Nhận thức của học viên tại trung tâm anh ngữ talky về những yếu tố ảnh hưởng tới việc nghe hiểu tiếng anh

Answer to the first research question: What are the most common factors affecting the students‟ listening comprehension during the two listening tasks?. The primary purposes of the disse

z

HANOI UNIVERSITY DEPARTMENT OF POST-GRADUATE STUDIES

STUDENTS’ PERCEPTIONS OF THE FACTORS AFFECTING

THEIR ENGLISH LANGUAGE LISTENING

COMPREHENSION

AT TALKY ENGLISH SCHOOL

Nhận thức của học viên tại Trung tâm Anh ngữ Talky

về những yếu tố ảnh hưởng tới việc nghe hiểu tiếng Anh

A dissertation submitted to the Department of Post-graduate Studies

in partial fulfillment of the requirements for the Master‟s degree

Student: Bui Quang Huy Supervisor: Dr Dinh Thi Bao Huong

Hanoi 2019

Table of Contents

ACKOWLEDGEMENTS i

STATEMENT OF AUTHORSHIP ii

ABSTRACT iii

LIST OF ABBREVIATIONS iv

LIST OF TABLES AND FIGURES v

CHAPTER 1: INTRODUCTION 1

1.1 Statement of the problem 1

1.2 Purpose of the study 2

1.3 Research questions 2

1.4 Significance of the study 3

1.5 Scope of the study 3

1.6 Organization of the dissertation 4

CHAPTER 2: LITERATURE REVIEW 5

2.1 Definition of listening 5

2.2 Listening comprehension processes 6

2.2.1 Top-down processing 6

2.2.2 Bottom-up processing 9

2.2.3 Combination of top-down and bottom-up processing 10

2.3 Factors influencing listening success 13

2.3.1 Cognitive factors 14

2.3.2 Affective factors 20

2.3.3 Contextual factors 23

2.4 Previous studies 25

2.5 Summary 30

CHAPTER 3: METHODOLOGY 31

3.1 Research design 31

3.2 Research setting 32

3.3 Participants 32

3.4 Data collection instruments 33

3.4.1 Justification for the use of the questionnaire and interview 33

3.4.2 Questionnaire 34

3.4.3 Interview 35

3.5 Procedure 36

3.6 Data analysis 39

3.7 Ethics 39

3.8 Summary 40

CHAPTER 4: RESULTS AND DISCUSSION 41

4.1 Answer to the first research question: What are the most common factors affecting the students‟ listening comprehension during the two listening tasks? 41

4.2 Answer to the second research question: Is there any relationship between the listening task types and the factors experienced? 48

4.3 Discussion 50

CHAPTER 5: RECOMMENDATIONS AND CONCLUSION 53

5.1 Recommendations 53

5.2 Limitations 54

5.3 Directions for future research 55

5.4 Conclusion 56

References 57

APPENDIX A: QUESTIONNAIRE (ENGLISH) 67

APPENDIX B: QUESTIONNAIRE (VIETNAMESE) 69

APPENDIX C: INTERVIEW GUIDE (ENGLISH) 71

APPENDIX D: INTERVIEW GUIDE (VIETNAMESE) 72

APPENDIX E: SAMPLE TRANSCRIPT 73

APPENDIX F: CODING TABLE 79

ACKOWLEDGEMENTS

Thanks to everyone who helped me to complete this M.A program and, ultimately, this dissertation Firstly, I would like to extend my grateful thanks to Dr Dinh Thi Bao Huong for her tremendous support and guidance throughout my coursework, study, and writing of this dissertation I am also grateful for the support of the librarian staff at Hanoi University who have assisted me in searching for coursework- and research-related material, as well as supported me in the processing of statistical data Thanks in general to the Department of Postgraduate Studies for their enthusiasm and practical assistance throughout the two years I have been a student at Hanoi University

I would also like to express my thanks to my friends and family for encouraging me to pursue this area of interest and providing me with valuable comments and suggestions Thanks to my students for their willingness to take part in this study and for being guinea pigs during some stages of my study Last but not least, thanks to the management at Talky English School for making it possible for me to carry out this research

STATEMENT OF AUTHORSHIP

I certify that except where due acknowledgement has been made, the work is that of the author solely; and the work has not been submitted in support of another degree or qualification from this or other university or institute of learning

The study in this thesis has been approved by Hanoi University

Hanoi, November 2019

Bui Quang Huy

ABSTRACT

This dissertation presents an exploratory approach to investigate the factors expected to have

an impact on listening skills of B1-level students at English Talky School The primary purposes of the dissertation were to explore the factors influencing learner listening comprehension with greatest frequency as well as the relationship between the listening task types and the factors experienced

A mixed-methods concurrent design was employed in the study Quantitative data were collected from 33 students at B1 level according to the Common European Framework of Reference for Languages (CEFR) Qualitative data were acquired from one-on-one interviews with four participants

Many of the factors reported were related to text, listener, task and environment characteristics The factors experienced by the students related to the gap-filling task type, in order of frequency of mention, were memory, attention and concentration, time available for processing, speech rate, vocabulary, physical conditions, speaker's clarity, culturally specific vocabulary and idioms, physical and psychological states, knowledge of grammar, concern about understanding every word, opportunities for second hearing and directness of text With regards to the factors influencing the students‟ listening in the other task type, in order of frequency of mention, attention and concentration, opportunities for second hearing, speech rate, memory, physical conditions, vocabulary and time available for processing were reported to be highly influential The findings also revealed that the number of factors with the highest impact was almost twice as many in the gap-filling task type as that in the multiple-choice task type Furthermore, most of the factors cited in the gap-filling task type were related to characteristics of the text and those of the listener (five each), whereas the number of factors reported in the multiple-choice task type, which were connected with the text and the listener, was rather low (two each) Last but not least, the overlap between the factors reported in the multiple-choice task type and those in the gap-filling task type signifies that the seven factors which are attention and concentration, opportunities for second hearing, speech rate, memory, physical conditions, vocabulary and time available for processing are equally important in both of the listening tasks

The results of this study are mixed in terms of corroborating and disproving the assertions in previous literature Notwithstanding differences, the findings from this paper carry a number

of pedagogical implications and recommendations for those in the teaching profession, material developers and future researchers who are interested replicating this work

PET: Preliminary English Test

TES: Talky English School

LIST OF TABLES AND FIGURES

Table 1: Factors reported in the two listening task types 48

Table 2: Percentages of students reporting factors experienced in two task types 49

Table 3: Groups of factors experienced in two task types 50

Figure 1: Most common factors experienced in pilot multiple-choice task 38

Figure 2: Most common factors experienced in multiple-choice task 41

Figure 3: Most common factors experienced in gap-filling task 42

CHAPTER 1: INTRODUCTION

1.1 Statement of the problem

The importance of listening comprehension is increasing the world over, and the need for global communication has made listening by non-native speakers even more crucial According to Rost (2002), of the four language skills being listening, reading, writing and speaking, listening skills are considered a “primary means of L2 acquisition” (p 103) Unfortunately, this „Cinderella‟ skill is often overlooked in many ESL or EFL programs (Mendelsohn, 1994; Nunan, 1997; Vandergrift, 1997) Given the role of listening in everyday life, such neglect is surprising; Regarding the time an individual is engaged in communication, 45% is devoted to listening, while speaking accounts for only 30% (as cited

in Hedge, 2001, p.228) Considering the growing importance of listening, it seems logical that language learners should be equipped with an insight into listening factors which learners usually experience in order to help them to avoid experiencing the same negative factors as well as to promote the positive ones in future listening tasks

