The effects of negotiated feedback in the writing conference onl2 error correction and l2 uptake

VIETNAM NATIONAL UNIVERSITY, HANOI UNIVERSITY OF LANGUAGES AND INTERNATIONAL STUDIES FACULTY OF ENGLISH LANGUAGE TEACHER EDUCATION GRADUATION THESIS THE EFFECTS OF NEGOTIATED FEEDBAC

VIETNAM NATIONAL UNIVERSITY, HANOI

UNIVERSITY OF LANGUAGES AND INTERNATIONAL

STUDIES

FACULTY OF ENGLISH LANGUAGE TEACHER EDUCATION

GRADUATION THESIS

THE EFFECTS OF NEGOTIATED FEEDBACK

IN THE WRITING CONFERENCE ON

L2 ERROR CORRECTION AND L2 UPTAKE

Supervisor: Dr Nguyễn Chí Đức Student: Phạm Nguyễn Quỳnh Anh Course: QH2016.F1.E1

HÀ NỘI – 2020

ĐẠI HỌC QUỐC GIA HÀ NỘI

TRƯỜNG ĐẠI HỌC NGOẠI NGỮ

KHOA SƯ PHẠM TIẾNG ANH

KHÓA LUẬN TỐT NGHIỆP

THẢO LUẬN LỖI TRONG BÀI VIẾT

CỦA NGƯỜI HỌC NGOẠI NGỮ VÀ TÁC DỤNG CỦA NHỮNG CUỘC THẢO LUẬN NÀY ĐỐI VỚI VIỆC CHỮA LỖI VÀ HỌC NGOẠI NGỮ

Giáo viên hướng dẫn: Dr Nguyễn Chí Đức Sinh viên: Phạm Nguyễn Quỳnh Anh

Khóa: QH2016.F1.E1

HÀ NỘI – 2020

I hereby state that I: (Student’s name, class), being a candidate for the degree of Bachelor of Arts (programme) accept the requirements of the College relating

to the retention and use of Bachelor’s Graduation Paper deposited in the library

In terms of these conditions, I agree that the origin of my paper deposited in the library should be accessible for the purposes of study and research, in accordance with the normal conditions established by the librarian for the care, loan or reproduction of the paper

Signature

Date

ACKNOWLEDGEMENTS

This paper would not have been possible without the help and guidance of Mr Nguyen Chi Duc, who provided me with clear instructions, critical comments, and prompt support I am profusely grateful for his constructive critique and devoted supervision, without which I would not have been able to complete this thesis successfully

I would also want to extend my gratitude to all the participants, who dedicated their time and efforts this research

In addition, I cannot disregard the help from my beloved ones, who motivated and provided me with unlimited mental support

Last but not least, I thank myself for always trying to move forward

ABSTRACT

In the world of L2 learning and teaching, feedback has always been a debated issue, with different researchers putting forward different claims about its effectiveness (Karim & Nassaji, 2019; Kang & Han, 2015; Liu & Brown, 2015) The conflicting results give rise to the question that perhaps it is the way feedback is provided and how students are engaged in the process that ultimately determines the efficacy of feedback This case study investigated the effects of negotiated feedback in writing conferences on students’ error correction rate and second language uptake Specifically, the study compared the quantity and quality of negotiated feedback in Teacher-to-Student writing conferences and Peer-to-Peer writing conferences to see how students’ engagement differed in these two types of conferences and whether the effects were more pronounced

much-in the case of Teacher-to-Student conferences By means of recorded writmuch-ing conferences, writing drafts, and post-tests of five intermediate/upper-intermediate learners, the results of this study revealed that negotiated feedback had a more prominent effect on students’ error correction rate and L2 uptake in the suggest conference than in the typical Peer-to-Peer one The disparity could

be attributed to several significant factors, including the use of scaffolding in sample analysis, the accuracy of the input provided, the students’ understanding

of the marked errors, and the provision of feedforward The findings of this research can be meaningful to teachers and program developers in increasing students’ engagement with writing feedback, choosing the suitable type of writing conferences, as well as helping students improve their future performance Furthermore, the study set a foundation for further larger-scaled studies on how different feedback negotiations would affect error correction and L2 uptake

TABLE OF CONTENTS

ACKNOWLEDGEMENTS 4

ABSTRACT 5

TABLE OF CONTENTS 6

LIST OF TABLES AND FIGURES 8

INTRODUCTION 9

1 Rationale 9

2 Research aim and overall research design 10

3 Expected contributions 10

4 Organization of this report 11

Chapter 1 13

LITERATURE REVIEW 13

1 Key concepts 13

1.1 Error vs mistake 13

1.2 Feedback vs feedforward 15

1.3 Writing conference 16

1.4 Negotiation episodes 17

1.5 Second language uptake: declarative vs procedural knowledge 18

2 Theoretical frameworks 19

2.1 Output Hypothesis 19

2.2 Interaction Hypothesis 20

2.3 Socio-cultural Theory 21

2.4 The backbone of this study 24

3 Empirical research into the effects of feedback, negotiated feedback, and writing conference on error correction and L2 uptake 25

3.1 Empirical studies into the effects of feedback on error correction and L2 uptake 25

3.2 Empirical studies into the effects of feedback engagement on error correction and L2 uptake 26

3.3 Empirical studies into the effects of writing conferences on error correction and L2 uptake 28

4 Research gaps 30

Chapter 2: 31

METHODOLOGY 31

1 Research Questions 31

2 Research site 31

3 Sampling 32

4 Research tools 32

4.1 Writing conferences 32

4.2 Writing drafts 36

4.3 Learner-specific tests 37

5 Data collection 39

6 Data analysis 40

6.1 Research question 01 40

6.2 Research question 02 41

6.3 Research question 03 42

Chapter 3 43

FINDINGS 43

1 Research question 01 43

2 Research question 02 48

Chapter 4 51

DISCUSSION & PEDAGOGICAL IMPLICATIONS 51

1 Discussion 51

2 Pedagogical implications 56

CONCLUSION 58

1 Summary of the findings 58

2 Limitations of the study 59

3 Suggestions for further research 59

BIBLIOGRAPHY 61

APPENDICES 70

LIST OF TABLES AND FIGURES

Table 1 ERROR CODING SCHEME 37 Table 2 POST-TEST MARKING SCHEME 38 Table 3 NEGOTIATION PARTICIPATION OF STUDENT N IN THE

SUGGESTED TEACHER-TO-STUDENT WRITING CONFERENCE 43 Table 4 NEGOTIATION PARTICIPATION OF STUDENT P IN THE

SUGGESTED TEACHER-TO-STUDENT WRITING CONFERENCE 44 Table 5 NEGOTIATION PARTICIPATION OF STUDENT T IN THE PEER-TO-PEER WRITING CONFERENCE 44 Table 6 NEGOTIATION PARTICIPATION OF STUDENT T IN THE PEER-TO-PEER WRITING CONFERENCE 45 Table 7 NEGOTIATION PARTICIPATION OF STUDENT Q IN THE

