive) approach involves correction o f all errors in learner’s text Focused (or selective), on the other hand, targets a number o f linguistic features only (e.g errors in the use o f English articles) Errors outside the focus domain are left uncorrected

Different predictions have been made w th respect to the relative effectiveness o f focused and unfocused CF Ellis et al (2008) claimed that there are theoretical reasons for expecting the focused approach to be more effective in accuracy development than unfocused CF They stated that learners are more likely to notice and understand corrections when they target a specific (set of) error types Sheen (2007) and Bitchener (2008) also argued that unfocused CF may not be the most effective correction method because L2 learners have a limited processing capacity They claimed that asking learners

to deal with CF which targets a broad range o f linguistic features at the same time might produce a cognitive overload, and prohibit feedback processing

There are also reasons to question the hypothesized superiority o f a focused CF approach Ferris (2010) noted that from a practical perspective, only targeting specific error types might not be enough, a teacher’s purpose in correcting his pupil’s written work is (among other things) improving accuracy in general, not just the use o f one grammatical feature Moreover, observing that some o f their errors have been corrected while others have not might be rather confusing for students.

2.3.2.2 Direct and indirect corrective feedback

The much discussed contrast is also made between direct and indirect error correction The main factor distinguishing these two types o f CF is the learners’ involvement in the correction process Whereas direct CF consists o f an indication o f the error and the corresponding correct linguistic form, indirect corrective feedback only indicates that error has been made Instead o f teacher providing the target form, it is left to the learner

to correct his own errors Indirect correction method can take different forms that vary in their explicitness (e.g underlining o f errors, coding o f errors) (Ferris & Robert, 2001)

Various hypotheses considering the relative effectiveness o f direct and indirect CF have been put forward, some in favour o f direct error correction, others supporting the indirect approach

Chandler (2003) asserted that direct correction minimizes the confusion that can happen

if learners do n’t understand the feedback and this is best for producing accurate revisions and for subsequent writing He argued that, whereas direct corrective feedback enables learners to instantly internalize the correct form as provided by their teacher, learners whose errors are corrected indirectly do not know if their own hypothesized corrections are indeed accurate In her study in 2002, Ferris emphasized that direct correction also gives learners right forms for errors, learners especially those with low proficiency find

direct correction less threatening and thus, helpful before they have acquired the adequate ability to correct their own errors Additionally, Bitchener and Knoch (2010) suggested that only direct CF offers learners the kind o f explicit information that is needed for testing hypotheses about the target language.

I lowever, direct CF does not receive much support from researchers due to a number o f drawbacks First, as Wei-chen (2005) claimed “giving students direct correction and believe hat it will w ork based on the belief that views human mind as an empty bottle into wh ch teacher can pour whatever knowledge they want, however, according to the current cognitive psychology, is well known that this is not the way human minds w ork” (p 133), the learners will therefore overlook their own role in correction process and become heavily dependent on the teacher Although teachers can let learners revise their writing, learners can just mechanically copy the ready-made correction without figuring out the reasons Second, teachers may become frustrated, though making exhausting efforts, when he sees that identical types o f errors appear repeatedly on the written paper

o f the same students (Phuong, 2011) In addition, from students’ perspectives, it is discouraging to find their compositions full o f errors crossed out with red -ink Some o f them a n embarrassed and lose confidence when they receive their written work corrected

in this way

Advocites o f indirect CF, on the contrary, have reported that indirect CF assisted stu den t to make accuracy progress in the long run while the direct feedback could only show progress in a short time period (Lalande, 1982; Ferris & Helt, 2000) or equally ( Ferris & Robert, 2001; Chandler 2003) What is more, indirect feedback triggered

“guided learning and problem-solving process” (Lalande, 1982) which enhances students’ awareness o f their own problems and helps nurture independent students writers who rmy be able to self- edit their writing Lee (1997) who studied 149 freshmen o f low language proficiency level in Hong Kong, found out that students’ failure in correcting

errors was attributed to their failure in detecting them Makino (1993) also suggested that the less detailed cues or indirect feedback should be given to more advanced students and the more detailed cues or direct feedback to less advanced students.

