The effects of indirect written corrective feedback on non english majored students improvements of grammatical accuracy in writing m a

VIETNAM NATIONAL UNIVERSITY – HO CHI MINH CITY UNIVERSITY OF SOCIAL SCIENCES & HUMANITIES FACULTY OF ENGLISH LINGUISTICS & LITERATURE THE EFFECTS OF INDIRECT WRITTEN CORRECTIVE FEEDBAC

VIETNAM NATIONAL UNIVERSITY – HO CHI MINH CITY UNIVERSITY

OF SOCIAL SCIENCES & HUMANITIES

FACULTY OF ENGLISH LINGUISTICS & LITERATURE

THE EFFECTS OF INDIRECT WRITTEN CORRECTIVE FEEDBACK ON NON-ENGLISH MAJORED STUDENTS’ IMPROVEMENTS OF GRAMMATICAL ACCURACY IN

CAO THỊ QUỲNH LOAN, Ph.D

HO CHI MINH CITY, January 2020

my pressure while doing this thesis I own her a great debt of gratitude

I also appreciate the assistance of academic staff of Nguyen Tat Thanh International Institution of Education who made it a good condition for me to complete the thesis and the students who gave

me invaluable support I was appointed to be in charge of teaching two Pre-Intermediate classes upon my request, which made it convenient for me to conduct this experimental study The students were also kind and supportive They fully cooperated and made their effort in the tests given and in fulfilling my requirements That helped data be collected sufficiently for the research

My appreciation also goes to my teammates who gave me amazing support during my hard time with the thesis Being in the thesis group and struggled to complete the mission, we thoroughly understand each other’s difficulties and were willing to listen to each other’s problem They were always by my side, gave me considerable encouragement and assisted me to the best of their ability It would be hard for my thesis to be fulfilled without Ngoc An’ s detailed comments that made me realize my weaknesses in language use and his diligence in revising the thesis format or brother Vu Bao’ s valuable advice which helped me keep moving on when I was stuck somewhere with my own “thesis sorrow” It could be a mistake without mentioning Diem, Vi and Phuc who gave me tender loving care and always respond to my questions regarding research areas I was not clear about I also want to send many thanks to Ba Tong and Minh Giang who were with their hectic schedule but still spent time sharing me their experience in doing thesis and giving me advice on what to do

Finally, I am grateful for my mother and husband who gave me spiritual support for this academic pursuit They were those who always showed care to the progress of my thesis and shared me whatever burden I had so that I could wholeheartedly concentrate on doing my thesis

ii

STATEMENT OF ORIGINALITY

I certify that this thesis entitled “THE EFFECTS OF INDIRECT WRITTEN CORRECTIVE

FEEDBACK ON NON-ENGLISH MAJORED STUDENTS IMPROVEMENTS OF GRAMMATICAL ACCURACY IN WRITING” is the intellectual product of my own work, to

the best of my knowledge and ability

I hereby declare that this thesis has not been submitted, either in full or in part to any institutions for any purposes

Ho Chi Minh City, November 8th, 2019

Nguyen Nhat Minh Chau

iii

RETENTION AND USE OF MY THESIS

I hereby state that I, Nguyen Nhat Minh Chau, being the candidate of the degree of Master in TESOL accept the requirements of the University of Social Sciences and Humanities regarding the retention and use of my thesis deposited in the library

In terms of these conditions, I agree that the original of my thesis can be made accessible to individuals for the purpose of research and study with the conditions established by the University’s library for the care, loan or reproduction of the thesis

Ho Chi Minh City, November 8th, 2019

Nguyen Nhat Minh Chau

iv

TALBE OF CONTENTS

ACKNOWLDEGEMENTS i

STATEMENT OF ORIGINALITY ii

RETENTION AND USE OF MY THESIS iii

TALBE OF CONTENTS iv

LIST OF TABLES ix

LIST OF FIGURES x

ABSTRACT xi

CHAPTER 1 INTRODUCTION 1

1.1 Background to the study 1

1.2 Research aims 7

1.3 Research questions 7

1.4 Research hypotheses 7

1.5 Significance of the study 7

1.6 Outline of the thesis 8

CHAPTER 2 LITERATURE REVIEW 9

2.1 Error making in second language learning 9

2.1.1 Errors and mistakes 9

2.1.2 Sources of errors 10

2.2 Providing feedback to learners’ performance 12

2.2.1 Definition of feedback 12

2.2.2 Categorization of feedback 14

2.2.2.1 Positive and negative feedback 14

2.2.2.2 Oral and written Corrective Feedback 15

2.3 Written Corrective feedback 17

2.3.1 Researches into the effectiveness of WCF 17

2.3.2 Strategies of giving WCF 19

v

2.3.2.1 Focused vs unfocused WCF 19

2.3.2.2 Direct vs indirect written corrective feedback 23

2.3.2.3 The effectiveness of WCF types 27

2.3.2.4 Values of WCF for treatable and untreatable errors 32

2.4 Grammatical knowledge in second language acquisition 34

2.4.1 The role of grammar in writing 34

2.4.2 Improving grammar in writing 37

2.5 Second language writing 39

2.5.1 The relationship between speaking and writing 39

2.5.2 The writing process 41

2.5.2.1 Model of the writing process by Hayes (1996) 41

2.5.2.2 Model of the writing process Bereiter and Scardamalia (1987) 42

2.5.2.3 Evaluating the two models 43

2.6 Conceptual framework 45

2.7 Gaps persisting 46

2.8 Summary 46

CHAPTER 3 METHODOLOGY 48

3.1 Research setting 48

3.2 Participants 48

3.3 Research design 51

3.4 Research instruments 53

3.4.1 Guided writing tasks 54

3.4.2 Interview 56

3.4.3 Pilot the research instruments 58

3.4.3.1 Piloting the tests 58

3.4.3.2 Piloting the interview 59

vi

3.5 Treatment and data collection procedure 59

3.5.1 Targeted errors for treatment 59

3.5.2 Treatment 60

3.5.3 Procedure 62

3.5.3.1 Placement test 62

3.5.3.2 Pretest 63

3.5.3.3 Posttest 65

3.5.3.4 Delayed posttest 65

3.6 Data analysis 68

3.7 Chapter summary 69

CHAPTER 4 70

RESULTS AND DISCUSSION 70

4.1 Tests 70

4.1.1 Preconditions of statistical analysis 70

4.1.2 Participants’ pre-existing knowledge to the treatment 71

4.1.3 The difference in the short-term effects of indirect WCF on the improvement of grammatical accuracy in the posttest compared to those of direct WCF 73

4.1.4 The difference in the long-term effects of indirect WCF on the improvement of grammatical accuracy in the delayed posttest compared to those of direct WCF 74