Although the practice of teaching English to second language (L2) learners has changed considerably over the last few decades, learners still find it challenging in different language tasks L2 learners continue to be faced with a plethora of problems inside and outside the classroom as they try to improve their listening comprehension skills (Vandergrift & Goh, 2012) In Vietnam, it is disappointing to see that language learners have devoted a great deal

of their time to practicing their listening skills by listening to recordings of various kinds Unfortunately, it seems that no return has been achieved as students are still in the dark about the key factors which contribute to their listening success and, to complicate matters further,

no explicit factors are available for the students to reflect on, based on which current pedagogical practices can be redirected in the way of helping them to overcome listening problems associated with certain factors Also, empirical findings from the existing body of L2 listening research, though informative, have done little to assist the researcher in generalizing those findings to his current research context

Regrettably, within the current research context, that is Talky English School, despite the generous offer of various learning resources, the scores obtained from recent listening tasks have shown that while some gained high marks, others received low ones This leads to the researcher believing that the differences in the test results lie in the factors they experience in particular listening task types Taking this as a starting point, a study into factors affecting

listening comprehension is therefore necessary not only to examine what lies behind students‟ listening success but also to see what implications they have for the practice of teaching English at the school

In sum, regardless of the existing research investigating the factors affecting listening comprehension, there is still a need for empirical work generalizable to the researcher‟s specific context Dunkel (1991) called for teachers and researchers “to increase vastly the number of empirical studies that investigate the ways in which factors inside and outside the L2 head affect comprehension of L2 discourse for beginning-, intermediate- and advanced-level L2 listeners” (p 445), while Vandergrift and Goh (2012) maintained that “more qualitative research methodologies such as interviews or stimulated recalls” (p 76) are required for a deeper understanding of relationship between certain factors and listening success This present study addressed those concerns by investigating factors influencing listening comprehension and any possible relationship between those factors and listening comprehension in specific listening tasks

1.2 Purpose of the study

This research aimed to determine the most common factors affecting second language listening comprehension in an EFL context, and the possible relationship between the factors experienced and the listening tasks

with necessary skills and exam strategies for a future IELTS test once they have completed all General English courses, it makes sense to conduct a study into these two listening task types, paving the way for the research results being generalized to students doing future IELTS courses For these reasons, only two listening tasks being multiple choice and gap-fill were dealt with throughout this paper

1.4 Significance of the study

On knowing the factors furthering and obstructing students‟ listening comprehension, those involved in the teaching of listening skills can make some changes to their pedagogical practices, aiming to boost their students‟ listening skills The research also adds to the existing body of knowledge since it looks at not only the negative aspects of the factors but their positive ones also

By taking part in the study, the students can have a clear idea of what factors facilitate and debilitate their listening comprehension Also, those involved in the interviews can be exposed to listening strategies (suggested by the interviewer in charge after each probing session) which they have not been exposed to before, resulting in them dealing with different listening tasks in the future more effectively Moreover, ways of dealing with specific listening tasks, which appear to have long been misused, will be rectified for the participants‟ sake once they agree to get involved in in-depth interviews offered by the researcher

1.5 Scope of the study

The researcher is a teacher of English at TES The current research was conducted in two months, including the piloting phase The research could not last longer, otherwise the syllabus might have been affected Also, the researcher presumed that there should not be too much of a skill being taught at a time, so the time should be spent on all four language skills reasonably Therefore, only two months was spent on the study A total of 33 learners of General English with a CEFR level of B1 were involved in this study in 2018 Only 33 took part in the research because other students at the time were not at the same level of proficiency or were on a different syllabus, which could affect the validity of the research The research findings are therefore confined to this population only and cannot be generalized

to a wider population with a different background, level of proficiency and within a different context Furthermore, only four participants, according to certain criteria, were purposively chosen and then contacted for further probing The interviewees were selected according to

four criteria: willingness to attend the interview, availability, experience of factors with the greatest frequency and possibility of giving relevant and reliable information Each interview lasted for around 20-30 minutes and each was conducted by the researcher There were two main interview questions and they all delved into the two research questions The first question aimed to invite the students‟ reflection on the factors which affected their listening comprehension the most, while the second one sought to triangulate the data collected from the questionnaire as well as their responses to the first interview question Also, the main questions were developed in the form of two general statements which were then followed by

a sequence of sub-questions for further probing Last but not least, due to the narrow scope of this study, simple research tools were used, which were the questionnaire and the interview

In sum, the thesis only deals with the most common factors affecting the students‟ listening comprehension in the two listening tasks being multiple choice and gap-fill, and the possible relationship between the listening task types and the factors reported

1.6 Organization of the dissertation

This dissertation is divided into five chapters Chapter 1 deals with the purpose of the study and the research objectives Chapter 2 reviews relevant literature in respect of the definitions

of listening, the factors affecting listening comprehension and the processes involved in listening comprehension Chapter 3 details the research tools, the data collection procedures and how the data were analyzed Chapter 4 presents the results of the research along with a discussion of the findings The last chapter – Chapter 5 discusses limitations and implications

of the study, based on which suggestions and recommendations for future research are made

CHAPTER 2: LITERATURE REVIEW

This chapter is a literature review that elaborates on the definition of listening, the factors affecting listening comprehension, the processes of listening comprehension and the relevant literature regarding listening comprehension within an EFL context

2.1 Definition of listening

In spite of the fact that the literature on listening has offered a wide range of definitions of listening, mainstream researchers (i.e., Buck, 2001; Lynch & Mendelsohn, 2002; O‟Malley et al., 1989; Richards et al., 1992; Vandergrift, 1999) all concurred that listening is as an active process

According to Anderson and Lynch (1988), arguing what successful listening really is,

“understanding is not something that happens because of what a speaker says: the listener has

a crucial part to play in the process, by activating various types of knowledge, and by applying what he knows to what he hears and trying to understand what the speaker means” (p 6) Underwood (1989) simplified the definition of listening as “the activity of paying attention to and trying to get meaning from something we hear” (p 1) For Mendelsohn (1994) listening comprehension means the ability to understand the spoken language of native speakers O‟Malley, Chamot, and Kupper (1989) offered a useful and more extensive definition of listening as “an active and conscious process in which the listener constructs meaning by using cues from contextual information and from existing knowledge, while relying upon multiple strategic resources to fulfill the task requirement” (p 19)

Mendelsohn (1994) also posited that, in listening to spoken language, the ability to decode the speaker‟s intention is required of a competent listener, as well as other abilities such as processing the linguistic forms such as speech speed and fillers, handling listening in an interaction, understanding the whole message contained in the discourse, comprehending the message without understanding every word, and recognizing different genres Listeners are also required to know how to process and how to judge what the illocutionary force of an utterance is, that is to say, what this string of sounds means in a specific context and under a particular set of circumstances

According to a definition provided by Purdy (1997), listening is “the active and dynamic process of attending, perceiving, interpreting, remembering, and responding to the expressed (verbal and non-verbal), needs, concerns, and information offered by other human beings” (p

8) Rost (2002) defined listening as a process of receiving what the speaker actually says (receptive), constructing and representing meaning (constructive), negotiating meaning with the speaker and responding (collaborative), and creating meaning through involvement, imagination and empathy (transformative)

To sum up, it is evident that individuals and researchers seem to define listening in terms of their personal or theoretical interests Despite this, the definition offered by O‟Malley, Chamot, and Kupper (1989) is adopted in this study since it appears to be comprehensive and overlaps some of the other definitions, as well as involves „task requirement‟ which is intimately related to one of the two research questions in this paper Also, the definition of listening and that of listening comprehension are interchangeable in this paper in order to avoid unnecessary confusion when the focus is more on the process of understanding English

as a foreign language

2.2 Listening comprehension processes

After concurring on a working definition of listening, it makes sense to delve into how the listening process takes place There are two types of listening processes being involved in the interpretation of spoken texts They are frequently referred to as top-down and bottom-up processing

2.2.1 Top-down processing

Top-down processing is defined as the use of background knowledge as the basis for understanding spoken discourse (Richards, 2008) This is concerned with an attempt to visualize such a „big‟ picture as an overview of the structure of the whole text This way of listening is sometimes equated with „gist listening‟ or „extensive listening‟ (Scrivener, 2011) Speaking in favor of this, Nunan (2002) considered the activation of context and situation as the core of the listening practice:

In this reconstruction process, the listener uses prior knowledge of the context and situation within which the listening takes place to make sense of what he or she hears Context and situation include such things as knowledge of the topic at hand, the speaker or speakers, and their relationship to the situation, as well as to each other and prior events (p 239)

How top-down processing is made use of can be exemplified by this extract:

Margaret

John

Margaret

John

And what other subjects did you take?