PEER-TO-PEER WRITING CONFERENCE 46 Table 8 STUDENT N’S SUCCESS RATE OF ERROR CORRECTION 48 Table 9 STUDENT P’S SUCCESS RATE OF ERROR CORRECTION 48 Table 10 STUDENT T’S SUCCESS RATE OF ERROR CORRECTION 49 Table 11 STUDENT L’S SUCCESS RATE OF ERROR CORRECTION 49 Table 12 STUDENT Q’S SUCCESS RATE OF ERROR CORRECTION 49

INTRODUCTION

1 Rationale

In foreign/second language (L2) writing instruction, giving written corrective feedback (henceforth referred to as WCF for short) to student writing is generally deemed to lie at the heart of formative assessment However, the question as to whether WCF is indeed conducive to learning has engendered a heated debate among both L2 writing researchers and instructors (please refer to Karim and Nassaji (2019) for a critical review and Kang and Han (2015) and Liu and Brown (2015) for a systematic review) This debate got to the climax when Truscott

(1996) put forward an idea that WCF was, in general, “ineffective” and even

“harmful” to both writing development as well as general L2 acquisition (p.327)

This standpoint immediately receives many critiques from various scholars far and wide (e.g., Ferris (1999); Bruton (2009); Chandler (2009))

Although Truscott’s view above is too strong a claim, he does draw our attention

to several limitations in the way that WCF is typically given in our traditional L2 writing classroom First of all, such feedback is often provided in a

unidirectional, but not a dialogic manner This consequently deprives L2

students of the opportunity to interact with the instructor to understand the feedback As this feedback is not fully understood, L2 students have a high tendency to copy the instructor’s correction into the new versions of their writing

verbatim without any mental engagement with the feedback Even if L2 students

themselves would like to further unpack this feedback, they rarely receive

adequate external support or scaffolding from their instructor in such a

self-initiative Therefore, the matter might not lie in whether we should continue to give WCF to student writing, but how WCF should be given

Given the limitations delineated above, the present study proposes a three-step writing conference in which L2 students first work with the instructor and fasten their focus on a particular erroneous language point, analyze a collection of samples of how this language point should be used in a standard way with a view

to figuring out the underlying patterns or rules, and then plan for their error correction as well as further study of this language point The ultimate goal of

this writing conference is to transform our traditional feedback sessions into learning affordances that L2 students can negotiate with their instructor to understand the given feedback, which, in turn, hopefully enables students to pick

up the underlying knowledge associated with this feedback

2 Research aim and overall research design

This study aims to examine the relative effects of the proposed three-step Teacher-to-Student writing conference on error correction and L2 uptake To this end, I shall compare the success rate of error correction as well as the amount of L2 uptake that the above writing conference brings about against that of a typical Peer-to-Peer writing conference To be more specific, the present study seeks answers to the following research questions:

(a) How are L2 students engaged in the suggested writing conference with their writing instructor and in the conventional one with their peers?

(b) What is the effect of this engagement on error correction? Is this effect more pronounced in the case of the suggested writing conference?

(c) What is the effect of this engagement on L2 uptake? Is this effect more pronounced in the case of the suggested writing conference?

Theoretically, this study is grounded on Swain’s Output Hypothesis (2005), Long’s Interaction Hypothesis (1996), and Vygotsky’s Socio-cultural Theory (1987) Unlike most of the previous studies in this area, the present research makes use of both negotiated feedback and feedforward as well as the notion of scaffolding to mentally engage L2 students in their error correction in the writing conference This study adopts a case-study method for its overall research design Research participants are five intermediate EFL students recruited from

a language learning center in Hanoi, the capital of Vietnam

3 Expected contributions

This study is meaningful in at least three different ways First, it is among the pioneers that incorporate both feedback and feedforward into the writing conference Second, it does not only examine the effects of the suggested writing

conference on L2 students’ immediate error correction (as in the case of most previous research in this area) but also their delayed L2 uptake/retention The latter helps me to answer the question as to whether the knowledge/skill L2 students obtain from the writing conference is retained over time and therefore, hopefully, enables them to avoid similar errors in their future performance The findings from this study could shed more light on the benefits that teacher/peer comments bring about for writing improvement and L2 uptake and thus provide more empirical support for Swain’s Output Hypothesis (2005), Long’s Interaction Hypothesis (1996) and Vygotsky’s Socio-cultural Theory (1987) It

is also expected that these findings would have useful implications for writing instructors and students both inside and outside of this teaching/learning context

4 Organization of this report

Apart from the Introduction and the Conclusion section, this research report includes four major chapters: (1) Literature review, (2) Methodology, (3) Findings, and (4) Discussion and Implications

Chapter 1 – Literature Review – has three main purposes First, it is to define the key concepts that will be used throughout this thesis Second, it describes the relevant theoretical frameworks that together form the backbone of the present study Finally, it reviews previous empirical research that examines the effects

of feedback, feedback engagement, and writing conferences on L2 writing improvement as well as L2 uptake with a view to highlighting research gaps this study aims to fulfill

Chapter 3 – Findings, as the name suggests, presents the core findings of this research project To be more specific, these findings include the number of errors that the students made, the number of negotiation episodes (regarding both the form and the meaning of the target language code) as well as the effects of these episodes on error correction and L2 uptake both in the suggested Teacher-to-Student and the conventional Peer-to-Peer writing conference