4.1.5 The effects of indirect WCF within groups 76

4.1.5.1 Between the pretest and posttest 76

4.1.5.2 Between the pretest and delayed posttest 77

4.1.5.3 Between the posttest and delayed posttest 79

4.1.6 Types of errors committed 81

4.1.7 The improvement of grammatical accuracy for different error types after being treated with indirect WCF 82

vii

4.1.8 Discussion 85

4.2 Interview 89

4.2.1 The needs for error correction in writing 89

4.2.1.1 Making grammatical errors is a perennial problem of students in English writing 89

4.2.1.2 Addressing students’ grammatical errors has still been inadequate 91

4.2.2 Students’ opinion regarding indirect WFC (vs direct WCF) 93

4.2.2.1 Students’ support for the method 93

4.2.2.2 The observed positive effects of indirect WCF on students’ writing 97

4.2.3 Students’ difficulties with indirect WFC 100

4.2.4 Students’ suggestions for the improvement of the indirect WFC 101

4.3 Chapter summary 103

CHAPTER 5 CONCLUSION 104

5.1 Summary of the current study 104

5.2 Pedagogical implications 106

5.3 Limitation of the study 107

5.4 Recommendation for further studies 108

REFERENCES 109

APPENDICES 117

APPENDIX 1 118

APPENDIX 2 123

APPENDIX 3 136

APPENDIX 4 140

APPENDIX 5 142

APPENDIX 6 145

APPENDIX 7 152

ix

LIST OF TABLES

Table 2.1 A taxonomy of oral CF strategies (Ellis, 2009) 16

Table 2.2 A Taxonomy of Written CF Strategies (Sheen and Ellis, 2011) 26

Table 3.1 Placement test results by the number of correct answers 50

Table 3.2 Placement test result by the percentage of correct answers 50

Table 3.3 T-test for the Placement test scores 51

Table 3.4 Error codes used in the study 61

Table 3.5 Procedure of the study 66

Table 4.1 Tests of normality for the scores of the experimental and comparison groups in the pretest, posttest and delayed posttest 71

Table 4.2 Results of Mann-Whitney U Test for the pretest scores 72

Table 4.3 Results of Mann-Whitney U Test for the posttest scores 73

Table 4.4 Results of Mann-Whitney U Test for the delayed posttest scores 75

Table 4.5 Results of Wilcoxon Signed Ranks Test for the Pretest and Posttest of the Comparison and Experimental groups 77

Table 4.6 Results of Wilcoxon Signed Ranks Test for the Pretest and delayed posttest of the Comparison and Experimental groups 78

Table 4.7 Results of Wilcoxon Signed Ranks Test for the Posttest and Delayed posttest of the Comparison and Experimental groups 79

x

LIST OF FIGURES

Figure 4.1 Grammatical aspects that participants struggled the most in pretest, posttest and delayed posttest 81 Figure 4.2 The number of grammatical errors committed in pretest, posttest and delayed posttest 84

Keywords: written corrective feedback, indirect written corrective feedback, written feedback,

grammatical accuracy, writing performance

1

CHAPTER 1 INTRODUCTION 1.1 Background to the study

Making errors is deemed inevitable in the process of learning a foreign language, especially in productive skills including speaking and writing As Edge and Garton (2013) maintain, the production of non-standard forms is a part of the main sequence that almost all learners follow when they acquire a new language Thus, there is always a need for teachers to provide feedback in the form of correction for the amelioration of students’ areas of weaknesses According to Harmer (2007), correcting is seen as a form of feedback, and error correction or corrective feedback (CF) is very much the same-the process that teachers employ

in response to students’ errors Brown (2007) also mentions that the learning process can be impeded if learners “do not commit errors and then benefit from various forms of feedback on those errors” (p.257) It means that making errors in language learning should not be treated as

an adversity that should be avoided, but rather, be considered a necessary process that allows learners to learn before getting a full command of a language At the same time, teachers are responsible for giving appropriate feedback so as to encourage language development, giving learners opportunities to learn from their committed errors That serves as a plausible reason why theories of both first and second language acquisition (SLA) have adopted various positions regarding the significance of corrective feedback They are particularly into learning how a second language (L2) is acquired and what measure should be taken for learners to overcome their errors while acquiring a target language (Bitchener & Ferris, 2012)

Many researchers have justified the supportive role of CF from various standpoints According to Long (2007), error correction allows learners to pay more attention to the gap between their language and the target form in order to make appropriate revisions That is very similar with the view of Schmidt (2001) as he believes one of the best ways for teachers to steer students’ notice to the linguistic forms is to offer CF It can be seen that these two perspectives have acknowledged the supportive role of CF in enhancing the notice on forms which is vital for the improvement of grammatical aspects in second and foreign language learning Making even a stronger point, Black and William (1998) (as cited in Harmer, 2007) assert that feedback on students’ work is likely to have more impacts on achievement than any other single factors These views are convincing to a great extent It is understandable that the

2

improvement in language learning does not merely depend on how much learners acquire from the input of knowledge from teachers, but it is also about the experience they can learn from their mistakes in order not to commit them again That is to say, it is beneficial for learners to pay close attention to their language deficiencies and to work on improving those aspects in order to progress in language learning This process is significantly aided by receiving instructor’s CF

However, these ideas have encountered contradictory perspectives At this point, it may

be a flaw without mentioning the claim of Truscott (1996) in which he firmly asserts that correction had only minimal effects on learners’ ability to write accurately, and that if there was any positive sign, it would not be significant This conclusion was made after he had synthesized, evaluated and analyzed a number of relevant studies When this contention was later rejected by Ferris (1999), another paper of Truscott (1999) was released with an attempt

to point out weaknesses in the work of his opponent and criticized that the opposition was unfounded, hence not being persuasive More importantly, it was Truscott’s conviction in the second paper that error correction embraced more harm than help to L2 learning, and if the correction of grammatical had any appeal, it would rest on bias That was not only an outright rejection to any idea going against his belief but also strengthened what he defended so far

Interestingly, that is not the end of the heated debate When Chandler (2003), from his experimental study, concluded that error correction was beneficial for writing by increasing the overall accuracy in revisions and subsequent writing tasks, and that it does not harm the fluency

or overall writing quality of students; Truscott (2004) gave another response to these claims

He suggested that some crucial distinction had been overlooked and thus, they were more like conjectures rather than research findings While emphasizing that grammatical error correction was detrimental, Truscott strongest conclusion was made that “correction is a bad idea” (p.342) In fact, being rather arbitrary, the affirmation from Truscott has sparked a large number of arguments, and those “fights” have yet to reach an end However, it is the driving force for many other studies to be conducted after that in order to verify the effectiveness of

CF in general and written corrective feedback (WCF) in particular, which can clarify the issue and is beneficial for educational practice Although Truscott’s claim seems to be drifted out from the mainstream of other perspectives, once the role of WCF is still under conflicts, it is

to ignore whatever type of errors learners make Depending on different genres of writing tasks, teachers’ focus on the kind of errors that learners commit may not be necessarily the same However, Hyland (2003) believes that when it comes to correction, errors related to grammar usually come out on top of the list as they are an obvious problem of L2 leaners It is not surprising to realize that most teachers seem to prioritize this area more than others Part of the reasons may come from the fact that the accuracy of a writing product can be of great importance, especially when it is made for audiences belonging to a certain community, as Harmer (2007) illustrated, including business or academic environment Therefore, teachers should avoid emphasizing on ideas and neglect forms when addressing learners’ errors in writing, and that the separation of forms and meaning seems impossible in the practice of teaching a language because a message can only be conveyed successfully once appropriate forms are selected