I got Maths, Physics and Engineering Drawing And English too But I only just passed that I took French but I failed it I‟m not much good at languages But you can take the National Certificate if you‟ve got four O levels so I was all right

And where are you working?

In a local firm They make parts for the motor industry, you know, crankshafts, crankshafts, connecting rods and so on Reynolds Supply Company They‟re very good They started me off in the model shop

(as cited in Hedge, 2000, p 232)

In this example, while a native a speaker of English has no problem in interpreting the spoken text, a second language learner may find it challenging at the mention of „the National Certificate‟ since a knowledge of this association should be required to understand its profile What is more, a highly competent learner of English, who has a rich vocabulary, may well know what „crankshaft‟ or „connecting rod‟ means, but he is expected to be furnished with even more specialized terms such as „model shop‟ in order to know why a young engineer should work in this place for some time (Hedge, 2000)

Listeners using top-down processing deduce meaning by picking up contextual clues from the discourse and then working out how those clues are connected with their prior knowledge of some kind, which is also termed „schematic knowledge‟ The clues to meaning may include knowledge of the setting, the topic and purpose of the discourse or that of what has been mentioned before Besides, it is noteworthy that misunderstanding may arise even when interlocutors of the same tongue communicate, which may be ascribed to cultural conflict A typical example of this is when the notion of „evening‟ signifies „after office hours‟ in one culture, but means „from three-thirty onwards‟ in another (ibid., pp 232-233)

According to Hedge (2000), schemata is divided into two sub-types being formal schemata and content schemata The former is concerned with the overall structure of events For instance, „once upon a time‟ heralds a certain kind of story that follows certain consequences, ranging from a description of characters, an event, an outcome, or perhaps a moral comment Some other event such as a wedding, a lecture or a church service follows the consequences

of their own, and thus, knowing how they are going to be like provides listeners with a general picture of the events and prepares them for what is going to be said up next The latter, on the other hand, uses a particular knowledge, which may be general, sociocultural and topical in nature, to infer meaning By way of illustration, when a speaker says this to his interlocutor, “I am going to the dentist this afternoon,” the utterance activates a schema for

„going to the dentist‟, including a setting (i.e., the dentist‟s office), participants (i.e., the dentist, the patient), goals (i.e., to have a medical checkup or to replace a filling), procedures (i.e., injections, drilling, rinsing) and outcomes (i.e., having the problem solved or discomfort) Therefore, if an utterance is produced later on the same day by his interlocutor, which might go like “So how was it?”, the speaker can simply reply, “Fine I didn‟t feel a thing.” Since both of the speakers share the same schemata of „going to the dentist‟, the details of the visit do not need to be repeated and not much is provided to enable both of them understand what happened (Richards, 2008)

As a way of teaching top-down processing, pre-listening activities that involve activating prior knowledge, making predictions, and reviewing key vocabulary should be taken into account (Richards, 2008) As Underwood (1989) put it:

It is unfair to plunge students straight into the listening text, even when testing rather than teaching listening comprehension, as this makes it extremely difficult for them to use the natural listening skills (which we all use in our native language) of matching what they hear with what they expect to hear and using their previous knowledge to make sense of it (p 30)

Richards (2008) also recommended some activities that can be used to develop students‟ down processing, such as:

top-● Students generate a set of questions they expect to hear about a topic, then listen to see

if they are answered

● Students generate a list of things they already know about a topic and things they would like to learn more about, then listen and compare

● Students read one speaker‟s part in a conversation, predict the other speaker‟s part, then listen and compare

● Students read a list of key points to be covered in a talk, then listen to see which ones are mentioned

● Students listen to part of a story, complete the story ending, then listen and compare endings

● Students read news headlines, guess what happened, then listen to the full news items and compare

2.2.2 Bottom-up processing

According to a definition by Nunan (2002), bottom-up processing means “a process of decoding the sounds that one hears in a linear fashion, from the smallest meaningful units (phonemes) to complete texts” (p 239) How bottom-up processing works is described by the author as follows:

Phonemic units are decoded and linked together to form words, words are linked together to form phrases, phrases are linked together to form utterances, and utterances are linked together to form complete, meaningful texts In other words, the process is a linear one, in which meaning itself is derived as the last step in the process (p 239) Sharing this view, Richards (2008) maintained that lexical resource and grammatical range are central to bottom-up processing This can be illustrated by this spoken text – “The guy I sat next to on the bus this morning on the way to work was telling me he runs a Thai restaurant in Chinatown Apparently, it‟s very popular at the moment.” By adopting bottom-

up processing, this spoken text is broken down into pieces, which provide listeners with the underlying core meaning of each utterance The pieces include „the guy‟, „I sat next to on the bus‟, „this morning‟, „was telling me‟, „he runs a Thai restaurant in Chinatown‟, „apparently, it‟s very popular‟ and „at the moment‟ The pieces are also referred to as „chunks‟, based on which listeners can work out the meaning of each utterance To take the above-mentioned example further, listeners can expand on each chunk to fully understand what each chunk means Some propositions that are expanded on can be „I was on the bus‟, „There was a guy next to me‟, „We talked‟, „He said he runs a Thai restaurant‟, „It‟s in Chinatown‟, and „It‟s very popular now‟ These propositions are what they remember and are not the form in which they first heard them It is the grammatical knowledge, the speaker‟s intonation and hesitation that assist listeners in finding the appropriate pieces

Pre-listening activities help listeners to develop not only top-down but also bottom-up processing through reviewing vocabulary or analyzing the features of spoken discourse (Richards, 2008) Underwood (1989) claimed that previewing the language which will be heard in the listening text is appropriate when the teacher, though unwilling to neglect the

content of the recording, wants to sensitize his students to the language forms in „authentic‟ situations, which his have recently studied Richards (2008) also offered a wide range of tasks that seek to develop listeners‟ bottom-up processing, which can be seen below

● Identify the referents of pronouns in an utterance

● Recognize the time reference of an utterance

● Distinguish between positive and negative statements

● Recognize the order in which words occurred in an utterance

● Identify sequence markers

● Identify key words that occurred in a spoken text

● Identify which modal verbs occurred in a spoken text

2.2.3 Combination of top-down and bottom-up processing

Brown (1990), Scrivener (2011), Hedge (2000), and Richards (2008) all concurred on the idea that both top-down and bottom-up processing take place at the same time Richards (2008) argued that the extent to which one dominates the other “depends on the listener‟s familiarity with the topic and content of a text, the density of information in a text, the text type, and the listener‟s purpose in listening” (p 10) How two types of cooks listen to the radio is a good advertisement for this An experienced cook can listen to a radio chef describing a recipe for a meal to compare it to his own Due to his prior knowledge of the meal, he might listen to discern the differences and similarities only A novice cook without a previous knowledge of the meal, on the other hand, might listen attentively to each step so as to write down the recipe Thus, it seems to hold water that the former requires more top-down processing, whereas the latter requires a larger amount of bottom-up processing

Field (1998) suggested that a typical listening lesson should involve a three-part sequence consisting of pre-listening, while-listening, and post-listening, which contains activities revolving around bottom-up and top-down listening (as cited in Richards, 2008, p 10)

Scrivener (2011), in favor of this idea, advanced a possible route map for a listening lesson in which listeners go from the „big‟ background and overview tasks (top-down processing) to the „smaller‟, detail-focused and language-focused issues (bottom-up processing), as seen below