Chapter 4 – Discussion and Implications – first sums up the findings of this study

in accordance with the research questions Subsequently, these findings are

compared with those of the previous research in the same area Finally,

implications for instructional practices are also provided for L2 writing

instructors both inside and outside of this teaching/learning context

Chapter 1 LITERATURE REVIEW

The purpose of this chapter is threefold First, it is to define the key concepts that will be used throughout this research report Second, it disseminates the relevant theoretical frameworks that together form the backbone of the present study Finally, it briefly reviews previous empirical research that examines the effects

of the writing conference together with its accompanied components (e.g., teacher comment and scaffolding) on error correction and L2 uptake with a view

to highlighting research gaps this study aims to fulfil

1 Key concepts

1.1 Error vs mistake

As learners begin to study another language, they gradually build up their understanding of linguistic rules, and it is common to see erroneous language output during this period Therefore, many researchers have looked at these inaccuracies in expressions to determine the factors affecting learners' language acquisition process In a study conducted by Corder (1967), he distinguishes two popular terms for linguistic inaccuracies: mistakes and errors According to this scholar, learners make mistakes when they have already acquired prior knowledge of the language However, due to memory lapses or physical and psychological factors like exhaustion and extreme emotions, they are unable to recall the language rules at the time of production Mistakes are sometimes referred to as performance errors As a result, mistakes occur randomly, and can often be corrected by the learners themselves Errors (also referred to as competence error), on the other hand, refer to inaccuracies caused by learners' inadequate language competence and thus are much more consistent and structured Therefore, researchers can rely on these systematic language variations to determine the learners' linguistic ability

Overgeneralization is a demonstration of errors and how they differ from

mistakes For instance, “He standed there” as an incorrect form of the sentence

“He stood there” This may be attributable to the learner's unawareness of the

irregular verb “stand” and instead conjugates the verb according to the regular

past tense rule This type of problem may exist mainly in users of beginner-level

In contrast, when an advanced or native speaker of English says “He walk home every day” instead of “He walks home every day”, that could be an occasion when the person has made a mistake This grammatically incorrect expression could presumably be a “slip of the tongue error” from the speaker that is caused

by the individual's carelessness or other performance factors These mistakes can occur to those who are studying the language as well as proficient users

Cognitive linguists believe that teachers and learners should concentrate on dealing with errors, because they are essential to the process of language learning For those who follow the cognitive approach (i.e., an L2 learning theory that is based on cognitivist psychology), language acquisition mainly involves rule learning According to this approach, the ability to connect concepts, deduce patterns from correct examples, and make generalizations contributes to the formation of a mental representation of the target language (e.g., Ellis & Beaton, 1993)

Linguistic scientists following the cognitive approach place much emphasis on the distinguishing qualities of errors and mistakes to investigate learner's ability

to obtain L2 knowledge While researching L2 errors, Corder (1974) classifies them into three separate categories based on the learner's level of understanding

of the target language The first type, pre-systematic errors, appears when learners are unaware of the existence of a particular language rule and are random as users mainly rely on guessing the correct expression A learner may also demonstrate multiple variations of the same error on different occasions For instance, the learner can produce a grammatically incorrect sentence such as

“He isn't work” one time and a different inaccurate expression such as “He don't

work” another When learners make a false assumption of a rule in the language

system, they may encounter systematic errors As opposed to pre-systematic errors, these errors occur more regularly, and learners can repeat a particular error several times Post-systematic errors emerge when the learner understands the correct target language rule but uses it inconsistently

For researchers, these errors provide them with more in-depth insight into the language acquisition process However, according to Corder (1967), the benefits

of errors extend beyond research purposes The process of learning and teaching the target language can also benefit from the attention to errors Teachers can also benefit considerably from errors to determine how far students have progressed in the language course, and devise an appropriate plan for these students Finally, errors are essential for learners themselves, because they assist students in making progress in their learning process Through errors and teachers’ reactions to those errors, learners can dispel wrong hypotheses about the target language

In this paper, errors, conceptualized as an integral part of language learning, receives more focus than mistakes To be more specific, they are considered a visible proof that learning is taking place, which, in turn, provides the teacher with insightful information to choose an appropriate teaching strategy to help students correct the errors as well as pick up the underlying knowledge/skill associated with those errors

1.2 Feedback vs feedforward

Various terms have been used in the existing literature to label corrective

feedback Some of the most frequently used terms include “corrective

feedback”, “negative evidence” or merely “feedback” Chaudron (1977), Nassaji

and Kartchava (2017), for example, describes “feedback” as any reactions that

the teacher initiates to help their students to improve the quality of an oral utterance or a written sentence Long (1996), however, classifies the feedback

that the teacher provides their L2 learners with as “positive and negative

evidence” While the former praises them for any instances of their correct L2

use, the latter draws their attention towards erroneous ones In this light,

“corrective feedback” can be viewed as a type of “negative

evidence” Regardless of its different labels, corrective feedback, in nature,

refers to the information through which learners realize that their language output

is incorrect, thus evoking relevant changes in their inter-language system

Corrective feedback can be either oral or written This paper focuses on the written corrective feedback that the teacher often gives to student writing Apart from feedback, another key term that is used in the present study is feedforward While the former involves an evaluation of what L2 students can and cannot do against a set of learning goals, the latter is to suggest what they should do in the future to improve their performance of a language task at hand (Walker, 2013) In other words, feedback describes the outcome of their past learning On the other hand, feedforward is more associated with their future learning

In this research, students working with the teacher will be given feedback to revise their writings as well as feedforward to improve future performance

1.3 Writing conference

Writing conferences is a one-on-one strategy that takes place between the teacher and a particular student (i.e., Teacher-to-Student) or between a pair of students (i.e., Peer-to-Peer) Gillet and Beverly (2001) believe that the writing conference can be done one-on-one, in a pair, or even in a small group Likewise, Routman (2005) also confirms that writing conferences can be led either by the teacher or the student

Writing conferences are often carried out for different purposes They may be used to encourage, teach, and assess students or merely to set teaching/learning goals (Routman, 2005) Calkins (1986) suggests that writing conferences may help students with writing content, process, and evaluation Such a conference is probably the best opportunity for direct and immediate teaching of the complex processes and skills involved in writing These conferences can occur at any stage of the writing process They can foster the instruction on a wide range of writing skills, strategies, and concepts According to Graves (1994), the purpose

of writing conferences is to help children tell the teacher about what they know/do not know so that the teacher can help them more effectively with their writing Similarly, Lucy Calkins (1994) states that Teacher-to-Student and Peer-

to-Peer writing conferences are at the heart of writing teaching through which students learn to interact with their own writing