In this sense, it is very much similar to the context of language teaching in Vietnam According to Van (2009), although the development of practical communication skills is stressed under the direction of Vietnamese Ministry of Education and Training, it does not necessarily reflect the reality in classrooms Much of the attention has been paid on the enhancement of reading comprehension, vocabulary and structural patterns at high school levels to prepare students for the entrance examinations to colleges and universities which is considered the milestone of their learning journey In tertiary education, English is still the compulsory module that students across majors have to fulfill, and the enhancement of English skills are emphasized to bring students to a certain level of proficiency to prepare them for future jobs Learning “how to use the grammatical structures and how to use them in appropriate contexts” (p.13) has still been highlighted when it comes to the writing skill Therefore, it is worth bearing in mind that the emphasis on forms in general and in teaching

is especially true for non-English majored students who do not often passionately devote themselves to learning this language because it is not the priority of their study Nevertheless,

as long as the accuracy in using a language is still stressed no matter what major students are from, finding measure to help students better deal with grammatical aspects in writing is significant Feedback in the form of grammatical error correction can be one of the solutions

to this problem, and conducting studies in this area will be meaningful for educational practice

Of the factors that contribute to the effects of WCF on students’ improvement in writing, teachers’ strategies for providing WCF have also been an area of attention that initiates

a number of researches First, depending on the outcomes that teachers wish their students to achieve and the aspects that teachers want to steer students’ attention to, WCF can be made on specific types of errors in the case of focused CF or alternatively on all sorts of errors committed in the case of unfocused CF (Ellis, 2008) Having said that, it is the second aspect that attract the researcher ’s attention the most; to be more specific, the indirect and direct WCF

As mentioned above, providing WCF as a way to respond to students’ written work is

as an essential part of teaching According to Edge and Garton (2013), of the many things that teachers have to do, giving feedback is among the most important tasks that they have to fulfill because it has always been students’ need to know how close they are getting to the target Nevertheless, the researcher has many times been in a dilemma to adopt correction methods in the way that benefit students to the utmost One of the aspects of WCF is whether the feedback should be made implicitly or explicitly In order words, whether providing them with direct correction known as the conventional method that has long been employed by the majority of teachers of English or the indirect one that allows them to be more responsible on their errors made would yield better improvement In fact, the available theoretical background has been with so many different opinions While several studies of second language acquisition (SLA) (Bitchener, 2008; Bitchener & Knoch, 2008; Ellis et al., 2008; Sheen, 2007) are in favour of

5

direct WCF, others (Ashwell, 2000; Ferris, 2006) support the indirect one, ascertaining that it

is helpful as students have a chance to get involved in the process of solving the problems of their own and take more responsibility for their progress

There have been some arguments that indirect WCF may not work for students’ improvements It is because students may feel reluctant as they are not provided with enough information to correct errors by themselves Even when their language competence is sufficient for them to do so, they are not sure whether their corrections are accurate or not (Chandler, 2003) However, as Ellis (2008) once said it was the level of language competency that should

be taken into account before adopting either strategy so that students’ load of attention could

be in accordance with their language capacity Thus, whether direct WCF is still appropriate for students regardless of their English level of proficiency and learning habits, or the indirect one yields a more positive effect is worth investigating Also, when grammatical errors are targeted, whether direct or indirect WCF is more effective to help learners reduce the number

of grammatical errors committed and boost the accuracy of their overall writing performance also needs to be clarified

In fact, at this point, Van Beuningen et al (2012) have come up with a conclusion after carrying out a well-designed study in which a comparison between the different effects of direct and indirect WCF is examined They find that the active role of various types of WCF is associated with certain sort of errors, meaning that their values are not necessarily the same for various error categories In detail, while direct WCF is effective for grammatical errors, the indirect one is beneficial in writing in the way that it helps shrink the quantity of non-grammatical errors and foster long-term acquisition However, it seems that such a conclusion makes the question regarding the constructive role of indirect WCF for grammatical errors as well as the one of direct WCF for non-grammatical errors left unanswered Moreover, as this idea is neither found in any previous studies nor verified by subsequent studies, it is still of great importance to confirm the validity of this claim by further experimental studies

More importantly, it is worth mentioning the fact that while the majority of studies related to strategies of giving WCF have been conducted in foreign counties, only a few have been carried out in the context of Vietnam In fact, the topic of feedback has sparked interest

in some Vietnamese researchers who have recently investigated a number of topic-related aspects Some of the remarkable studies to be mentioned include the one by Nguyen (2016) on

6

the role of peer feedback on writing or the study by Dang (2016) on noticing-based collaborative feedback on writing There is also a study on the effectiveness of corrective feedback for the acquisition of L2 pragmatics which targets at speaking by Nguyen et al (2018) It is clear that these studies focus on learning either feedback given by students or feedback on oral performance The researcher realizes that while the effects of different feedback giving strategies in general and of direct and indirect feedback by teachers on writing

in particular is an important issue to learn about due to its practical values of language learners, this subject matter seems to be an area that has yet to be paid sufficient attention in Vietnam Therefore, the researcher has a strong motivation to have a step into this new area

In light of the mentioned aspects, the motives for the researcher to conduct the study emerge As discussed above, making errors in using a foreign language is a natural process that enables language learners to step forward However, it is of great importance for language teachers to adopt appropriate strategies to support students during this stage of learning When

CF is regarded as one of the methods that bring positive aspects, especially to the improvement

of linguistic forms, choosing the most effective approach, namely direct or indirect CF to respond to the written work has sometimes place teachers into a dilemma The thing is that a direct contrast between the two strategies has yet to be the emphasis in empirical studies That

is not to mention the fact that findings related to this area have yielded either inconsistency or conclusion that needs further verification

Getting an insight into the effects of indirect WCF compared to the direct one on learners’ grammatical accuracy in writing will be meaningful It helps give a confirmation to the available knowledge, strengthening the valuable theories that can be used as the background for further researches It also holds practical value as the researcher can be more confident to employ suitable correction strategies to benefit students of EFL, better improving their grammatical accuracy in writing and enhancing their English proficiency Without doubt, teachers of English in Vietnam will also get a great deal of benefits from the study findings because this can be the very first time that the study on this topic is made in writing classroom contexts in Vietnam and for Vietnamese students with their typical characteristics in language learning

7

1.2 Research aims

Due to the unresolved aspects related to the effects of error correction strategies in general and the indirect WCF in particular, it is worth investigating the issue from a different angle Therefore, this study is conducted aiming at:

- examining the effects of teachers’ indirect WCF on the improvement of grammatical accuracy in writing compared to the conventional method, namely direct WCF In finding out the answer, it is expected that grammatical aspects that students are struggling with will also emerge

- finding out students’ perspectives towards the employment of indirect WCF as a way

to respond to their errors in writing for pedagogical implications

1.3 Research questions

This study seeks to answer the following two questions:

a Does indirect WCF enable students to improve the grammatical accuracy of their writing as opposed to the direct one?

b What do students think about the effects of indirect WCF on their writing performance?

b Students will perceive that indirect WCF has a positive effect on their writing

1.5 Significance of the study

The research conducted promises to yield an insight into the role of indirect WCF in improving grammatical accuracy in paragraph writing Using indirect WCF has yet to be a common practice that teachers of language apply to respond to their students’ written products Besides, those who want to have a try are often in a dilemma as they are not certain whether the strategy would bring positive effects on students’ ability to write more accurately The