Possible route map for a listening lesson from Scrivener (2011)

Discuss the general topic

Learners start to think about the topic, raising a number of issues that will be discussed later on the recording This preparation may help them to hear these things being discussed later

Predict the specific content Students hypothesise specific issues that may be

raised

Predict the structure

Students consider / discuss possible organisational structures for a phone-in (who speaks? what kind of questions? typical exchanges?) This may help learners to recognise the content more easily

Gist listening for overview

Learners get an overall impression of the content without worrying about small items or individual words

Gist listening for attitudes Learners interpret intonation, paralinguistic features

Listening to pick out specific

small language details

This focused work (eg on pronunciation) may raise learner awareness (eg of weak forms) and thus help students to listen better in future

Upon offering this modality, the researcher also maintained that this route map is flexible, which means that it is worth trying out if the teacher wants to do it the other way around, that

is, sensitizing his students to detail in the first place

Based on this route map, it certainly bears deducing that both top-down processing and bottom-up processing are closely linked with gist listening and listening for specific details, which are associated with multiple-choice and gap-filling listening task types The extent to which listeners are familiar with the L2 changes the way they use top-down and bottom-up processing While highly proficient listeners are able to interpret the speech signal using bottom-up processing (Tsui & Fullilove, 1998) and combine this bottom-up input with

background knowledge using top-down strategies to facilitate their listening (Field, 2004), their less proficient counterparts tend to make use of top-down input, that is to say background knowledge, to make up for their being unable to interpret speech sounds (Goh, 2000; Tyler, 2001) The results are also consistent with Wolff (1987), who observed that for listeners given a harder text to work on, they were prone to resort to top-down processing as a coping strategy A study by Mueller (1980) also took a closer look at L2 listeners‟ utilization

of top-down processing, suggesting that visual aids furthered the comprehension of less able listeners far more than they did that of their more able counterparts Another conclusion was reached by Long (1990), who hypothesized that non-native listeners might have used top-down input Although this was taken hypothetically, the evidence from her work strongly supported this theoretical assumption, revealing that background knowledge had a big part to play in interpreting spoken texts

Despite the combination of bottom-up and top-down processing as a „neutral‟ solution, Hedge (2000) challenged that even for first language listeners, comprehension “is always only selective and partial” (p 235) The reason why this is the case is that unlike computers, human beings, with limited memory, are not able to process information totally, not to mention the fact that the spoken text may be incomplete due to the distortion of radio signals

or the lack of explicitness in a speaker Brown and Yule (1983) added that comprehension is achieved when a listener interprets the speaker‟s intention reasonably (as cited in Hedge,

2000, p 235)

Hedge (2000) claimed that listening is not just about comprehension As such, many listening situations require response To respond correctly, listeners need to infer over and above what

is being intended, realizing the purposes and the outcomes of the discourse Speaking in favor

of this, Nunan (2002) held the view that when listening to a monolog, either live or through the media, listening is nonreciprocal, listeners have no chance to reply, to check their understanding or to check if they interpret what has been said correctly In the real world, it seems that no listener wants to take the role of nonreciprocal „eavesdropper‟ on a conversation By all this, it is meant that the ultimate aim of listening, regardless of top-down processing, bottom-up processing or combination of the two being used, is interaction, which seeks to make listening comprehension really „work‟ in interpersonal communication

While exploring top-down and bottom-up processing, Hedge (2000) mentioned an important facet of listening which is purposes for listening She said that teachers should make listening activities purposeful because different skills will be used for different purposes What really lies behind this is whether listening is done out of curiosity, for enjoyment, or for note-taking,

the skills involved will depend This was further clarified by Nunan (2002) when he pointed out that whether listeners listen for a general idea or for specific information, they still need different processes and strategies This opens up another worthwhile dimension of using top-down and bottom-up processing in English language classrooms which is concerned with designing listening tasks According to Nunan (2002), students should be trained to make use

of different listening strategies This can be achieved by holding the listening text constant (i.e., working with a news broadcast reporting a series of international events) and allowing learners to listen to the text several times, at which different instructions are given They might be asked to listen for gist first (i.e., the countries where the events have taken place) When they listen for the second time, they can be asked to match the places with a list of events Finally, they might be set the task of listening for detail (i.e., distinguishing between certain aspects of the event or comparing the differences between the accounts given by the radio and that by a newspaper)

In short, top-down processing enables listeners to predict the organization and content of spoken discourse, based on which they can make sense of its overall meaning On the other hand, bottom-up processing allows listeners to build up the message from the bottom, that is individual small linguistic pieces While the former seems more useful in dealing with tasks requiring a broad interpretation of a particular spoken text, the latter appears to work better in tasks requiring a narrower interpretation It also bears repeating that a combination of both listening processes are sometimes preferred as a neutral approach, allowing listeners to make used of the advantages of both processes

Since listeners tend to use top-down processing strategies such as inferencing and elaboration

to make sense of spoken texts, particularly when they do not recognize every word in the input (Goh, 1998a, 1998b) and since phonological and verbal cues from the input are used to attend to micro-features of text such as the form of individual words and grapheme/phoneme connections (bottom-up processing) (Carter & Nunan, 2001), it makes sense to argue that the former processing strategy has a role to play in listening success in the multiple-choice task and the latter strategy in the gap-filling one

2.3 Factors influencing listening success

According to Vandergrift and Goh (2012), cognitive factors include linguistic knowledge (knowledge of vocabulary and syntax), discourse knowledge, pragmatic knowledge, metacognitive knowledge, prior knowledge, first language listening ability, sound

discrimination ability and working memory capacity Affective factors incorporate factors such as anxiety, motivation, and self-efficacy Contextual factors include informal real-life listening outside the classroom, formal real-life listening in the classroom such as lectures, formal classroom listening practice, interactive listening, and listening assessment This section details studies into the cognitive factors claimed to have an impact on listening success, after which the affective factors which take a crucial role in listening success will be dealt with Lastly, an examination of some of the contextual factors impacting on listening comprehension will be offered

2.3.1 Cognitive factors

2.3.1.1 Knowledge of vocabulary

A considerable amount of literature has been published on the impact of cognitive factors on listening comprehension Lund‟s (1991a) study which involved learners of German revealed that readers at lower levels of proficiency were able to recollect greater detail than listeners who were able to recall more main ideas Without linguistic knowledge, listeners created plausible constructions as a compensatory strategy for filling in the details that they failed to understand or retain during listening Lund also maintained that the listening mode forces listeners to handle the listening task in a different fashion to make up for the transitory nature

of the spoken text

Another study conducted by Mecartty (2000) investigated the extent to which lexical resource and syntactic knowledge correlate with listening and reading comprehension of fourth semester learners of Spanish According to the data collected, lexical resource was cited as a strong predictor for both reading and listening, which explained approximately one fourth of L2 reading ability and 14% of L2 listening ability Based on the findings, she concluded that reading and listening comprehension processes may have some shared features, and that knowledge of L2 vocabulary may be of less importance in the comprehension process underlying L2 listening compared with reading, and identifying the factors which can explain the remaining variance in listening is also of great importance

Bonk (2000) also examined the correlation between knowledge of vocabulary and L2 listening comprehension Sharing Lund‟s (1991a) methodology, the participants in Bonk (2000) was required to listen to spoken texts and then demonstrate their comprehension by freely rewriting down what they could recollect out of their memory after listening to a spoken text The participants had to listen to four passages of increasing lexical complexity,

wrote recalls and then took dictation of the passages The results showed that high comprehension scores were closely related to good knowledge of vocabulary Additionally, the recall scores varied widely as the level of lexical difficulty rose Interestingly, some listeners achieved high comprehension scores even though they had knowledge of 75% or fewer of the vocabulary items used in a passage It was very possible that the listeners in question were able to make use of productive strategies like referencing to make up for what they failed to get Based on this, it can be concluded that better listening comprehension was closely correlated with higher dictation scores