From language assessment perspectives, conference assessment usually involves the students’ appointed visits to the teacher's office in order to discuss a certain piece of work or learning process, or both (Brown & Hudson, 1998) The most notable difference between conferences and other forms of assessment is the direct focus on learning processes and strategies (Genesee & Upshur, 1996) For example, when (a) conference(s) is/are carried out to discuss multiple versions

of students’ writing work, students' perceptions and concerns about the learning processes can be concentrated on

The present research employs both Teacher-to-Student and Peer-to-Peer writing conferences as a platform for students to negotiate errors

1.4 Negotiation episodes

According to Nassaji (2011), negotiation is a process that takes place through the back and forth interactional strategies used to reach a solution to a problem in the course of communication Form negotiation and meaning negotiation are two kinds of negotiation Meaning negotiation is regarded as the side sequences to the actual conversation to solve communication problems, making what was being conveyed more comprehensible (Pica, 1994; Van den Branden, 1997) Form negotiation is produced when attention is given to form or the language codes and takes place when “one interlocutor tries to push the other towards producing a formally more correct and/or appropriate utterance” (Van den Branden, 1997, p 592) During conversations, both of these types of negotiation can lead to interactional feedback

Given the literature on negotiation and negotiation episodes as mentioned above, negotiated feedback can be considered the type of feedback provided through oral or written communication and is negotiated both in terms of meaning and form In this research, negotiated feedback occurring during oral interactions are studied In contrast, non-negotiated feedback takes place when the learners take

in the provided input from their teacher or peers without attempting to negotiate form or meaning

1.5 Second language uptake: declarative vs procedural knowledge

Schmidt (1993) proposes that for a new language code (e.g., a new vocabulary item or a new grammar point) to be learned, L2 learners need to direct a part or the entirety of their attention to it Ellis (1999) further expands this view by introducing the notion of focal and peripheral attention, which helps create a clear distinction between incidental and intentional learning In Ellis’ (1999) conceptualization, L2 learners’ focal attention is directly placed on a new language code when they learn it intentionally Some common classroom techniques that can exemplify this learning process in the case of vocabulary learning, for example, are flashcards and mnemonic methods (see Nation & Webb, 2011, pp.01-76, for a detailed account of different techniques for intentional vocabulary learning) It is evident from those techniques that L2 learners’ primary aim is to strengthen, in their mental lexicon, the form-meaning connection of the new word Therefore, their mental processing is entirely devoted to the form-meaning retrieval process as in the flashcard method or the creation of a mental image for the target word meaning as in the mnemonic method Meanwhile, in incidental learning, only peripheral attention is drawn towards the target language code since the focal one is concurrently occupied for text comprehension or production In other words, incidental learning is just a by-product of a meaning-focused activity (Nation, 2013) The learners’ primary aim in this learning process is to understand or compose the target text However,

to realize this aim, they might need the knowledge of the target language code; otherwise, their text comprehension or production can be hindered The result of such an incidental learning process is often referred to as L2 uptake In fact, this uptake can happen for both declarative and procedural knowledge

Declarative knowledge refers to the ability of L2 learners to understand and analyze a new language code, while procedural one is their ability to use this language code for communicative purposes In this study, declarative and

procedural knowledge will be measured respectively by the students’ ability to recognize or recall the language codes (previously negotiated) in a familiar or a novel context through their individual post-tests

2 Theoretical frameworks

Since this study aims to examine the effects of the suggested writing conference

on L2 writing improvement and L2 uptake, there are at least three relevant theories in the field of second language learning, which include Swain’s Output Hypothesis (2005), Long’s Interaction Hypothesis (Long, 1996) and Vygotsky’s Socio-cultural Theory (1987) In what follows, I shall describe the core tenets of these theories and justify why these theories can altogether form the backbone

of the present research project

2.1 Output Hypothesis

As an output task, the writing practice in the present study might foster L2 learning in accordance with the four main tenets of Swain’s Output Hypothesis

(1985, 1995, 2000, 2005) First, it promotes “noticing” (Swain, 1995, p.125)

When students are required to produce a piece of writing, they may notice gaps and holes in their lexical and grammatical resources and subsequently direct their attention to relevant lexis and syntax as they are exposed to L2 input further down the path of their learning process Second, this output task also creates the opportunity for them to experiment with using a newly met word and structure

in a particular context and therefore helps them to consolidate their knowledge

of this lexical or structural item – a benefit that Swain refers to as the

hypothesis-testing function of output tasks The third tenet of this hypothesis, which has

perhaps received less attention in the research literature, is that output

enhances fluency of L2 use That is because using a recently encountered

word/structure in an output activity fosters memorization of this lexical/grammatical item, which, in turn, increases the degree of automaticity in recognizing it from a given input text and also retrieving it from memory for communicating ideas Finally, when students are engaged in the writing

conference, they can talk about a particular word or structure Such a

meta-linguistic talk is often found to benefit their internalization of such a word and

structure into their interlanguage system

2.2 Interaction Hypothesis

The Interaction Hypothesis (Long, 1996) has four major constructs, which are input, interaction, feedback, and output The present study is concerned with feedback and interaction

The Interaction Hypothesis maintains that feedback resulting from the negotiation of form/meaning is essential for second language uptake Such a negotiation episode is triggered when a breakdown in communication occurs in conversation leading to a modified utterance either from the L2 students or their interlocutor Either the students need to change what is said (i.e., modified output) to be understood by the native speaker or the native speaker needs to simplify speech (i.e., modified input) to be understood by these non-native

speakers Both modified input and modified output may lead to noticing gaps in

the students’ interlanguage system Such noticing may improve the chances of subsequent acquisition of the noticed form White (1987) argues that it is the comprehensible input which promotes second language acquisition because such input creates opportunities for a language student to notices the gap between his/her current state of interlanguage and the target language Further, Swain’s Output Hypothesis (1985, 1995, 2000, 2005) suggests that L2 input itself is not sufficient for language acquisition Second language output may trigger specific cognitive processes for second language learning Negotiation of meaning is able

to help this process, because by becoming consciously aware of one’s production, the output can serve the metalinguistic function of helping to internalize linguistic form, as already mentioned above In Long’s interaction hypothesis (1996), he argues that interactive input is more important than non-interactive input Interpreting what Long (1996) claims, Gass and Mackey (2007) indicate that through interaction, students’ attention is directed to problematic features of knowledge It may lead students to pay attention to

something new such as a new vocabulary item or a grammar point, which promotes the second language uptake