8

result of this study is expected to bring to light the effects of an alternative of giving WCF, an indirect approach, on the improvement of using some grammatical points that are basic but vital in writing That helps bolster teachers’ confidence in adopting the most suitable strategy

of giving WCF that benefits their students’ writing capacity

Furthermore, by obtaining an insight into students’ perspectives regarding the effects

of indirect WCF on their writing performance, the result of this study can be used as an implication for teachers to modify their teaching practice in order to better support students during their English learning process At the same time, the research result is believed to be a valuable source of reference for further research on related issues

1.6 Outline of the thesis

The thesis is comprised of 5 chapters The first chapter gives a background to the study

It includes the role of feedback in language teaching and learning, important facets that affect the effect of the practice and existing debates on the effectiveness of different WCF strategies

as well as the research aim, the questions to which the study is seeking answers, the formulated hypothesis to anticipate the study result, and the contribution of the study The second chapter which is literature review goes deeper into theoretical background and existing knowledge in company with findings of some other studies in the same area, which helps clarify aspects that this study is connected with The following chapter describes how the study is to be operated

To be more specific, it illustrates the research methodology used to get the answers to the research questions including research design, setting, participants, the procedure to carry out the study and the method of data analysis Chapter four provides the analysis of data and the discussion to compare the results gained with other studies on the same subject matter Finally,

a summary of the main points of this study is presented in the final chapter

9

CHAPTER 2 LITERATURE REVIEW 2.1 Error making in second language learning

2.1.1 Errors and mistakes

Errors can be a common feature that almost all language learners have in the process of learning a new language A number of definitions of errors have been put forth According to Chun et al (1982), errors indicate faulty or incomplete learning in the use of linguistic items compared to what produced by fluent speakers of that language Liski and Putanen (1983) suggest that an error arises when speakers unsuccessfully follow the patterns or manner of speech of “educated people in English speaking countries” (p.227) The two definitions, notwithstanding their detailed description, focus only on errors in the spoken language while productive skills also involve writing Moreover, errors are somehow defined in an unsystematic way, not to mention the ambiguity of some phrases in the definitions Therefore,

a more comprehensive definition of errors by Lennon (1991) can be relied on In this view, errors are seen as “a linguistic form or combination of forms which, in the same context and under similar conditions of production, would, in all likelihood, not be produced by the speakers' native speaker counterparts.” (p.182)

However, as errors denote something incorrect in language use, it is necessary for a distinction to be made between ‘errors’ and ‘mistakes’ Brown (2007) states that mistakes are very much the same as slips, represent a failure in using a language system correctly On the other hand, an error can be a signal of a deficiency in one’s target language competence, which

is not in the case of a mistake To put it another way, mistakes are more of performance errors that learners commit when they know the language system but fail to use it correctly Errors,

by contrast, occur when learners’ existing language system is incorrect In this sense, a mistake can be corrected by language users themselves when it is pointed out, but errors require more explanation from instructors to be rectified

Due to the fact that a language learner makes both mistakes and errors during their learning process, teachers should be fully aware of the difference between the two in order to treat them appropriately Like Harmer (2007) suggests, mistakes or errors affect the type of feedback teachers are to give to as a support Learners’ mistakes can be put right thanks to teachers’ reminders, yet an error can only be corrected by learners if they thoroughly know

10

what goes wrong with their systematic language competence Thus, supporting learners with their errors calls for teachers’ devotion of more time and effort What’s more, teachers may be sometimes left with no way to identify whether learners are making mistakes or errors because learners’ self-correction ability is somehow hidden Therefore, it is necessary for teachers to pay close attention to observe the whole learning process to judge learners’ current levels of language competence and assist them effectively That is to say, learning about the radical differences between mistakes and errors has significant impacts on language teaching practice

2.1.2 Sources of errors

As mentioned earlier in the previous chapter, making errors during the process of learning a second language is something inevitable In a sense, that fact can be explained by the complexity of second language acquisition (SLA) VanPatten (2004) admits that L2 learning is very challenging because of the existence of implicit linguistic system with an abundance of “lexical entries and their features and forms, an abstract syntactic system, a phonological system, and rules on pragmatic use of language, among other components related

to language.” (p.5) Moreover, SLA, at the same time, contains multiple processes that a language learner has to go through to be able to master the target language Thus, it can be inferred that being proficient in a foreign language requires learners the ability to handle a wide variety of elements of linguistic knowledge, and it seems hard for one to be a language master without experiencing trials and errors

Another explanation for the cause of errors is proposed by Harmer (2007) He explains, when the acquisition of a second language is added to the existing deep knowledge of the mother tongue, confusions will unavoidably come into being, and errors arise as a consequence

In classroom environment, as the linguistic features of first and second language embrace many differences, teachers of English may frequently find that their students’ poor use of a second language is the result of their first language’s interference It is because language users tend to relate and integrate what they have already known into learning new things Having said that, Harmer believes errors exist in every stage of language development, and instead of being considered as a weakness, such errors should be regarded as a tool with the function of re-shaping learner language capacity towards a full language mastery

Brown’s (2007) elaboration in justification of errors sources is based on a well-rounded view By analyzing the cognitive processes learners go through while learning a second

11

language, he points out three sources where most learners’ errors are generated from, including interlingual transfer, intralingual transfer and the context of learning In fact, the term

‘interlingual’ was first coined by Weinreich (1953) and was later adapted to “interlanguage”

by Selinker (1972) who made it a more commonly used term to describe a language system having features of both native and target language Interlingual transfer, according to Brown (2007), often occurs at the beginning stages of learning a new language when learners tend to draw upon their native language to have preference for the formation of the target language In that sense, the concept of interlingual transfer is similar to Harmer’s (2007) proposition on the transfer from students’ first to second language In can also be inferred that teachers’ familiarity with learners’ native language can assist their ability to detect the cause of their students’ errors Intralingual transfer happens at the much later stages of learning compared to interlingual transfer It is when language learners already acquire the target language to a significant extent and use that knowledge and experience to overgeneralize the use of new structures Such overgeneralization is very much the same as incorrect inferences and often leads them to some inappropriate usage of the new language, especially in terms of grammar Last, the context of learning is mentioned as another important factor contributing to the emergence of errors In this aspect, teachers’ misleading explanation, learning materials’ faulty presentation of knowledge or even teachers’ provision of wrong information are all the illustrations for how learners’ acquisition of the target language can be negatively affected by the learning context As can been seen, while the first two sources of errors come from learners themselves, the last one is more from teachers, materials and learning environments

However, making errors should be seen as a positive sign of language development Brown (2007) believes that L2 learning should be seen as the acquisition of whatever types of linguistic knowledge that are different from their L1 The process of acquiring a new language

is very much the same as a “creative construction” (p.255) It is where learners try to test certain hypothesis related to the target language from various knowledge sources, such as the available knowledge from their L1, the limited knowledge L2 they have had so far, general knowledge about a language and even the knowledge about life, people and the world around them For this reason, making errors is a part of this process, and such errors primarily emerge before learners can reach any confirmation about the how the new language system is like Indeed, Brown emphasizes that by undergoing stages of learning with developmental errors and

12

hypothesis testing, learners slowly move closer to the language system of native speakers That

is to say, errors in L2 learning should not be necessarily seen as a reflection of an imperfect feature of language learning Other than that, it is involved in logical and systematic stages of acquisition Besides, if errors committed can help learners to realize their problems and move closer to language mastery, it is not necessarily to treat them as a negative sign In this sense, having a broader perspective on learners’ errors is necessary for teachers of language to judge learners’ capacity