Another study carried out by Staehr (2009) which investigated a larger population of Danish learners of English at advanced level revealed compelling evidence of the relationship between L2 lexical resource and listening comprehension The findings showed that over one half (51%) of listening variance could be justified by vocabulary, whereas 49% could be attributed to lexical resource Quality of knowledge regarding different aspects of a word and other words related to it, on the other hand, comprised a small percentage (2%) Also, an in-depth analysis of the data collected also demonstrated that 27 out of 56 participants who scored below the 5000-word level on the vocabulary scale were able to achieve a test score of 60% or higher This confirmed the findings in previous studies by Bonk (2000) and Graham, Santos, and Vanderplank (2010), who observed that the participants could deal with their linguistic weaknesses, probably basing their comprehension of what they did not understand

on what they already understood Graham et al (2010), however, added that there must be prerequisite linguistic recognition attained before listeners can use inferencing strategies to make up for their gaps in comprehension

2.3.1.2 Knowledge of syntax

Knowledge of grammar is considered to have great impact on successful L2 listening Mecartty (2000) examined the possible contribution of syntactic knowledge to L2 reading and listening comprehension Although the correlation between syntactic knowledge and reading and listening comprehension was not too close, multiple regression analyses revealed that knowledge of syntax did not always ensure L2 listening success Conrad (1985) hypothesized that listeners at beginner level would pay greater attention to syntactic cues while listeners with greater proficiency would devote their attention more to semantic cues

However, a recent study by Field (2008) published conflicting findings In the study, Field aimed to find out which types of words listeners depend on most, that is to say content words

or function words He required participants to write down the last four or five words they heard each time a recording was paused L2 listeners underperformed when compared to native language listeners, in terms of correctly identifying content and function words All L2 learners of English identified a larger percentage of content words One of the native language groups was able to identify all words correctly, regardless of whether or not they were content

or function words Field ascribed the findings to poor working memory as L2 listeners need to choose where to pay their attention In view of this, L2 listener will choose to pay attention to content words, based on their stress

2.3.1.3 Knowledge of discourse

Discourse knowledge, also known as script knowledge (Dunkel, 1986), refers to how the type

of information found in spoken texts is deemed, how the information is organized, and how it can be used to facilitate comprehension In one study, Jung (2003) concluded that listeners who benefited from discourse markers correctly recollected main ideas or low-level information units, which facilitated the construction of the meaning of the main ideas Jung also suggests that listeners may make the most of discourse markers when the structure of the spoken text is not clear, or when the listers are familiar with the passage type, or when they have background knowledge of the passage topic, or when the text is not planned

2.3.1.4 Pragmatic knowledge

According to Rose and Kasper (2001), pragmatic knowledge refers to the application of information related to a speaker‟s intention which goes beyond the literal meaning of an utterance Accordingly, listeners often work out a speaker‟s intention by making use of linguistic, cultural and contextual information

Cook and Liddicoat (2002) conducted a study on listener comprehension of request strategies

In the research, native speakers, high-proficiency, and low-proficiency listeners were asked to listened to excerpts associated with direct, indirect, and unconventional indirect questions Listeners‟ interpretations varied according to levels of proficiency While native speakers found it easy to interpret the questions, high-proficiency found it more difficult to understand the unconventional indirect questions; And understandably, the other group of listeners encountered difficulty in understanding both types of indirect questions Cook and Liddicoat then justified the difference by pointing to two factors being the processing demands of more

indirect information requiring processing of both linguistic and contextual information, and the poor working memory of listeners at a lower level of proficiency

Concurring on this, Garcia (2004) shared more or less the same conclusions in respect of the comprehension of conversational implicature and speech acts, which mean working out what

a speaker really means based on the indirect requests and understanding requests and corrections respectively The findings revealed that learners of English with greater proficiency outperformed those with lower proficiency on all measures assessing linguistic ability and pragmatic appropriacy What is more, Garcia also posited that the development of pragmatic knowledge did not necessarily go hand in hand with that of linguistic ability, concluding that L2 learners could benefit from a targeted focus on pragmatic comprehension

2.3.1.6 Prior knowledge

Prior knowledge refers to “what a learner already knows and which is available before a certain learning task, such as knowledge of vocabulary, syntax, the first language, or background knowledge about a topic or event” (Richards & Schmidt, 2013, p 457)

Long (1990) provided empirical evidence for the influential role of prior knowledge in L2 listening In the study, a large population of American university students of Spanish listened

to texts which were expected to have some shared features but topic In each case, the

participants were required to complete a background questionnaire on their knowledge of the topic and then listened to the text two times After second hearing, they were asked to summarize the content of the text based on what they had understood They then completed a checklist containing some statements of the text content (distractors were also provided), which were paraphrased in English For the second text, the same procedures were then repeated The results showed that the participants had much less prior knowledge of the topic

of the first text, compared to that of the second text and this impacted how much information could be recollected after listening As for the written summary of the text, Long observed an average of 53% for the first text and 68% for the second one Running contrary to this, the checklist showed that the mean score for the second text was 28% higher than for the first one Similar findings by Chiang and Dunkel (1992) were also consistent with this in terms of knowledge of religions of various kinds

Another important study by Tsui and Fullilove (1998) also deserves a mention as it took place

in a context of a popular standardized high-stakes examination The research involved responses by a large number of listeners to questions on listening comprehension texts Two types of listening texts were employed in the study – one being non-matching schema type texts where initial linguistic information did not fit in well with subsequent linguistic information, and the other being matching schema type texts where the initial linguistic information fitted in well with the subsequent information Dealing with these types of passages required the listeners to handle the input with care and to revise the initial schema if

a mismatch was noticed There were also two types of questions presented in the study being global-type questions which required overall comprehension and the ability to make conclusions or inferences, and local-type questions which required comprehension of particular details

Tsui and Fullilove concluded that proficient listeners did better than less proficient ones on both types of questions on non-matching schema-type passages This finding is understandable in the light of the flexibility of listeners noted in Vandergrift‟s (2003) work Less proficient listeners can still handle matching-schema type texts better if they activate their prior knowledge to make up for their gaps in comprehension The key role of prior knowledge in facilitating listening comprehension calls for the provision of a context for listeners Contextualization through pre-listening activities can serve as a predictor, and enable listeners to govern their attempts to understand

Previous studies such as Ginther (2002), Chung (2002) and Flowerdew and Miller (2005) have substantiated the effects of pre-listening activities on listening performance Listeners

who are provided with a context have the resources to activate their prior knowledge as well

as to make use of top-down processing to facilitate their listening comprehension This enables them to handle the input more effectively and to free up their working memory resources Tyler (2001) reported that when listeners could access the topic through an advance organizer, usage of L1 working memory and that of L2 were not different However, when listeners had no access to the advance organizer, usage of working memory was far higher among L2 listeners

The literature available on prior knowledge during listening has highlighted its importance in listening comprehension Also, activating prior knowledge matters when it comes to teaching adult learners as they already have a prior knowledge of various topics to activate, based on which their listening comprehension can be facilitated Furthermore, pre-listening activities should also constitute an integral part of any listening curriculum as they can prepare listeners for any listening tasks with some background knowledge, thus better listening performance

2.3.1.7 L1 listening ability

Although L2 listeners are already competent in their first language, the extent to which this contributes to L2 listening success remains to be investigated Vandergrift‟s (2006) study conducted on adolescent learners of French revealed that L1 listening ability and L2 proficiency together could justify around 39% of the common variance in L2 listening ability L2 proficiency explained around 25% while the figure for L1 was about 14% This was also observed by Hulstijn in a number of studies concerning L2 reading and writing Hulstijn (2011) noted that for Dutch learners of English, the correlation between L1 and L2 seemed to

be function of lexical resource and knowledge of grammar, processing speed, metacognitive knowledge and other general, language-independent skills