In a study o f the efficacy o f different types o f error feedback on accuracy o f revision and subsequent writing, Ferris (2002) found that students who received indirect feedback did much better than the students who received direct feedback She also found out that in the short- term, direct feedback had more effective results, whereas, in the long-term, indirect feedhack had more positive effects Moreover, if teacher decides to use indirect rather than direct feedback, they need to make decisions on the explicitness o f it; the teacher may indicate errors through underlining, highlighting, verbal rules, or a code Teacher may decide to combine both forms o f feedback, direct or indirect, depending on whether she/he expects students to focus on some pattern o f error, i.e “treatable” or “unbeatable errors” Therefore, it has been suggested that before choosing what type o f feedback to use, teacher should take such intervening factors into consideration as students’ level o f proficiency, strength and weaknesses in writing grammar because students at different level o f proficiency can benefit from different ways o f providing corrective on form (Gucnette, 2007 as cited in Ha, 2010) Accordingly, teachers should carefully pay attention to the development o f students’ proficiency levels to provide appropriate types

o f feedback

2.3.2.3 Indirect coded and non-coded corrective feedback.

Studies examining the effect o f indirect feedback have tended to make a further distinction between those that do or do not use a code Coded feedback points to exact location o f an error, and the type o f error involved is indicated with a code, e.g “ WF” (signals the error o f word form), or VT (signals the error o f using wrong verb tense) Non-coded feedback, on the other hand, refers to instances when the teacher underlines

an error, circle an error, leaving students to figure out the types o f errors themselves (Bitchener et al, 2005).

There are some researchers who evaluated the effectiveness o f coded and non-coded feedback (Ferris & Roberts, 2001; Chandler, 2003; Delgado, 2007; Ha, 2010) Chandler (2003) noted that there is a difference between coded and non-coded group Delgado (2007) and Ha (2010) also claimed that coded feedback is the most helpful compared with non-coded and no feedback while other authors found that there were no significant differences between the group treated with coded feedback and the group given non­coded feedback in students’ development in grammar accuracy From the teachers’ point

o f view, it would be assumed that the indirect non-coded feedback is likely to be more favorable as it is time saving and helps teachers to avoid mistakes in identifying types o f errors when using codes (Ferris & Robert, 2001) However, Ferris (2002) cautioned that dearth of input from feedback with underlining may pose threat on leaner’ linguistics development in the long run compared with coded feedback According to Harmer (2001,

as cited in Jimena et al., 2011) coded feedback makes correction much neater due to the simple and systematical codes Besides this, this method involves learners in the self­correction process and helps them learn more effectively Meanwhile, it arouses learners’ responsibility in correction and improves their writing accuracy in the long run (Ferris,

2002).

Wood (1993) claimed that using the correction code to guide students in correcting their own errors makes writing a more o f learning activity Using the correction code does not make marking paper more efficient, but it does provide students with an opportunity to learn from their mistakes and to gain confidence in their ability to write She also suggested that at the beginning o f the term, teacher should give each student a copy o f the correction code that will be used in the course or put the correction code on the blackboard for students to copy on the first day o f class Students are asked to keep the

correction code handy so that they can refer to it when rewriting the compositions It will take time for the students to get used to the symbols in the correction code and to get used

to correcting and rewriting their compositions on their own

However, since indirect coded feedback just covers the common errors and limited, those individual errors may be ignored In addition, the number o f the correction codes that teacher use to give feedback to students’ writing appears to be a dilemma (Gower, Phillips & Walters, 1995 as cited in Phuong, 2011) If teachers utilize too few codes, they might not cover most types o f errors I f the number o f codes, on the other hand, is overloaded, the situation that students may confuse and may not remember the codes is quite possible (Phuong, 2011) Besides, coded feedback is threatening and hard to be self­corrected for low-proficiency learners (Ferris, 2002) Furthermore, when delivering coded feed, one point teachers and learners must bear in mind is that they must understand what the codes mean, be consistent with and accustomed to the codes Otherwise, chaos may occur due to the misinterpretation o f the codes (Jimena et al,

2011).

Delgado (2007) conducted a study to investigate the effects o f different error feedback (coded, non-coded and no feedback) on the ability to self -e d it L2 writing o f 30 Spanish students in second semester university level and found that students benefit from having coded feedback over both non-coded feedback and no feedback She suggested that teacher should not address every single error in the students’ paper as this practice may overwhelm students Thus, students should only correct errors based on acquired knowledge o f the target language In other words, the teacher should only code those errors where the students can be expected the grammatical rules behind their errors An important issue that needs to be addressed is to familiarize students with codes and symbols used to mark their errors