2.2 Providing feedback to learners’ performance

of learners’ work It tells whether students’ answer to a question is true or false or to what extent it can be altered for the appropriateness Hyland and Hyland go on to state while the source of such feedback can be varied, teachers’ feedback is considered the most widely-used type of feedback in L2 classroom This idea is completely justifiable It is because among the available sources of feedback such as peer feedback or automatic feedback provided by programmed software that has become very popular in the era of technology, teachers are still the most reliable source as they not only give feedback but also are capable of explaining and directing corrections based on their expertise That is especially popular in educational environments where a teacher is at the center of a classroom, and students heavily rely on them for every stage of learning

Sadler (1989) defines feedback as the “information about how successfully something has been or is being done” (p.120) This definition is very much the same as the proposition by Hyland and Hyland and Hyland (2019) mentioned above because the way feedback is perceived is identical Besides, indicating how close students’ performance is to a language target, feedback, as defined by Sadler, can also be made either at the end of a process, that is

13

when students complete a given task or during an ongoing process where feedback may be given more than once to support students at the right time He also explains that different types

of skills (such as physical, intellectual or social skills) cannot be achieved by being told what

to do, but by practicing in a supportive environment where feedback comes into play Obviously, language skill is not an exception in this view and the role of practice has always been highlighted when language learners want to improve their language competence That is why teachers of language usually motivate their students with the popular saying “Practice makes perfect” However, just because students devote a great deal of time to practicing, it does not mean they are able to master the target language Without feedback given to guide students to the correctness, they are likely to keep repeating their errors from one piece of work

to another, and that can be detrimental to their ability to write accurately Therefore, overcoming errors and mistakes thanks to feedback is the same as the learning process in which teachers’ feedback carries with it an instructional purpose to direct a task or performance Upon reflecting the definition of feedback by Ramaprasad (1983), Sadler emphasizes that a piece of information can only be seen as feedback only when it helps fill or alter a gap, and therefore, an effective feedback cannot be reached if feedback provider is not capable of changing an outcome due to lack of knowledge or power This perspective comprises of two aspects with important indications First, feedback seems to have lots of things to do with correction, and it is carried out with the final purpose to better something problematic such as students’ inaccuracy in using certain language items to fulfill a task Second, it can be seen that Sadler’s view, at the same time, highlights the role of feedback given by teachers who supportively assist their students to shorten the difference between what they have done and what they should do This opinion is consistent with the view of Hyland and Hyland (2009) in that teachers’ feedback is the most common and reliable source of feedback which students can rely on for the improvement of their language use

Furthermore, feedback is considered an indispensable element of formative assessment conducted during the process of learning Feedback helps to modify teaching and learning so that students could improve their language competence In this sense, it is regarded as an assistance for learners to monitor their strengths and weaknesses in their performance so that what is of good quality can be boosted and developed while aspects relating to defects can be modified or deterred Hattie and Timperley (2007) contend that feedback not only provides

14

information for the confirmation of how good something is, but it also embraces certain forms

of direction that students could make use of for the enhancement of their work or performance

In order words, feedback could reduce the mentioned gap by restructuring students’ understanding, giving them confirmation of the correctness and incorrectness, indicating the necessity of adding information and other types of elements

In short, giving feedback to learners’ work is an inseparable practice of language teaching Feedback itself tends to be directed at areas of weaknesses with an aim to improve them By responding to learners’ language performance, teacher feedback during the learning process helps learners to know where their language proficiency is at and how to work harder

to reach positive learning outcomes

2.2.2 Categorization of feedback

2.2.2.1 Positive and negative feedback

Harmer (2007) states that language learners tend to receive teachers’ assessment in form of “praise or blame” (p 138) To make it simpler, the former can be seen as a kind of compliment given to signal students’ good work; the latter, on the other hand, indicates an existing problem This is very much the same as Petty’s (2004) idea where he regards the two mentioned aspects as “medals” and “missions” “Medals” are offered to students when they are doing well while “missions” are direction for the correction As pointed out earlier, feedback

is an element of assessment, so these views on assessment suggest that feedback has the same nature That means feedback can be made to indicate students’ good work and that they are on the right track, which is like a confirmation of good work from teachers so that students can confidently keep moving forward However, in the other case, feedback is made to signal errors that students have to overcome for the improvement, and after such feedback, working to improve their persisting areas of weaknesses is necessary

Indeed, according to Ellis (2009), feedback provided to students can be positive and negative Positive feedback affirms the correctness of students’ responses, and it is agreed that such feedback brings with it an impact on students’ motivation and progress Meanwhile the negative one indicates “a lack of veracity or linguistically deviant”, or to put it another way, negative feedback is “corrective in intent” (p.3) Such negative comment can be made by different ways It can be obtained through explanation or presentation of grammatical rules or

15

made in form of corrective feedback on students’ errors, and when that comes into practice, teachers can either make it directly through overt correction or indirectly by strategies that signal an existence of errors (Long, 1996) This interpretation by Long is very much in line with Ellis’s view when he suggests that corrective feedback constitutes a type of negative feedback From these opinions, the basic qualities of corrective feedback are emerged First, corrective feedback serves as an indication of language defects or certain type of errors from students’ work or performance; and it is necessary to know that “negative” here does not mean

to describe feedback as bad or harmful, but it is just an implication of something imperfect and

in need of revision Second, corrective feedback is made by teachers with their attention paid

to how linguistic elements are presented It is not separated from an intention to improve certain areas of weaknesses, directing correction either directly or indirectly

2.2.2.2 Oral and written Corrective Feedback

In terms of negative feedback, it is worth taking a close look at how it is operated for different skills Compared to receptive skills including reading and listening, productive skills with speaking and writing have certain features requiring feedback to have a more active role

As a matter of fact, teachers’ responses to students’ work on reading and listening are usually

in the form of ‘right’ or ‘wrong’, and it is not common that teachers make notes to give comments regarding how good or bad students have performed It is partly because the processes that students went through to come up with an answer seems to be out of teachers’ control, so instead of giving detailed feedback, just the confirmation of the correctness is provided by teachers Speaking and writing, by contrast, are a demonstration of how students use their available language to develop a product of language By accessing students’ oral or written work, teachers are able to obtain a picture of the way students use different materials

to construct language Thus, appropriate feedback can be offered to address different aspects

of students’ performance to help them improve these two skills

Within the same CF system, how teachers respond to students’ oral and written work is not the same although it is true that such kind of feedback is closely associated with correction

to improve the accuracy Basing on theoretical distinctions between explicit-implicit learning and explicit-implicit knowledge, Ellis et al (2006) suggest that oral CF can be classified depending on how explicit and implicit it is; in other words, the state of being whether overt or covert Concerning explicit CF, there is a clear indication of errors students make in their

16

utterance, while those are concealed in implicit one Regarding categorization of oral CF, it seems that Ellis’s classification is not the only one In fact, there are also other six strategies of oral CF suggested by Lyster and Ranta (1997) including explicit correction, metalinguistic comments, elicitation, recasts, repetitions and clarification requests However, as can be seen, they are not the two different systems because the system by Lyster and Ranta is just a more elaborate one Therefore, when these types of oral CF are considered under Ellis’s mentioned viewpoint, it is clear that while the first three belong to the explicit CF, the last three are of the implicit one