In other words, interaction adds more values to output as L2 students can receive the correct sample of L2 use via a negotiated input (i.e., feedback) Thanks to this feedback, they can modify their output to render it more comprehensible to their interlocutors

2.3 Socio-cultural Theory

Sociocultural Theory (henceforth referred to as SCT for short) originated from the work of Vygotsky (1978) Vygotsky theorizes that human mental activities are, in nature, the evolutionary capability to consciously and intentionally control

thinking by using higher-level cultural tools such as language, literacy,

numeracy, categorization, rationality, and logic These tools enable advanced mental activities of humans, such as decision-making and problem-solving (Lantolf & Thorne, 2006, p 198)

From the SCT perspectives (Vygotsky, 1978, 1987), all human features and functioning are combined effects of natural biological endowment and our interaction with historical, social, and cultural factors from the world (Yaroshevsky & Gurgenidze, 1997, as cited in Lantolf & Pavlenko, 2001, p 352), and that it is more of our sociocultural history than biology that provides

us with these features/functioning (Lantolf and Pavlenko, 2001, p 143) Accordingly, human features like language ability are dynamically subject to change during our interaction with the world In other words, our students’ L2 competence might not be fixed traits and can be socio-culturally co-constructed via their interaction with their peers and us

SCT views learning as both a cognitive and social activity It emphasizes that human mental activities are mediated, especially by a powerful cultural artifact that is language Social interaction is a crucial aspect of learning, providing opportunities for the process of regulation and internalization, which, in turn, facilitate cognitive development Learning occurs in the zone of proximal development (hereafter as ZPD), with the scaffolding of others In what follows,

I discuss three key concepts of SCT – ZPD, social interaction, and scaffolding –

in greater detail

Zone of proximal development

For Vygotsky, learning emerges through interaction within the ZPD The ZPD

is “the distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem-solving under adult guidance or in collaboration with more capable peers” (Vygotsky, 1978, p.86) The ZPD used to be considered a fixed trait of the learner; however, it has now evolved into an emergent and open-ended one that unfolds through interaction and increases the learning potential by creating unanticipated opportunities (Wells, 1998) In the context language learning, the notion of the ZPD creates a multi-faceted picture that encompasses the elements of the language learning situation, namely “the teacher, the learner, their social and cultural history, their goals and motives, as well as the resources available to them, including those that are dialogically constructed together” (Aljaafreh & Lantolf, 1994, p.468) In brief, in a Vygotskian view, learning occurs chiefly within the learner’s ZPD

Social interaction

Social interaction is another key construct of SCT and refers to the process of using language to communicate to share experience and construct knowledge Social interaction from a sociocultural standpoint is the communication between teachers and students, or between students and peers, who are together engaged

in the learning process

SCT proposes that learning happens within social and cultural practices, and learning is a co-constructed activity, where students develop understanding while communicating about their experience According to McNeil (2012), social interaction enables students to develop understanding by doing a task together, receiving support from others, and later completing the task independently In other words, the cognitive and linguistic development of an individual can be achieved through social interaction with assistance from others Therefore, learning is constructed and occurs in social interaction

Walqui (2006) notes that “Social interaction is the basis of learning and development Learning is a process of apprenticeship and internalization in which skills and knowledge are transformed from the social into the cognitive plane” (p 160) Working in language-learning situations (including foreign and second language classroom contexts), Gibbons (2007) claims that language acquisition within SCT is a joint activity between students, and language learning and language use are situated in social and cultural situations The role

of interaction is crucial for second language learning because it provides students opportunities to use and comprehend the target language In conclusion, learning originates in interaction, and interaction creates an environment for learning language and learning through language (Wells, 1999)

Scaffolding

Another notion crucial to a Vygotskian framework is the notion of guided support provided to the less knowledgeable partner (the novice) as they collaborate with a more knowledgeable partner (the expert) This idea, which is known as scaffolding, refers to a “situation where a knowledgeable participant can create supportive conditions in which the novice can participate, and extend his or her current skills and knowledge to higher levels of competence” (Donato,

1994, p.40) This notion implies that scaffolding and support provided through Teacher-to-Student and Peer-to-Peer interactions are necessary to learners’ completion of complicated and advanced tasks while learning a second language Scaffolding, which is based on learner’s needs, should be differentiated from helping the learner in a unidirectional way (McCarthy & McMahon, 1992) Here, support is generated as a joint effort and social interaction between the novice and the expert working within the learner’s ZPD (Donato, 1994) From this perspective, the type of feedback does not determine the effectiveness of corrective feedback It is the way that feedback evolves in interaction and the way feedback is negotiated between the novice and the expert

Aljaafreh and Lantolf (1994) gathered data on three learners of English who received corrective feedback collaboratively and within the zone of proximal development The findings showed that the nature of the transaction and

mediation provided by the expert is a determining factor in the usefulness of corrective feedback Results also revealed that every type of error treatment was effective as long as it was negotiated between the learner and the teacher and was given at the right time or within the learner’s ZPD

Aljaafreh and Lantolf (1994) named some features of effective scaffolding in an L2 context Scaffolding should be offered only when needed and removed at signs of students’ independent functioning; graduated when learners’ proficiency increases; and must be provided in a dialogic conversation actively involving both teacher and student Williams (2004) suggests that teachers can scaffold interaction and the learner’s movement towards autonomy by marking critical features in the text, simplifying the task, and “modeling” indirectly by eliciting

an improved performance from the writer

2.4 The backbone of this study

As already mentioned, this study aims to investigate the effects of the suggested writing conference on error correction and L2 uptake When L2 students are required to produce a piece of writing, they are involved in a pushed output task

As a result, they can enjoy four potential benefits of an output task that Swain (2005) suggests in her Output Hypothesis After the writing phase, these students participate in either a Peer-to-Peer or a Teacher-to-Student writing conference where they discuss a particular error to correct this language deviation In such interaction, especially negotiation episodes (if any), they can also enjoy the benefits that Long (1996) describes as inherent features of negotiated interaction

in his Interaction Hypothesis Thanks to such interaction, L2 students can also

be able to move out from their Current Developmental Level to their Zone of Proximal Development In this very case, language (including both L1 and L2)

is used as a potent mediator to draw students’ attention to their L2 writing errors (noticing), provide them with further L2 input (comprehensible input), scaffold them with input analysis to draw out language rules or patterns (language-focused learning), guide them to modify their output (error correction or modified output) and plan their future actions to internalize the above rules

and/or patterns (feedforward) Put differently, all three theories described above together can work as the backbone of the present research project