Lee (1997) stated that the use o f codes should be based on students’ understanding o f what concepts represented by the codes and if students were not aware o f the codes, then the effectiveness o f codes should be questioned He also warned that teachers should be careful in using error codes, which can be seen by teachers’ consistency in using codes when giving feedback on students’ writings It can be widely seen that teachers o f EFL :lassrooms often introduce the writing codes but hardly follow them from the beginning

to the end o f the writing course Conceivably, it is really time consuming for teachers to detect errors and identify error types at the same time for a huge number o f writing or :odes sometimes are too short to convey teachers’ comments on students’ writings As such, the effects o f coded feedback cannot be fully demonstrated

2.3.3 Feedback, self- correction and language acquisition

Ferris (1999) claimed that though it is arguable whether grammar feedback and instruction will be consistently effective for all L2 students writers, it seems clear that the absence o f any feedback or strategies training will ensure that many students never take seriously the need to improve their editing skills and that they will not have the knowledge or strategies to edit even when they do perceive its importance (p.8) Ferris and Hedgcock (1998) also agreed that most students benefit from grammar instruction and feedback which eventually leads to self- correction techniques

A number o f teachers believe that students are not able to correct themselves and that teachers must give them a right model to copy Many L2 teachers feel that they are morally responsible for correcting all the errors in students’ written work despite the fact that it is time-consuming and may be a burdensome task to some extent However, if the teachers continually do this for the students, they will become passive and lazy in self- :orrection (Phuong, 2011) She also argued that the learners must eventually correct themselves if they are really to learn the language and the art o f helping students for the

teachers here, therefore, is to get them used to self-correcting and to give them support only when necessary.

According to M akino (1993), self-correction has two advantages: one is that “self­correction gives students an opportunity to consider and activate their linguistic competence, so that they can be active participants in written compositions rather than passive recipient o f feedback” (p.338); the other is that “self- correction can activate students’ linguistic competence in correcting their own errors and also improve their linguistic creativity through self-correction” (p.338) In other words, teacher feedback and students self-correction may well develop students’ abilities to write more explicitly and accurately in a foreign language

Wood (1993) suggested that in EFL/ESL writing classes teachers should have two aims

in mind: developing fluency and developing accuracy In order to help develop students’ fluency, teachers should provide plenty o f opportunities for them to express ideas on paper To help develop accuracy, teachers should correct student errors, offer constructive criticism and suggestions, and provide remedial exercises when necessary She also states that quality and quantity are another two important factors that writing teachers expect from their students’ written work and self -correction o f student work addresses these issues: fluency and accuracy, as well as quantity and quality According

to the author, self-correction not only provides students with opportunities to develop both rluency and accuracy but also helps “w ean” students from dependency on the teacher for correction The teacher is available to answer questions and to help students whenever necessary, but he/she allows them to first figure out on their own reason for their errors and how they can correct them Putting more o f the responsibility on students for correction develops a sense o f self-sufficiency It boosts their confidence, particularly

w hen they com pare their rewrites to the originals and can clearly see the improvement

they have made Students become more active participants in their own learning and in helping classmates to learn.

Regarcing the relationship between feedback and self-correction, Lalande (1982) sugges:ed that feedback would be the most beneficial to learners when they took actions

to correct errors by themselves after receiving teachers’ feedback, which implies that self-correction is part o f writing process In the future prospect, besides knowing how to correct errors, students should be trained to self-edit their writings after finishing them In other v ords, as suggested by Ferris (1995), learners should become “ independent editors” (p 18), which means they are able to proofread and edit their own writings even when teacheis are not available to locate, label or correct errors for them This should become the goil o f EFL writ"ig classes because it would benefit both learners and teachers Learne'S can be more confident in realizing and correcting their errors, which helps relieve the teachers’ painstaking task o f giving feedback

In terns o f language acquisition, self correction also increases students’ awareness o f the language, which is indentified by Schm It (1990) as one o f three aspects o f consciousness involved in language learning These aspects include awareness, intention and knowledge The first sense, consciousness as awareness, embraces noticing Accorcing to Schmidt (1995, p.20), “the noticing hypothesis states that what learners notice n input is what becomes intake for learning” Schmidt also stated that a) whether a learner deliberately attends to a linguistic form in the input or it is noticed purely uninteitionally, if it is noticed it becomes intake; and b) that noticing is a necessary condition for L2 acquisition Ellis (1997) helped clarify Schmidt's hypothesis and the place cf noticing in L2 acquisition by proposing the following model:

Figure 2: The process o f learning implicit knowledge (Ellis 1997, p 119)