Besides, input-providing and output-prompting represent the two ways in which different oral CF strategies vary Being coined by Ellis (2013), the former means that teachers provide learners with the appropriate forms of the target language and the latter involves students’ self-correction of their committing errors In reality, these two are supposed to engage students into various types of cognitive processing and have values for different stage of learning It is also clear that they are not another categorization of oral CF but more of another dimension to view oral CF

On account of the mentioned classifications, Ellis (2009) develops a taxonomy of oral

CF strategies as follows:

Table 2.1 A taxonomy of oral CF strategies (Ellis, 2009)

Input- providing  Recasts  Explicit correction

Output- prompting  Repetition

a supportive role in students’ oral performance However, the significance of recast is valued

17

in content-based classrooms where the meaning of conversation is prioritized over the how language is used By giving recast as a feedback to students, teachers provide an alternative model of students’ attempted utterance, hence letting students realize the difference between their interlanguage and the target-like forms (Nicolas et al., 2001) Therefore, in the context of writing where the form of language is heavily emphasized, whether the effects of indirect methods of feedback provision still yield the same result is an area that need to be clarified Although oral CF is not the focus of the study, taking a look at how it works to improve students’ oral performance helps draws an analogy with written CF that may embrace certain significant differences

2.3 Written Corrective feedback

2.3.1 Researches into the effectiveness of WCF

As noted earlier, giving corrective feedback on students’ work has long been a common practice that has never been separated from language teaching In terms of writing, such pedagogical practice involves teachers’ attempts to rectify grammatical errors in L2 learners’ writing, which is considered essential and “non-negotiable” in L2 writing classrooms (Kang & Han, 2015, p.1) By addressing and responding to students’ impaired or problematic pieces of language use, WCF aims at arousing their awareness of what goes wrong and generating an effort for the enhancement

Having said that, researches into the importance of WCF for specific dimensions have been concluded with mixed results In particular, the role of WCF is examined when it is involved into the two distinct processes, the revision and the production of new pieces of writing For the former, Ashwell (2000), Ferris and Roberts (2001) recognize the positive effects of WCF as an editing tool during the revision for students’ improvement of accuracy

In their studies, students who received error correction on their pieces of writing made fewer errors upon redoing the same paper on another occasion Nevertheless, such conclusion soon encounters challenges regarding the value its holds As Ellis et al (2008) and Sheen (2007) points out, those studies do not reflect the effects of WCF for future compositions but just stay within the limitation of revisions Such a critical evaluation is justifiable because it is obvious that revision is not the ultimate goal of learning to write Being able to successfully amend a written work helps language learners step closer to mastering the writing skill, but it is more of

18

a practice during an ongoing process Only when learners can be able to skillfully use language

to produce new pieces of writing for various purposes and in different contexts, is their writing competence recognized

While the arguments against the effectiveness of WCF during revisions have not been thoroughly addressed in literature, the quest for empirical proof as regards the long-term effects

of WCF on learning has paved ground for a rise in investigations into this aspect Supporting the long-lasting learning effects of WCF is the result of the studies conducted by Van Beuningen et al (2005, 2008) who investigate the effects of indirect and direct WCF on written accuracy of L2 learners of Dutch Their findings are that WCF do not only enhance the precision of the revised version of particular productive writing tasks but also bring about significant learning effects to those who received WCF, meaning that fewer errors are committed for the subsequently new writing pieces Van Beuningen et al (2012) again stresses that WCT carries a potential for both revising and learning with effects retained over time rather than in a specific time for a particular text The idea is also in agreement with Chandler (2003) and Ferris (2006) when they examine into WCF in SLA

While the efficacy of WCF is proven to maintain for several weeks in these studies, the call for longitudinal evidence is a driving force for other researchers to continue with their investigations into WCF in longer time period A prime example can be the 10-month study by Bitchener (2008) on the effects of WCF English article system to 52 low-intermediate ESL students in Auckland, New Zealand The contribution of this study to the existing literature is that it highlights the long-standing effects of WCF on the acquisition of certain linguistic forms/structures (articles) over a period of 10 months

Nevertheless, Truscott and Hsu (2008) denies the result above They believe that there

is an insufficient evidence for such a conclusion and that students’ full awareness of errors in their writing during the revision does not necessarily result in the acquisition for subsequent writing pieces The assertion was made after they had investigated the writing performance of

47 ESL students divided into experimental group who received comprehensive WCF and the control group with no treatment Though their findings still confirm the significance of WCF

in helping students reduce errors on the revised piece of writing containing WCF, they saw no difference between the experimental and control group when they performed a new writing task a week later Thus, it is their strong belief that successful error reduction during revisions

19

could not be a predictor of learning, and error correction itself does not hold necessary values

as a teaching device However, this study also suffers from certain shortcomings as the treatment focused on a single type of correction which was in forms of making the location of errors (uncoded correction), so the effects of coded correction is left unanswered in their study

In light of the mentioned studies, it is confirmed that WCF has positive effects on writing, at least for revisions WCF allows students to pay attention to errors committed and successful edit their texts with a higher degree of accuracy However, its long-term effects have still been a matter of controversy Although most of the studies conducted see WCF with merits

in helping reducing a number of certain types of errors in subsequent writing, the opponents still have evidence to prove that it is not always the case Therefore, seeking for a confirmation from further studies will be meaningful

2.3.2 Strategies of giving WCF

2.3.2.1 Focused vs unfocused WCF

A distinction is made regarding the two approaches to giving WCF, in particular, the utilization of unfocused or focused WCF Focused and unfocused WCF are distinguished in terms of the quantity of error targeted for the correction For unfocused WCF, teachers may choose to correct all errors detected from students’ writing, which is considered a common practice in writing instruction that most teachers employ to treat students’ written work Sheen

et al (2009) regard unfocused WCF as the traditional approach that “involves providing correction on a wide range of errors in each piece of students’ written texts” (p.559) This definition reveals that unfocused WCF is not constrained within a particular type of error but a large number of error categories Besides, Van Beuningen et all (2012) and Shintani et al (2013) use the term ‘comprehensive error correction’ as an alternative to unfocused WCF, emphasizing the variety of errors that this type of feedback covers

In spite of these variations in the way unfocused WCF is regarded, the interpretation is consistently remained static in literature To be more specific, in the case of unfocused WCF,

no error category is eliminated from teacher’s correction, and in order to do so, teachers always have to pay meticulous attention to examining the all-round aspects of students’ written work

In the language classroom, addressing students’ errors this way requires teachers to devote time and certain level of concentration to complete the marking because any distraction may make