3 Empirical research into the effects of feedback, negotiated feedback, and writing conference on error correction and L2 uptake

3.1 Empirical studies into the effects of feedback on error correction and L2 uptake

The role of feedback in the language classroom has been researched for over two decades There exists a vast amount of research on the effects of feedback on second language uptake However, there are still clashes in the existing literature Many researchers supported the necessity of feedback in second language writing as well as on second language uptake Ferris (1999, 2004), for example, stated that most students in his study needed and wanted to be corrected by their teachers; therefore, error correction could not be excluded from L2 writing classrooms Chandler (2003) conducted a study to examine the relative effects

of the four kinds of feedback on L2 writing improvement, especially regarding the accuracy aspect He found out that all four types of feedback led to a significantly improved accuracy in L2 students’ later writing Likewise, Bitchener, Young, and Cameron (2005) examined the effects of feedback and writing conferences on L2 students’ development of three English aspects, including preposition, simple past, and definite article They found that the combination of feedback and individual conference improved students’ accurate usage of the past simple and definite article to a great extent and of preposition

to a noticeable level Additionally, Leki (2007) investigated the usefulness of written feedback provided by faculty staff to L2 graduate students in a US university by analyzing teacher comments and interviewing students about the feedback usefulness Most students reported that they found the feedback very useful Many even said they would like to receive more feedback as such, especially in terms of content and technical writing skills Li’s (2010) meta-analysis has further confirmed the effectiveness of feedback for L2 learning To

be more specific, by meta-analyzing a collection of 33 empirical studies on the

effects of feedback on L2 writing performance, he found an overall medium effect size for feedback on L2 writing improvement It is also noteworthy that such an effect size appeared to maintain over time

However, some researchers argued that L2 feedback was of no or, at least, little use to writing development Semke (1984), for example, carried out a 10-week study in which students were divided into four groups and received four different treatments, including direct error correction, coded feedback (but with self-correction), comments on content only, and both direct error correction and comments on content As a result of the study, she detected no significant difference in the writing outcomes between the three groups, which received error correction of one form or another and the remaining group who only received comments on the writing content She concludes that it might be the quantity of the writing rather than that of feedback session that matters Similarly, Kepner (1991) found out that his/her L2 participants made almost no improvement in their writing before and after receiving direct feedback on grammatical errors After reviewing relevant research in this area, Truscott (1996) concludes that feedback does not have benefits for either L2 writing improvement or L2 uptake

Considering the review above, it is clear that the effects of feedback on L2 writing improvement, as well as L2 uptake, were yet to be conclusive That is why more empirical research in this area is still welcomed

3.2 Empirical studies into the effects of feedback engagement on error correction and L2 uptake

As feedback itself is often found not to be conducive to L2 learning, people start

to direct their attention towards feedback engagement However, very few studies until now have looked at the relationship between feedback engagement and L2 writing improvement or L2 uptake

Han and Hyland (2015), for example, examined the engagement of four average college students with corrective feedback through a qualitative study They found that students’ beliefs and learning experiences with the interactional

context of receiving and processing feedback could lead to differing degrees and manners in their engagement In another case study, Zhang (2017) investigated one university student’s engagement with computer-generated corrective feedback given to her writing He found out that although the participant’s overall behavioral pattern was unclear, her cognitive engagement, which was positive, and her emotional engagement was affected by the writing scores In addition, the emotional responses like surprise, happiness, dissatisfaction, and frustration could also affect her understanding and uptake of corrective feedback Zheng and Yu (2018) conducted another case study of 12 Chinese lower-proficiency students, which aimed to explore their engagement with teacher-written corrective feedback in the EFL writing class Research data were collected from multiple sources, including drafts of student essays, teacher-written feedback, student immediate oral reports, and semi-structured interviews The study found that while the participants’ affective engagement was relatively positive, their behavioral and cognitive engagement was not extensive in the sense that their behavioral engagement did not necessarily result

in greater language accuracy There was slight awareness at the level of understanding the written corrective feedback, especially for the indirect written corrective feedback

These studies together show that the nature of feedback engagement is, in essence, far more complex than we used to think The quality of such an engagement depends not only on cognitive but also on affective domains of L2 students It is also apparent that the mentioned studies merely focus on feedback, but not comment as a fusion of both feedback and feedforward In addition, according to SCT, no matter how much L2 students are engaged with feedback, they might not be able to move out from their current development level to the zone of proximal development without the support of their teacher or peers Thus, writing conferences might be more fertile land to study the effects of feedback or comment on L2 writing improvement and L2 uptake In what follows, critical studies in this area will be reviewed

3.3 Empirical studies into the effects of writing conferences on error correction and L2 uptake

In 2005, Bitchener, Young, and Cameron conducted a study on the effect of different types of corrective feedback on ESL student writing The study investigated whether the kind of feedback (direct, explicit written feedback and student–researcher 5-minute individual conferences; direct, explicit written feedback only; no corrective feedback) given to 53 post-intermediate ESOL (migrant) students on three types of errors (prepositions, past simple, and definite article) resulted in improved accuracy in new pieces of writing over 12 weeks The participants were placed into three treatment groups Group one studied full-time (20 hours per week) in a post-intermediate class Another attended a part-time post intermediate class for 10 hours per week The other enrolled in a part-time post-intermediate class for 4 hours per week Among these three groups, group one received direct written corrective feedback and a 5-minute student–researcher conference after each writing Group two was provided direct WCF only Group three did not have corrective feedback on the targeted features but were given feedback on the quality and organization of their content for ethical purposes The study found a noticeable effect of the combination of written and conference feedback on accuracy in using the past simple aspect and the definite article in the new writing pieces

In 2013, Guetzlaff carried out a research project on the effects of Peer-to-Peer writing conferences on writing revisions in a second-grade classroom This study acquired both qualitative and quantitative data The research compared the effects of student training on Peer-to-Peer writing conferences on their L2 writing performance The results denoted that the students who were trained on how to carry out Peer-to-Peer writing conferences were more likely to make statements that clarified their writing or thinking They also tended to ask their partners more questions than those who were not trained at all Overall, the former also outperformed in their post-treatment writing task than the latter Bayraktar (2013) investigated the nature of interactions during the Teacher-to-Student Writing Conference and revisiting the Potential Effects of Self-Efficacy