According to Bernhardt and Kamil (1995), how L1 listening contributes to L2 listening proficiency needs researching since one may be accidentally measuring L1 listening ability in his or her assessment of L2 listening, mistakenly calling it L2 listening ability This matters as L2 listeners may perform badly as a result of their poor performance in L1 listening

2.3.1.8 Sound Discrimination Ability

One justification for listeners‟ overall weakness in comprehension which may transfer from L1 to L2 is the ability to distinguish sounds There is some evidence that phonological

memory skills affect growth in listening ability and vocabulary learning, especially with children at a beginning level of language proficiency (French, 2003) Thus, there still remains

a research gap in the role of discrimination of sounds in L2 listening

In short, the findings from the above-mentioned studies emphasize the impact if cognitive factors on L2 listening comprehension While factors with great impact on L2 listening comprehension such as pragmatic knowledge and metacognition have been widely documented in the literature, the impact of some other factors such as L1 listening ability and sound discrimination ability on L2 listening still remain to be further investigated

2.3.2 Affective factors

According to Vandergrift and Goh (2012), L2 listening concerns more than paying attention

to linguistic input and understanding different cognitive requirements given to listeners As a matter of fact, listener ability to maximize their comprehension efforts can be affected by a number of affective factors These emotional listener-related characteristics will eventually determine how listeners respond to a listening task and affect their listening performance accordingly The three affective factors that have been widely researched within the context

of L2 listening are anxiety, motivation and self-efficacy

2.3.2.1 Anxiety

Mendelsohn (1994) noted that classroom practices associated with assessment contributed to

a high level of anxiety Pieces of research such as Horwitz (1986), Horwitz and Young (1991), and Horwitz, Tallon, and Luo (2009) on causes and effects in L2 anxiety and on the development of a scale to assess anxiety in language learning have also been pursued by Elkhafaifi (2005) regarding L2 listening in Arabic Elkhafaifi was able to distinguish L2 listening anxiety from L2 general anxiety within a classroom context, using existing scales adapted for listening Another finding was negative correlations between listening anxiety and end-of-course grades Elkhafaifi‟s (2005) findings were later echoed in Mills, Pajares, and Herron‟s (2006) research involving learners of French, which revealed negative correlations between listening competence and listening anxiety, that is to say the higher the level of listening ability, the lower the level of anxiety observed

Learners of Spanish involved in Vogely‟s (1999) study aiming to identify the causes of anxiety reported factors related to L2 input being speed, clarity and unavailability of visual

support, which were followed by processing factors like strategies used inappropriately According to the data collected on the participants‟ perception of what could be done to reduce anxiety, the largest percentage of responses fell into two categories being making input comprehensible and improving instructional factors such as more time for processing and integration of other skills into listening

Much of the literature regarding L2 listening anxiety involves the context of testing Arnold (2000) used relaxation and visualization exercises as a method for alleviating anxiety before participants took a listening test each week during the course of eight weeks The control group underperformed on the listening test at the end of the research, when compared to the experimental group Differences in a pre- and post- questionnaire indicated a positive attitude towards the exercises in promoting self-confidence and alleviating anxiety He also added that changing attitudes about listening skills is of great importance when it comes to changing how much listeners are willing to attempt during listening

According to Vandergrift and Goh (2012), little research has been carried out into what instructors can do to reduce anxiety It is also noteworthy that not all anxiety is necessarily negative in nature Horwitz (2010) argued that while anxiety can be debilitating as usually considered, a certain level of anxiety can be facilitating, serving as a motivator helping listeners to stay more focused, thus greater performance

2.3.2.2 Self-efficacy

According to Graham (2006), high levels of anxiety often result in low levels of confidence and self-efficacy since L2 listeners often ascribe good L2 listening performance to factors beyond their control Self-efficacy as the basis for self-confidence and motivation refers to learners‟ beliefs about their own ability to take part in learning activities successfully Based

on previous experience, listeners with great self-efficacy can feel confident dealing with different listening situations since they know how to deal with challenges Listeners with poor self-efficacy are not confident about their listening ability, feeling reluctant to take part in any listening activities as they fear that their inadequacies might be exposed Unlike listeners with great self-efficacy who ascribe their listening success to their own efforts, those with poor self-efficacy feel unable to improve their listening skills as they often blame factors outside their control for their poor listening performance Bandura (1993) posited that learners would

be more motivated to handle future tasks if they ascribed their listening success to factors that they had control over This can be hypothesized that teaching L2 learners to better control

their listening processes could result in listeners believing that listening success is within their control, thus improved self-efficacy This was confirmed by Graham and Macaro (2008), concluding that teaching of listening strategies facilitated listening comprehension and had salutary impact on their self-efficacy

Mills et al (2006) also investigated the impact of self-efficacy on L2 listening performance When effects for anxiety were monitored for, results revealed that L2 listeners‟ personal judgements of their self-efficacy affected how they handled listening tasks and their performance as well Despite this, the findings were valid for female participants only Mills

et al (2006) justified the difference according to the fact that the study was conducted on a voluntary basis and that the male participants might not have been serious enough about the study as their listening test results were much lower than their actual performance in class and the listening results for their female counterparts

2.3.2.3 Motivation

According to Vandergrift and Goh (2012), although the role of motivation in L2 learning has been widely investigated, there is very little literature on the relationship between L2 listening and language learning motivation There is, however, some evidence that language learners involved in tasks which develop metacognitive knowledge about listening feel motivated eventually (Goh & Taib, 2006; Vandergrift, 2002)

Vandergrift (2005) provided some empirical findings on the potential correlation between motivation, metacognitive control of listening processes, and comprehension performance The research results revealed that although listening comprehension correlated negatively with amotivation (-.34), correlations with extrinsic motivation such as personal gain and intrinsic motivation such as enjoyment were modest at 21 and 34 respectively Another finding was that a greater awareness of listening process correlated with higher levels of motivational intensity A pattern of increasingly higher correlations between the three levels

of motivation (from amotivation to extrinsic to intrinsic motivation) and metacognitive awareness of listening strategies was also observed Those who scored low on motivation, presumably as a result of a lack of self-confidence and self-efficacy, demonstrated a passive attitude towards L2 learning, and reported to have used less effective listening strategies Those scoring high on motivation, on the other hand, seemed to engage in listening behaviors which were increasingly metacognitive in nature

In summary, the three affective factors mentioned above also have great impact on listeners‟ perception of a listening task, their applying themselves to the task and their success experience during listening The factors are also interrelated as confident L2 listeners are likely to be more motivated, to experience less anxiety and to have greater self-efficacy, which has significant implications for the teaching of L2 listening (Vandergrift & Goh, 2012)

2.3.3 Contextual factors

2.3.3.1 Interactive listening

According to Vandergrift and Goh (2012), interactive listening is an important part of listening proficiency Interactive listening most often occurs in more informal contexts for language learning and reflects the type of listening language learners would like to develop in order to communicate with L2 speakers There are however constraints on interactive listening that can impact on the process and product of the listening event How listeners deal with a listening comprehension problem within an interactive listening context will depend on numerous affective factors such as willingness to take chances, fear of exposure of inadequacies, assertiveness and motivation The extent to which the interaction is affected by these factors will depend on the relationship between the interlocutors as status relationships may affect listening and the freedom listeners feel to work out meaning Differences in age, sex, language competence and power relationships often make interactive listening a context where disadvantaged listeners feel inferior This sense of powerlessness often impact on how much is comprehended as a result of heightened anxiety and the extent to which listeners dare

to ask for clarification in order to avoid losing face

2.3.3.2 Listening in informal learning contexts

Informal contexts are another factor to consider In one study, participants with a five-week study abroad experience were compared with a peer group at intermediate level taking a similar Spanish course on campus (Cubillos, Chieffo, & Fan, 2008) The participants were required to complete a pre- and post-listening test, a questionnaire on strategies use, and self-assessment of Spanish skills An interesting finding was that the group studying abroad did not perform better than the on-campus one in the listening test The researchers then ascribed the unexpected finding to the nature of the test, which was not sensitive to gains in interactive listening ability It is also interesting to note that the participants with greater proficiency of the off-campus group made bigger gains compared to their counterparts who were at a lower

level of proficiency The questionnaire responses also revealed that the off-campus group showed greater confidence in communicating in Spanish