20

an error missed out That is probably why when discussing the practice of marking students’ written work, Bitchener and Ferris (2012) describe teachers as those who “would diligently and assiduously correct their students’ papers” (p.29) That highlights the fact that this strategy

is associated with a large amount of hard work and great care teachers have to put into Teachers’ motives for applying unfocused WCF in language teaching somehow arouses curiosity In fact, schools normally do not regulate methods teachers use to mark students’ constructed-response writing tasks However, there may be certain reasons why a multitude of language teachers still opt for unfocused WCF which is proven to be demanding rather than a more simplistic approach to correcting students’ texts From a practical standpoint, most teachers reach a general consensus that it is really uncomfortable to let any imperfection exist

in form of errors in a written work of students Compared to the oral performance when the emphasis is usually on meaning, written performance where students’ every single work is scanned and checked carefully is given with an importance on the linguistic form As what Bitchener and Ferris (2012) say, because a good piece of writing is considered free of errors, it

is instructors’ responsibility to provide “extensive, comprehensive correction of student themes

on an almost daily basis” (p 29) Therefore, in terms of writing, it is hard for teachers to forgive any error that may interfere with the writing performance

On the other hand, the case of focused WCF is in a sharp contrast with the unfocused one With focused WCF, only specific errors or problematic linguistic features are selected for the correction at a time while others are neglected (Ellis, 2008; Sheen et al., 2009, Van Beuningen et al., 2012) It can be seen that the range of errors targeted for correction is narrowed down in this strategy of correction Instead of examining all types of errors at the same time, teachers’ attention is paid on only one type, meaning that each type has a chance to

be reflected much on The provision of focused WCF, for that reason, is very much the same

as a highly selective process and is made with teachers’ clear purposes to call for students’ concentration on a targeted grammar point Although it can be insufficient to ignore other kinds

of errors in students’ written work, using focused WCF can be a good idea to address certain perennial error source and lead to improvements by reducing the number of that error type for subsequent writing pieces

It is worth noticing that focused WCF can also be further categorized into smaller units Ellis et al (2008) classified this approach into “highly focused CF” that targets a single error

21

type and “less focused CF” referring to the selection of more than one type of errors, but “still restrict correction to a limited number of pre-selected types” (p.356) For example, instead of paying attention only to the use of past simple verbs in the case of the former, the targeted grammatical areas for correction can also include past simple tense with articles and prepositions in the latter, meaning all errors of these types would be picked for corrections In reality, highly focused WCF is usually adopted in some particular contexts Teachers may use the method to address some kinds of errors that the majority of students tend to commit repeatedly from one writing piece to another, and those errors belong to an important linguistic group that affects meaning significantly without being addressed properly For other cases, less focused WCF with a group of errors directed for correction can also be teachers’ choice when those types of errors are closely related to a specific genre of writing The use of be, comparatives and superlatives, for example can be targeted when it comes to descriptive writing due to their roles in making description

A growing interest has been put forward to figuring out the different effects that focused and unfocused WCF exert Although in the study aiming at addressing the distinct effects between the two, Ellis et al (2008) do not find out any significant difference, it is still their strong belief that focused WCF will be more advantageous for learners since the attention to specific error types enables them to have a more thorough understanding about the nature of the errors and the correction They further add that if attention and understanding are the two important features for language acquisition, focused WCF promises to bring about more positive results for learners (Ellis, 2005, as cited in Ellis et al., 2008) Upon making a comparison with focused WCF, Sheen et al (2009) view unfocused CF as the approach that

“constitutes a relatively unsystematic way of correcting errors” and may “run the risk of overloading students’ attentional capacity” (p.559) Indeed, when it comes to writing, students’ reluctance to choose the most appropriate grammatical points to construct their writing pieces usually results in a large number of errors of many kinds That also means when unfocused WCF is adopted, it is inevitably that these students find it overwhelming to absorb all the corrections teachers make regarding what went wrong in their writing Gradually, it will be a source of distraction because students find it hard to handle each error type appropriately That may be why many teachers encounter situations where their students keep repeating the same types of errors from one text to another

22

Nevertheless, focused WCF is proven with outstanding merits Ellis (2009) contends that whereas unfocused CF may probably disperse learners’ notice, targeting fewer problematic features results in greater concentration, hence being more effective (Ellis, 2009) Likewise, Van Beuningen et al (2012), while admitting the validity of unfocused CF in the long run, propose that focused CF can result in learning from intensive attention as learners have chances

to receive the repeated evidence with multiple correction on the same error In reality, focused WCF is not the choice of many teachers as it is believed that neglecting any error is likely to lead to students’ later writing defects However, receiving multiple correction on the same type

of errors repeatedly gives students a chance to keep looking at certain perennial errors, and such attention is very valuable for the acquisition of a foreign language

To illustrate that point, it is necessary to take a look at the positive role of focused WCF found in Sheen et al (2009)’s study In this study, the different effects of focused WCF which targeted English article errors and the unfocused WCF which aimed to address errors of five grammatical features (articles, copula ‘be’, regular past simple, irregular past simple, temporal and locative prepositions) were examined on the written work of 80 ESL students at a US College The result of the study is that the focused WCF group outperforms the unfocused one with greater accuracy for the use of articles, both in short term and long term More interestingly, there is a decrease in the number of errors made in the focused group, not only for articles but also for other grammatical features which were not the targeted in the experiment To explain for such an interesting result, Sheen et al stated that the systematic way

of correction found in the case of focused WCF orients students to attend to forms in general, and that is not only beneficial for the article system, but also for a range of grammatical features

All in all, in the operation of focused and unfocused WCF, being ‘focused’ or

‘unfocused’ here does not depend on the number of errors teachers aim for correction, but sorely on the extent to which teachers’ attention is drawn on Focused WCF is not restricted to only one targeted grammatical error but is also for a small group of errors that may be closely connected to a certain genre of writing What’s more, the effectiveness of focused WCF on the improvement of grammatical accuracy outweighs that of unfocused WCF in a way that more attention from students is generated for the former than the latter However, it is suggested that

23

further studies need to be conducted in order to investigate how many structures could be included in a focused WCF so that its effectiveness would be still ensured

2.3.2.2 Direct vs indirect written corrective feedback

Researchers’ interests are not only kindled by the question of whether WCF should be comprehensive or selective, but also on whether WCF should be direct or indirect Regarding direct CF, it is consistently regarded as teachers’ correction of learners’ errors with the supply

of correct target language forms (Van Beuningen et al., 2012, Ellis et al., 2008) However, for

a detailed description, it must be the one made by Bitchener and Knoch (2009), who elaborately define direct CF as “the provision of the correct linguistic forms or structures by the teacher to the students above or near the linguistic error” (p.198) They further detail that the practice consists of “crossing out of an unnecessary word/ phrase/morpheme, the insertion of a missing word/phrase/morpheme, or the provision of the correct form or structure” (p.198) As can be seen from the definition, although direct WCF takes different forms, its different strategies still share a common feature To be more precise, the correction of students’ errors is made available with direct WCF, meaning that students who receive this form of feedback are fully aware of where problems are persisting in their writing and how such errors should be revised in order

to better the quality of their written works For that reasons, students normally do not have to put in further effort to seek for the correctness, which is directly in contrast with the case of indirect WCF that will be later discussed However, while addressing and correcting all sorts

of errors in students’ writing is still a common practice in most language classrooms, whether including a large number of corrections at the same time with this direct method is too overwhelming for any improvement to be made remains a matter of concern