Beliefs The purpose of this study was three-fold First, the nature of Student writing conferences was examined to determine if they were either balanced, more student-centered, or more teacher-centered The second goal was

Teacher-to-to investigate whether students’ levels of perceived self-efficacy could influence the nature of their writing conferences Third, this study hoped to highlight the prevalent interaction patterns taking place during the conferences with students with low and high levels of perceived self-efficacy towards writing This study employed a qualitative design, observing and analyzing the Teacher-to-Student writing conferences over ten weeks The participants were students in fifth grades from a public primary school in the Southeastern United States The researcher adopted the Writing Self-Efficacy Scale (Pajares, Miller, & Johnson, 1999) as adapted from Shell, Murphy, and Bruning (1989), as well as audio and video-taped Teacher-to-Student writing conferences, audio-taped interviews with the teacher and students, and observations to collect data Gathered evidence was described and interpreted using qualitative methods The research found that none of the teacher-to-student writing conferences was entirely teacher-centered The classroom teacher was skillful at conducting conferences, which possessed a balance of teacher-centered and student-centered features Also, the nature of writing conferences changed among students with different self-efficacy levels in terms of focus, ownership, conference schedule, turn-taking, speech frequency, quantity and functions of the questions asked, numbers

of compliments given by the teacher, and the number of external interruptions occurred during conferences

From the review of prior empirical studies into the effects of writing conferences

on L2 writing improvement and L2 uptake above, it has been confirmed that negotiation episodes in writing conferences bring about both L2 writing improvement and L2 uptake However, these studies often focus on the effects

of only teacher-to-student writing conferences or those of peer-to-peer writing There has rarely been any comparison of the effects of these two types of conferences

4 Research gaps

The review above has highlighted several research gaps First, there has been a large number of empirical studies on the effects of feedback on L2 writing improvement and/or L2 uptake; however, their findings are rather inconclusive Several studies found that feedback was beneficial for L2 learning Others, however, found the exact opposite Therefore, more studies in this area are still

welcomed Second, most of the previous research centers on feedback per se, but

not students’ cognitive engagement with feedback This has led the researcher to believe that feedback is not inherently good or bad It is feedback engagement that matters In addition, previous studies often conceptualize feedback as the teacher’s or peers’ response to L2 students’ writing performance without any endeavors to help L2 students to deal with the same errors in the future In other words, they merely focus on feedback, but not feedforward Again, it is the latter that is more conducive to L2 learning The next research gap is associated with

a comparison of the effects on L2 writing improvement and L2 uptake between the suggested Teacher-to-Student and the conventional Peer-to-Peer writing conference In the existing literature, such a study remains scarce Finally, there

is also still a paucity of research into the effects of writing conferences in the context of Vietnam, especially with immediate EFL students in an unofficial learning context like a foreign language learning center, as in the case of the present study Thus, this research project aims to fill all of the above gaps

Chapter 2:

METHODOLOGY

In this chapter, the author will first specify the research questions that this research project aims to answer Subsequently, key characteristics of the research participants, especially those related to their L2 profile, are presented Followed

up are justifications of the overall research design and the research methods that this study has adopted This section ends with a detailed description the procedure of data collection and analysis

1 Research Questions

First, this study aimed to examine how L2 students engage with the teacher’s written comment on their writing performance in the suggested Teacher-to-Student and the conventional Peer-to-Peer writing conference Second, it was to investigate whether negotiation episodes (if any) in these conferences led to successful error correction and L2 uptake Finally, this study compared whether the Peer-to-Peer or the Teacher-to-Student conference was more conducive to learning In other words, the present research project sought answers to the following research questions:

(a) How are L2 students engaged in the suggested writing conference with their writing instructor and in the conventional one with their peers?

(b) What is the effect of this engagement on error correction? Is this effect more pronounced in the case of the suggested writing conference?

(c) What is the effect of this engagement on L2 uptake? Is this effect more pronounced in the case of the suggested writing conference?

2 Research site

This research took place in a language center in Hanoi during the first half of

2020 Speakers of English a foreign language often had to attend lessons to improve their English proficiency, which usually involved writing practice To investigate the effects of negotiated feedback on learners’ error correction and language uptake, the researcher needed to organize Teacher-to-Student and Peer-

to-Peer writing conferences to examine how the leaners were involved in the process of negotiation and how that would affect their error correction rate for subsequent writing drafts The researcher also customized tests for each student

to examine whether negotiated feedback could transfer to long-term language uptake Given the limited time and resources, a language center in Hanoi was chosen to carry out this study owing to the available participants This research site provided the researcher of this project with convenience and feasibility

3 Sampling

The research involved five participants of equivalent English proficiency (approximately B1 to B2 level) Participants were divided into two groups One group consisting of two students took part in the suggested Teacher-to-Student writing conference, while the remaining three students took part in the typical Peer-to-Peer writing conference Participants were enrolled in an English course

at an English center in Hanoi, and their English proficiency was determined based on the center’s placement test

Given the limited time and resources, the researcher decided to carry out the study at an English center in Hanoi for convenience sampling, in which the researcher chose available and accessible individuals to act as participants (Cohen et al., 2007) The five participants above were chosen with a random sampling strategy from a class of 25 students with the same level of proficiency

4 Research tools

This research was conducted as a case study Qualitative data were collected through the writing conferences, both Teacher-to-Student and Peer-to-Peer and quantitative data were collected through the participants’ writing drafts and test papers

4.1 Writing conferences

To gather data on how the students negotiated with different partners, the researcher employed two types of writing conferences One was the suggested

Teacher-to-Student writing conference, and the other was the typical Peer writing conference