As can be seen from the Figure 1, Ellis has based his model on current theories o f L2 acquisition, where two main stages are seen to be involved in the process o f input becoming implicit knowledge The first stage, in which input becomes intake, involves learners noticing language features in the input, absorbing them into their short-term memories and comparing them to features produced as output With regard to short-term memory, Kihlstrom (1984) suggested that 1) consciousness and short-term memory are essentially the same; 2) that for language items to be stored in long-term memory they must be processed in short-term memory; and 3) that items not processed into short-term memory or not further encoded into long-term memory from short-term memory will be lost Schmidt therefore concluded, "If consciousness is indeed equivalent to the short term store, this amounts to a claim that storage without conscious awareness is impossible" (Schmidt, 1990, p 136)

The second stage is one in which intake is absorbed into the learner's interlanguage system and changes to this system only occur when language features become part o f long-term memory

Despite these above-mentioned benefits, Wood (1993) also mentioned two unavoidable drawbacks o f self correction technique First, there is initial resistance on the parts o f some students to having to rewrite their work Some find it tedious and time-consuming because they have to not only correct and rewrite their work but also get acquainted with the correction code and get accustomed to figuring out how to correct errors on their own Second, as the students are working on the same assignment twice, there will be fewer assignments for them.

In summary, what has been discussed in the previous section touched upon the close relationship among feedback, self-correction and language acquisition Although there are several drawbacks, the benefits that self-correction technique brings about far outweigh its disadvantages

2.3.4 Feedback on different types o f errors

Apart from theorizing about the most effective CF methodology, researchers have also been concerned with the questions which errors to target when providing CF Various proposals have been advanced in relation to this issue

Corder (1967) made a distinction between errors and mistakes He uses the term “errors”

to refer to a systematic errors o f the learners’ underlying knowledge o f the language These errors display the learner’s current developmental level o f the target language On the other hand, he uses the term “mistake” to refer to incorrect forms caused by memory lapses, slips o f the tongue and other instances o f performance errors According to his view, teachers should be tolerate with students mistakes because “ it would be quite unreasonable to expect the learner o f a second language not to exhibit such slips o f the tongue (or pen), since he is subject to similar external and internal conditions (memory lapses, physical states such as tiredness and psychological conditions such as strong

emotioi) w hen performing in his first or second language (p.25) He also stated that L2 learners can correct their own “mistakes” with assurance, but their “errors” are not amendable since their current linguistic developmental stage, interlanguage, does not have the ability to recognize the difference their utterance and that o f the native speakers Corder argued that errors are indispensible in language learning because through errors, learners test their hypotheses about the nature o f the language they are learning Corder pointec out that despite teachers’ best efforts, the occurrence o f errors is inevitable because errors occur for many reasons The reasons can be: interference from L I, overgeneralization, an incomplete knowledge o f the target language, the complexity o f the target language and fossilization Therefore, teachers should be more concerned with how to deal with students’ errors than the simple identification o f them However, he also admitted that the problem o f determining what is learner’s mistake and what a learner’s error is one o f som e difficulty involves a much more sophisticated study and analysis o f errors than is usually accorded them.

Burt (1975; Burt & Kiparsky ,1972) distinguished between global and local errors

Global errors are defined as those that violate rules involved in the overall structure o f a sentence, the relations among constituent clauses or, in a simple sentence, the relations among major constituents Hendrickson (1979) noted that these errors will cause a native speakei either to misinterpret a written message or to consider the message incomp'ehensible within the textual context o f the errors Local errors, on the other hand, cause trouble in a particular constituent, or in a clause o f a complex sentence If compared with global errors in the aspect o f degree o f comprehensibility, local errors may mike a sentence appear awkward but cause a native speaker little or no difficulty in understmding the intended meaning o f a sentence Hendrickson stated that the teacher can determine which error is global or local by the degree o f communication destruction that the error has caused in the native speaker’s comprehension o f the conveyed message

Burt and Kiparsky (1972) further stated that global errors are higher on the hierarchy than the local ones Thus, a sentence with both global and local errors improves much more when a global error is corrected than a local one, or even a group o f local ones.