Indirect WCF, on the other hand, contains no correction but an indication of errors committed According to Bitchener and Knoch (2009), the practice can be carried out in one

of the following ways, “underlining or circling an error; recording in the margin the number of errors in a given line; or using a code to show where an error has occurred and what type of error it is” (p.198) They continue “rather than the teacher providing an explicit correction, students are left to resolve and correct the problem that has been drawn to their attention.” (p.198) Although the description above of indirect WCF is detailed and helps provide a visual imagination about how indirect WCF is operated as a form of feedback to written work, Bitchener and Knock’s mention of the absence of “explicit correction” in indirect WCF is

of their work needs to be improved to enhance the accuracy of the whole paper

Without involving correct forms of errors, indirect CF encourages leaners to manage to fix their errors themselves by calling for their attention to the problematic piece of language According to Lalande (1982), such type of WCF opens up opportunities to foster guided-learning and problem solving which are extremely beneficial in terms of language learning He suggests that students’ foreign language writing abilities will be enhanced substantially by strategies that encourages guided-learning techniques That also means Lalande highly evaluates the merits of indirect WCF because this strategy is very much a direction of correction rather than the provision of an access to the available correction that students may

‘abuse’ Besides, the problem-solving ability gives students a chance to reconstruct their grammatical structures with an intention to make them better From a practical standpoint, if all students can involve in this process, teachers are able to set their mind at rest because it will

be easier for students to make progress and study with a greater level of independence and autonomy

More importantly, what it means by direct or indirect WCF should also be concerned

in the relation with metalinguistic features As stated in Ellis (2008), metalinguistic comments can be made in one of the two ways, the use of error codes or the provision of metalinguistic

25

explanation regarding the nature of errors Error codes usually take the form of abbreviated labels to represent different types of error, and they can be put right over the location of an error or at the margin of the text While the former is a more popular approach, both of them still help provide students with a clue for the correction Regarding metalinguistic explanations, they are given in the form of written comments including brief grammar description with regard

to the sort of error students commit in their writing pieces While such form of feedback can

be informative in a way that it helps students fully aware of the nature and root causes of their errors, it is less likely to be employed popularly in the reality of language classrooms Obviously, making detailed comments in the form of writing can be really time-consuming, and how exhausting the activity of giving feedback of a teacher can be imaginable when they have many students with many errors persisting in their written work; not to mention the fact that it can also be hard work as teachers have to have sufficient knowledge for accurate explanations for various error types While metalinguistic feedback is separated from direct and indirect WCF as an independent type with distinctive features (Ellis, 2008), the following taxonomy of WCF strategies by Sheen and Ellis (2011) is more convincing:

 Use of error code (e.g., symbols such as VT to signal a verb tense error or WO a word order error are inserted into the text)

 Use of brief grammatical explanation (e.g., error types are numbered in the text and then a brief explanation of each type of error is provided at the end of the text)

 Errors are indicated and located but the correct form is not supplied (e.g., an error is underlined in the place in the text in which it occurs)

According to the taxonomy above, direct and indirect WCF are the two big branches and metalinguistic information is seen as following features With metalinguistic information, direct WCF provides both correct forms and grammatical description which are absent when metalinguistic information is not provided Indirect WCF can be accompanied by either error codes or a brief grammatical explanation at the end of a text followed by the numbers provided

in the location of errors when metalinguistic information comes into play, otherwise errors are either indicated only or located and indicated at the same time when metalinguistic information

is not available However, whether metalinguistic information is added or not, direct and indirect WCF still maintain its very characters To be more specific, direct WCF is associated with the provision of correct forms of errors why it is not the case of indirect one where correction is just in in the form of clues or signals

27

2.3.2.3 The effectiveness of WCF types

The difference in effectiveness of direct and indirect WCF have been put forth Those who support direct CF base their arguments on the fact that it is beneficial for learners who possess limited language proficiency and have to struggle with the correction themselves (Ellis, 2008; Ferris and Roberts, 2001) Therefore, it can be seen that direct WCF can be suitable for those who are incapable of coming up with the correct forms of their errors as this strategy is highly straightforward and does not require students to manage for correction As can be seen, such availability of corrections can be advantageous in the way that less time and effort from students will be put into comprehending what goes wrong in their writing pieces

At the same time, the merits of indirect WCF have also been stated widely Of the many forms of teachers’ written feedback, Hyland (2003) mentions the “minimal marking” (p 181) which he described as “in- text, form-based feedback” and “indicating the location and perhaps type of error” It can be seen that this form of feedback is identical in every way with the indirect WCF that has been discussed so far Hyland believes that this minimal marking when accompanied with error codes is advantageous in “stimulating a student response” and

“developing self-editing strategies” (p.181) It is evident that students with this strategy of correction have to manage to revise their written work without relying on teacher inputs of correct forms This is very much the same as how students solve their problems on their own, and what they learn from this process will benefit further writing activities

To be in line with that, Harmer (2007) also supports the use of error codes rather than

a full correction of errors He believes that when students can identify their errors from those codes, they are in the position of correcting them, and when they can make changes to their writing and resolve the mistakes, the process of providing feedback yields a positive outcome This view suggests that indirect WCF in general and with error codes in particular fosters independent learning which is beneficial for the writing ability That emphasizes the merits of using errors codes as they provide useful clues for students to spot an error

Researches into the effectiveness of various types of WCF have paid most attention to the two main strands, the studies comparing direct WCF of different types with its combinations and the ones comparing direct and indirect WCF For the former, Bitchener et al., (2005) examined the roles of direct CF and direct CF plus oral metalinguistic in the form

of conference on the accuracy improvement of 53 adult students over a 12-week period Basing

28

on the result of the study, they conclude that the combinations of CF yield more positive effects

on the precision enhancement of past tense and articles use than single type of WCF alone However, as pointed out by Truscott (2007), the result of such a study is not justifiable as there

is a considerable imbalance in instruction hours allocated for the experimental group and control group that receive no correction

However, other studies on the same strand do not produce uniform results In an extended investigation by Bitchener and Knock (2009), no difference is perceived between the effects of direct CF plus oral-written metalinguistic, direct CF with written metalinguistic explanation and direct error correction only, which is not consistent with the result of the study above Afterwards, Sheen (2007) investigates the differential effects of direct and written metalinguistic CF on intermediate ESL learners An important part of the study findings showcases the outperformance of direct metalinguistic CF in comparison with direct CF only, but this only became evident two months later in her delayed post-test Thus, for most of the cases, it seems that the combination of direct WCF with metalinguistic information helps bring students to a greater level of accuracy in terms of grammar Such a positive sign can be explained by the fact that when students’ errors are elaborated either in written forms or made verbally, they have an opportunity to learn about their mistakes and understand the root causes

of those mistakes more clearly

The second branch, however, can be seen as the main source of controversy and is also the researcher’s area of interest In fact, a number of studies have aimed at a comparison of effects exerted by direct and indirect CF and the differential effectiveness each of them brings, but it is interesting to note that these studies have ended up with inconsistency

The first study to mention is the one by Chandler (2003) The study aimed at examining the effects of direct and indirect CF on 36 students who were required to write five chapters of autobiography Any change in four aspects, including the fluency measured by the time it takes

to write a text, the writing quality measured by teachers’ holistic ratings for different types of errors, student attitudes towards each type of marking and the time students need for the revisions of their errors was measured Among the four, the second aspects can be seen as closely related to the measurement of accuracy improvement of students’ texts after their errors are treated by different methods The CF strategies used in the study were divided into two big groups, teacher correction and teacher marking for student correction The former was very