Peer-to-There were two reasons why these two types of writing conferences were selected Firstly, Goldstein and Conrad (1990) found that only students succeeding in meaning negotiation in the writing conference were capable of making extensive and quality revisions This finding was born out by Williams (2004), who found better employment of teacher’s advice regarding revisions when there were explicit suggestions from the teacher and active participation as well as negotiation at the conference However, some L2 learners encountered cultural or social inhibitions about engaging in informal conversations with teachers, seen as figures of authority This meant that these students might not pose questions to teachers during the process of meaning negotiation and could submissively follow the teacher’s suggestion without thorough reflection In contrast, peer feedback was noted to provide students with a greater sense of control and autonomy, encouraging active involvement in the feedback process

by Mendoca and Johnson (1994) Furthermore, social and affective support can also be given to students when they interact orally with peers (Guerrero & Villamil, 1994) From a Vygotskian perspective, social interaction with peers was deemed essential as the Vygotskian theoretical construct of the Zone of Proximal Development (1978) proposes that writing skills could emerge with the mediation and help of others Hence, the participants were divided into two groups as mentioned above to tackle the existence of this sense of social hierarchy and to examine the differences in negotiation episodes when students work with their teacher versus when working with their peers, along with impacts

of these episodes on immediate correction rate and on long-term language uptake

Secondly, the quality and delivery of feedback were taken into consideration Some researchers (Leki, 1990; Nelson & Murphy, 1992; Lockhart & Ng, 1993; Mendoca & Johnson, 1994) found that students had problems detecting errors and providing quality feedback, sometimes using to formulaic comments on each other’s writing, or they may give inappropriate and over-critical feedback

(Amores 1997) or overfocus on surface errors (McGroarty & Zhu, 1997) Within the Vygotskian framework (1978), the guided support given to the novice (here, the students) during collaboration with the expert (here, the teacher), or scaffolding, was a critical element of feedback effectiveness This notion implied that learners need to be scaffolded and supported in their complex task of learning a second language as they interact with the teacher or peers From the Vygotskian perspectives, the effectiveness of corrective feedback did not rely much on the type of feedback, but on its evolution in interaction and the way it was negotiated between the novice and the expert (Donato, 1994) Notably, scaffolding was a joint process constructed based on the learner’s needs and Zone of Proximal Development While teachers were generally more informed about this process and possess higher expertise, students did not have an adequate background to provide their peers with quality scaffolded help (Donato, 1994) By dividing the participants into two groups as described, the researcher aimed to examine the effects of the quality of feedback as well as the effects of how it was provided on immediate correction rate and student’s language acquisition

The suggested teacher-to-student conference

In this study, the researcher adapted the following procedure of scaffolding for the suggested Teacher-to-Student conference Aljaafreh and Lantolf (1994) put forward a regulatory scale for giving feedback

Figure 1 REGULATORY SCALE (INDIRECT TO DIRECT) (Aljaafreh & Lantolf, 1994)

According to this scale, the teacher first pointed the sentence with the error and later narrowed down the location of the error if the students could not identify it After that, the teacher provided hints about the type of error If the students were still unable to correct the error, then the teacher offered cues about how to amend

it Finally, the teacher provided the correct form and gave an explanation as well

as examples of the right pattern

Though inspired by the guidelines devised by Aljaafreh and Lantolf (1994), the scaffolding steps used in this study were modified Specifically, the teacher in this study followed the guidelines below for each error

1 The teacher marks the errors made in the first draft with error codes

2 The teacher asks the student to correct the error independently

3 The teacher rejects unsuccessful attempts to self-correct

4 The teacher provides a number of good samples and asks the student to deduce the correct rules or patterns themselves

5 The teacher rejects unsuccessful attempts to correct

6 The teacher provides further cues to help the student arrive at the correct form

7 The teacher rejects unsuccessful attempts to self-correct

8 The teacher provides the correct form

9 The teacher provides an explanation for the correct form

This adaptation was made based on the assumption that if students were mentally engaged in the process of realizing the standardized knowledge rather than passively receiving it, it might be more beneficial to their long-term uptake of language

Additionally, the teacher also asked the students to devise a plan explaining how they intended to address the errors and gave them further at-home practice after receiving the feedforward in the conference

The conventional Peer-to-Peer conference

Regarding the typical Peer-to-Peer conference, students discussed the errors marked implicitly by the teacher one by one Students were encouraged to refer

to any resources to foster their error correction

4.2 Writing drafts

To gather data on students’ error correction rate, the researcher relied on their writing drafts All essays were on the same topic and answered using the same ideas after brainstorming with the teacher The 1st drafts were written in class and later, commented by the teacher with implicit error codes The errors were divided into two main types, local and global (Burt & Kiparsky, 1974) These authors defined local errors as erroneous forms that did not impede communication and comprehension of the intended meaning Local errors encompassed noun and verb inflections, as well as the use of articles, prepositions, and auxiliaries On the other hand, global errors were more severe

as they interfered with communication and caused disruptions in meaning Inaccurate word order or inappropriate word choice in a sentence could be some examples of global errors

Students’ errors were marked based on the coding scheme below

Table 1 ERROR CODING SCHEME

Type Code Meaning

W O Word order

WC Word choice [ ] Idea organization E? Elaboration

IR Irrelevant Local

? Unclear meaning

Awkward expression Deletion

The 2nd drafts were edited versions of the 1st drafts after the writing conference

4.3 Learner-specific tests

To gather data on students’ L2 uptake, the researcher designed individualized tests for each student Given the lack of time and feasibility, the tests only covered rule-governed language problems (local errors) in the students’ essays The number of questions in each test depended on the number of errors each student made To gauge students’ L2 uptake for each type of error, the researcher designed a series of 4 questions targeting each error following the order below

- The first question was a novel sentence with a gap targeting the error the students made Students were not given any options and were asked to fill in the

blank This question measured the students’ ability to recall the new language code in a novel sentence If the students succeeded, they could move on to another error Otherwise, they would continue with question 2 in this series

- The second question was taken directly from the students’ writing with the target error removed Students were not given any options and were asked to fill

in the blank This question measured the students’ ability to recall language codes in the same sentence as in their original writing If the students succeeded

in this question, they could move on to another error Otherwise, they would continue with question 3 in this series

- The third question was a novel sentence with a gap targeting the error the students made Students had three options to choose from and fill in the blank This question tests students’ ability to recognize language codes in a new sentence If the students succeeded in this question, they could move on to another error Otherwise, they would continue with question 4 in this series

- The fourth question was taken directly from the students’ writing with the target error removed Students had three options to choose from and fill in the blank This question measured the students’ ability to recognize language codes in the same sentence as in their original writing

Post-test marking scheme

For each series of 04 questions targeting one error, the teacher gave each student

a score from 0 to 4 based on their performance Level 3 and 4 refer to their procedural knowledge, while Level 1 and 2 manifests their declarative knowledge only (see section 1.5 chapter 1 for further explanation)

Table 2 POST-TEST MARKING SCHEME