Moyer (2006) also investigated the listening development of a group of L2 speakers of German at advanced level, who kept contact with native speakers living within an English-speaking context The results from a listening test and a language contact questionnaire showed that the quality and quantity of language contact, understandably, correlated strongly with listening ability and more confidence in speaking

2.3.3.3 Listening in formal learning contexts

According to Vandergrift and Goh (2012), academic listening refers to listening in formal academic contexts to learn subject matter content Research in academic listening has investigated the specific features of lectures Studies concerning this have revolved around the perceptions, problems, and strategies for comprehension as viewed by learners (Flowerdew & Miller, 1992), native-speaking lecturers (Flowerdew & Miller, 1996), and non-native speaking lecturers (Flowerdew, Miller, & Li, 2000) Common problems reported in these studies were speed of delivery of the lecture, difficulties with terminology of a specific course, cultural differences, and note-taking skills

Miller (2009) also investigated the characteristics of lectures which facilitated L2 engineering students‟ comprehension The participants were asked to determine linguistic features such as simple language and accent, and pedagogical characteristics such as examples, visuals, humor, prior preparation, and how the lecture was organized The findings presented suggested some important pedagogical implications for lecturers, putting forward a number of teaching strategies to help non-native students to improve their comprehension in their classes

Some other studies which focused on academic listening also carried implications for learners such as Flowerdew and Miller (1992), suggesting that learners should read course material before or after class; They should also ask their peers for help as well as ask questions in class; Moreover, they are advised to stay more focused and add notes to handouts or readings

in class as well

Kinesics is also an important factor for consideration Kinesic behavior refers to all body movements related to communication, such as gestures, head and lip movements, facial expressions, gaze, posture and interpersonal distance (Kellerman, 1992) Concurring on this,

Gullberg and McAfferty (2008) maintained that kinesic behavior such as gesture can play an important role in L2 input comprehension within classroom contexts According to Harris (2003), kinesic cues are often culturally bound and can add to, or change the literal meaning

of an utterance

To sum up, some of the cognitive, affective, and contextual factors that are supposed to have

an influence on L2 listening comprehension have been discussed in this section While some positive and negative correlations were observed, more of in-depth research related to this area of knowledge needs to be done before the relationship between a particular factor and listening success is considered definitive causality It is also worth mentioning that since the literature review is targeted, not all possible factors impacting on L2 listening comprehension are addressed in this chapter In other words, the three groups of factors dealt with in this chapter build on cognitive processing, and with the factors likely to affect the quality of that processing, which results in different listening success rates A large number of factors dealt with in this dissertation are associated with this three main groups of factors, whereas the other factors are based on different theoretical frameworks, and therefore may be impossible

to group into the three specific categories in question That said, the absence of any particular factor in this chapter should not be construed as a dismissal of its importance by other graduate researchers or material developers

2.4 Previous studies

Many a study has been carried out the world over to ascertain what listening problems L2 learners are often faced with Gao (2014) identified the listening problems and their causes, which confronted first-year intermediate-level Chinese undergraduates A key finding suggested that the main problems as well as their causes lied in the students‟ poor lexical resource by sound, limited knowledge of phonology and poor awareness of the nature of connected speech The study also suggested that as cognitive processes matter when it comes

to effective listening comprehension, Chinese intermediate-level university students should enhance their word-recognition skills as well as promote an awareness of how sounds in English connected speech are organized However, since Chinese students may have unique listening problems when compared with students in other geographic regions in the world, the research is advised to be replicated and compared with students from other cultures to provide

an insight into how differently students from different territories across the globe experience listening problems, hence a huge research gap to be addressed

Another piece of research by Yuan (1998) provided an insight into the listening difficulties encountered by proficient and less-proficient listeners The main difference between Yuan (1998) and Gao‟s (2014) study was that the participants in Yuan‟s (1998) were of different L2 proficiency Another difference lies in the research purpose of each study While Gao‟s (2014) study aimed to address the causes of the listening problems experienced by the students in question, Yuan‟s (1998) research also sought to identify the strategies employed

by the students in order to facilitate their listening comprehension According to the data collected, the less proficient listeners seem to have experienced listening difficulties more frequently than those of greater proficiency Also, both of the groups of listeners experienced the same factors being „EFL proficiency‟, „speech rate‟, „memory‟, „passage length‟ and „the ability to stay focused‟, which impacted on their listening comprehension materially It is also noteworthy that Yuan‟s (1998) framework was adapted from another framework having been devised by Teng (2002), which offered no satisfactory definitions of the factors affecting listening comprehension, in which case she had to decode the participants‟ listening difficulties according to their personal experiences, thus a possible arguable weakness in the connection between Yuan‟s (1998) research methodology and Teng‟s (2002) framework Another study of this type was conducted by Hsu (2005) into both the listening problems and the strategies used Hsu‟s (2005) work is different from Gao (2014) and Yuan (1998) in that while Hsu‟s (2005) and Gao‟s (2014) participants were at the same level of proficiency, Yuan‟s (1998) research population was of mixed ability Another difference lies in the methodology used in each study While Yuan (1998) preferred the interview and other data-collecting types of the same kind as research instrumentation, Hsu (2005) and Gao (2014) both used the questionnaire as a research tool to obtain data According to Hsu‟s (2005) findings, the main listening problems experienced by the participants were „vocabulary‟,

„speech rate‟ and „sentence length and complexity‟, two of which partially overlapped Yuan‟s (1998)‟s findings

Chien (1998) also investigated the factors responsible for 15 Taiwanese freshmen‟s understanding of English oral input and diagnosed possible causes of their poor listening comprehension, which turned out to overlap Gao‟s (2014), Hsu‟s (2005), and Yuan‟s (1998) findings to some extent A surprising finding from Chien‟s (1998) study was that the students‟ knowledge of vocabulary is weakly correlated with their overall listening comprehension performance This seems to be dissimilar to the findings from the previous studies by Gao (2014) and Hsu (2005), reporting that a lack of vocabulary resulted in poor listening comprehension Based on the findings, pedagogical implications for language learning were

also offered Although the teaching of grammar and vocabulary lays the foundations for foreign language education, it itself should not be taken for granted as the boundaries within which listening competence is well defined Rather, assisting students in activating their schematic knowledge and their ability to grasp a greater number of idea units should also be taken into account, and this can be done by providing students with socio-cultural knowledge

of the target language and by raising listeners‟ sensitivity to stress and intonation

Nowrouzi et al (2015) also examined the listening comprehension problems using the data collected from Iranian EFL students at tertiary level Unlike the other studies, Nowrouzi et al.‟s (2015) study used the questionnaire as instrumentation for the collection of data The research results revealed that the listening problems that the participants experienced can be divided into three subcategories and are associated with the three phases of listening comprehension being perception, parsing and utilization Accordingly, the most common listening problems turned out to be experienced during the perception, parsing and utilization phases, which were distraction and missing or misperceiving sounds and words, chunking difficulties and sentence forgetting, and confusion as to the main idea respectively Because of the problems experienced in all of the three phases, Nowrouzi et al (2015) suggested that all

of those problems should be „emphasized equally‟, but on the other hand posited that other areas of difficulty such as pragmatic and discourse-related problems could be neglected as the problems which are associated with the first two phases were more dominant