Ferris (1999) proposed a distinction between “treatable” and “untreatable” errors The

former “occur in a patterned, rule-governed way” (p.6); for example, verbs, subject and verb agreement, noun ending, articles, pronouns and spelling, while the later are

“indiosyncratic” and students can hardly refer to any handbook or set o f rules to correct these errors, namely lexical errors, missing words, unnecessary words, and sentence structure He suggested that CF would be most likely to be successful when directed at treatable inaccuracies

In fact, problems exist with the above mentioned classifications o f errors Ellis (2009) argued that tne dichotomy between errors and mistakes is not as strict as Corder (1967) presented it to be, and stated that “the gravity o f an error is to a very considerable extent a matter o f personal opinion” (Ellis, 2009, p.6) Ferris et al (2000) reported considerable progress in reducing errors for verb tense and form (‘treatable” ), little progress for lexical (“untreatable”) and noun ending errors (“treatable”), but regression was seen for sentence structure (“untreatable”) and article errors (“treatable”) Similarly, Ferris and Roberts (2001) found a decrease in verb and noun ending errors, but for the use o f articles, an increase in accuracy rate was reported Ferris et al (2000) and Ferris and Roberts (2001) agreed that students were more capable o f editing the “treatable” categories (namely verbs, noun endings, and articles) than “untreatable” categories (namely word choice and sentence structure) However, as Ellis (2009) claimed, there are

no theoretical grounds on which teachers or researchers can decide whether an error is simple or portable The same holds for Ferris’ dichotomy between treatable and untreatable errors For example, “articles” errors are classified as “treatable” errors because there are specific rules for using articles However, the study o f Ferris et al

(2000) showed that students make regression in using them Therefore, the treatability o f

“articles” errors can be doubted Ha (2010) mentioned that for “unbeatable” errors, teachers cannot expect students to improve in a short run They really need a lot o f time and efforts to enrich their knowledge as well as increase their exposure to the target language in order to select words appropriately and use correct expressions for what they

to express

2.3.5 Students’ perceptions and preferences for teachers’ corrective feedback

In addition to a great many studies on writing feedback, researchers have been much interested in students’ perceptions and preferences for teachers’ WCF Hedgcock & Lefkow itz’s (1994) study o f EFL and ESL college students revealed that EFL college students prefer and value teacher feedback and correction on grammatical, lexical, and surface level features more than those on content and style, whereas ESL students prefer feedback on content to feedback on form Their participants also expressed moderate preference for the use o f error codes, and both disliked the teachers’ use o f red pen

In order to find out what L2 students expect from the written teacher feedback, Leki (1991) used a questionnaire to investigate E S L ’ students’ preferences for error correction

in college-level writing classes The findings showed that students considered grammar, spelling and choice o f vocabulary to be important She discovered that having error- free work was also important to students, thus they expect their teachers to correct all the written errors The reason for students’ preferences for direct feedback was that indirect feedback sometimes cannot provide enough information to correct syntactic errors

R ahim i’s (2010) study o f EFL students at an Iranian University revealed that Iranian EFL students prefer to receive feedback on more global and meaning carrying grammatical items such as transitional words and sentence structure than the micro-aspects such as

spelling and prepositions, which do not have much impact on the meaning and com m inication ideas The results o f the study also showed that the students’ writing ability, as indicated by their writing scores, influences their views about three grammatical units- verb tenses, adverbs, and punctuation, which they consider moderately important The think-aloud data revealed that more skilled students writers attach more importance to these grammatical units because they are more concerned with their writing accuracy and, more importantly, with the clarity o f the message through their writing as well as communication o f ideas On the other hand, the less skilled

w ritersjust look at the effect o f errors at the sentence level and their views are strongly influenced by the frequency with which their teacher attends to an error Finally, the result hdicate that although the students consider direct teacher feedback necessary, they

do not Ind it helpful since they are not required to revise their essays The result showed that thi students regard self-editing illogical and indirect teacher feedback not very helpful due to the fact that their teachers have not explained the procedure o f self- editing and the logic behind indirect feedback to the students Rahimi (2010) suggested that teacher, should instruct and practice self-editing, indirect feedback, and peer feedback in class aid elucidate their logic and purposes to students As a result, the students would appreciite and value such practices

Lee (2005) investigated L2 secondary students’ perceptions, belief and attitudes about error o rre ctio n in the writing classroom The study demonstrated that a huge gap betweei teachers’ practices and student preferences in error correction does not exist For exampl;, although teachers tend to mark errors comprehensively, students also tend to prefer teachers to m ark errors comprehensively Similarly, although a significant number

o f teaclers correct all errors for students, students also prefer teachers to correct all their errors, n both cases there is a tendency for students to ask more o f teachers, as a greater percentige o f students prefer teachers to mark errors comprehensively and to correct them drectly Interestingly, this study has found that 76.3% o f the students prefer