Another study of this sort was conducted by Zou (2015), who not only investigated Chinese EFL learners‟ listening problems but also examined the relationship between the problems experienced and the participants‟ gender as well as their proficiency Similar to Nowrouzi et

al (2015), Zou (2015) also utilized Anderson‟s (1995) three-phase framework of language comprehension and employed the questionnaire as the only instrumentation for her study The questionnaire investigation involving 83 English-major sophomores uncovered the fact that there were two perception problems and two parsing problems obstructing most of the informants‟ listening comprehension The former included the students‟ inability to recognize words learned before and their limited lexical resource, while the latter were associated with the students‟ low processing speed and their inability to recall main ideas after listening Surprisingly, the number of listening problems reported during the perception and parsing phase in Zou‟s (2015) work is the same as in Nowrouzi et al.‟s (2015) Accordingly, both Zou (2015) and Nowrouzi et al (2015) cited the same perception problem which affected the participants‟ listening comprehension being their inability to recognize words, and the same parsing problem being the participant‟s inability to recall main ideas after listening However,

whereas Zou (2015) reported the listeners‟ limited lexical resource to have had the greatest impact on their listening comprehension, Nowrouzi et al (2015) reported this problem to not have had as much impact on the students‟ comprehension In spite of this slight difference, the finding reported by Zou (2015) was consistent with that of Hsu (2005) and Gao (2014)

Goh (1999) also examined the factors which influenced 40 Chinese students at tertiary level Data were collected through small group interviews and learner diaries Twenty factors were identified and these were categorized under five characteristics: text, listener, speaker, task, and environment Many of the factors identified were related to text and listener characteristics Five factors were reported by more than two-thirds of the forty language learners who participated in the study Also, the most common factors experienced by the participants in question, in order of frequency of mention, were vocabulary, prior knowledge, speech rate, type of input and speaker‟s accent, which were cited by more than two-thirds of those taking part in the study Like Yuan‟s (1998) study, Goh‟s (1999) work also involved learners at different levels of proficiency According to Goh‟s (1999) research findings, a large percentage of more proficient listeners cited twelve factors, while only four less proficient ones reported to experience four factors Goh‟s (1999) methodology was also different from that used in most of the above-mentioned studies in one respect, that is to say the use of diaries as a supplementary research tool This therefore ensured the validity of the research as a whole as it served as a means of data triangulation

Boyle (1984) reported some similar findings in a study investigating the factors affecting a sample of students having finished their secondary education in Hong Kong According to the data collected, the seven most cited factors were practice opportunities, educational level and background, command of English / difficulty of the English used, vocabulary/idiom, attention and concentration, speaker's production and speech rate As vocabulary and speech rate were reported in most of the abovementioned studies such as Hsu (2005), Goh (1999) and Yuan (1998), this therefore indicates the high likelihood of these two factors having a role to play in listening success

A different line of research has also looked at the relationships between certain task types and listening success Generally, selected-response items drawing on listening skills appeared to

be less difficult than items which require some writing Eykyn (1992) conducted a study comparing multiple-choice items with other task types (choose a picture, wh-questions, and vocabulary lists) to check high school French students‟ comprehension of radio texts The findings revealed that learners performed best on the multiple-choice format Teng (1998) examined three test methods (multiple-choice items, cloze and short answer questions), and

also found out that the university students scored far higher when working on the choice items Cheng‟s (2004) work investigating standard multiple-choice, multiple-choice cloze and open-ended questions, also confirmed that the participants scored the best on the multiple-choice cloze, but worst on open-ended questions It seems that the focus of these studies was on the specific listening task types and listening comprehension, rather than on the relationships between specific task types and certain factors affecting comprehension, thus

multiple-a potentimultiple-al resemultiple-arch gmultiple-ap to be filled

In Vietnamese context, Cao Mai Hanh (2018) carried out a study to investigate listening problems facing non-English major freshmen at Hanoi University of Natural Resources and Environment The research purposes of Cao Mai Hanh‟s (2018) work are more or less the same as Zou‟s (2015) as she also examined the relationship between the listening problems and the participants‟ gender The difference between Cao Mai Hanh (2018) and Zou (2015) lies in the research methodology While Cao Mai Hanh (2018) used both the questionnaire and the interview as two data collection instruments, Zou (2015) and Nowrouzi et al (2015) employed only one research tool being the questionnaire The research results revealed that the most frequently encountered problems were long spoken text, too many unfamiliar words, meaning of unknown words, text with too long and complex sentences, getting a general understanding from the first listening, speech rate, gliding over the words of speaker and poor acoustic conditions The findings therefore have confirmed the previous ones by Hsu (2005), Goh (1999) and Yuan (1998) and Boyle (1984) in terms of vocabulary and speech rate The fact that the poor acoustic environment affected the participants‟ listening comprehension in Cao Mai Hanh‟s (2018) work turned out to contrast with that of Boyle (1984), who argued that this factor did not have as much impact on the population‟s listening comprehension It therefore makes sense to call for further research into how this factor behaves during listening

Most of the above-mentioned studies, however, have been conducted overseas and have examined the factors affecting university students‟ listening comprehension (i.e., Cao Mai Hanh, 2018; Gao, 2014; Hsu, 2005; Goh, 1999), while some others employed only one data collecting instrument being the questionnaire (i.e., Nowrouzi, 2015; Zou, 2015) What is more, most of the previous researchers only looked at those factors as negative ones, rather than treating them neutrally, that is to say, examining both negative and positive aspects of those factors Most importantly, the more research conducted, the more findings from independent studies are confirmed, hence greater generalizability This means that there is still abundant room for research to be carried out within Vietnamese context, adding to the

existing body of knowledge of the factors affecting listening comprehension among students

at secondary level

2.5 Summary

This chapter reviewed the literature which provides a background for this study Firstly, various definitions of listening were provided and then justification for opting for the preferred definition was offered Next, the three groups of factors being cognitive factors, affective factors and contextual factors which are known to affect listening comprehension were discussed The next section reviewed two different kinds of processes involved in listening comprehension, and explored how they are combined to facilitate comprehension The last section reviewed the relevant literature regarding listening comprehension within an EFL context, based on which a research gap is to be addressed In other words, all the previous studies were conducted in foreign contexts with a different population, which makes

it impossible for the researcher to generalize the findings to the participants in his own research context

CHAPTER 3: METHODOLOGY

This chapter describes the methodology employed in this dissertation, which follows a methods design to investigate the most common factors influencing listening comprehension Quantitative data were collected from a total of 33 learners of English with a CEFR level of B1, while qualitative data were obtained from four participants from the same population Following this, a detailed account of the research tools is provided A description of the data collection procedures is also provided, which is then followed by an explanation of how the data were analyzed to address the research questions proposed Finally, ethical considerations are also discussed in the last section

mixed-3.1 Research design

The data were collected through two instruments being the questionnaire and interview Since the data collected from the questionnaire were quantitative, but those acquired from the interview were qualitative, the research design employed in this study was mixed in nature Moreover, the mixed methods strategy utilized in this study possesses a concurrent triangulation attribute, which means that both quantitative and qualitative data are gathered concurrently, after which the two databases are compared so as to ascertain whether there is convergence, differences, or some combination of the two (Creswell, 2009) As those in the field of applied linguistics increasingly report the advantages of mixed methods designs (Creswell & Plano Clark, 2007; Ortega & Iberri-Shea, 2005), it seems to make sense to use both quantitative and qualitative for investigation into complex issues such as listening comprehension

Since 33 students were involved in this study, it is advantageous to use the questionnaire as a tool for collecting data with the intent of generalizing from a sample to a population (Babbie, 1990) Also, this research tool is cost-effective and can reach a larger population (Bowling, 2014; Denscombe, 2003) On the other hand, the interview does not only act as a triangulation tool but also as an opportunity for those involved in the research to improve their listening skills based on their verbal descriptions of the factors experienced during their interviews with the researcher Also, the interview seems advantageous as the respondents are allowed to describe their experience in greater detail, thus a more comprehensive picture of the factors experienced