Assessing efl speaking skills in vietnamese tertiary education

English as a foreign language English language teaching English as a second language English for specific purposes Educational Testing Service International English Language Testing Syst

Assessing EFL Speaking Skills in Vietnamese Tertiary Education

by Thanh Nam Lam

BA (EFL Pedagogy), MA (TESOL)

A dissertation submitted to the School of Education, the University of Newcastle, Australia

in fulfilment of the requirements for the Degree of Doctor of Philosophy (Education)

October 2018

STATEMENT OF ORIGINALITY

I hereby certify that the work embodied in the thesis is my own work, conducted under normal supervision The thesis contains no material which has been accepted, or is being examined, for the award of any other degree or diploma in any university or other tertiary institution and, to the best of my knowledge and belief, contains no material previously published or written by another person, except where due reference has been made I give consent to the final version of

my thesis being made available worldwide when deposited in the University’s Digital Repository, subject to the provisions of the Copyright Act 1968 and any approved embargo

Thanh Nam Lam

Signature: ………

Date: …… October, 2018

I acknowledge my debt to the Government of Vietnam through the Ministry of Education and Training and the University of Newcastle, Australia, for awarding me a VIED-TUIT scholarship to support my doctoral studies

I am very thankful to the University of Newcastle for providing such a rich research resource and excellent student services and, in particular, to the friendly and supportive academic staff of the School of Education I must say a special thank you to Vietnam Aviation Academy, where I am working, for facilitating my study with the most favourable conditions they can

My profound gratitude goes to Professor Max Smith, Professor Allyson Holbrook, and Associate Professor James Ladwig, who gave me interesting lectures and useful

research skills in my coursework at the University of Newcastle

I wish also to express my great appreciation to Associate Professor Kylie Shaw, Associate Professor Mitchell O’Toole, Dr Maura Sellars, Dr Rachel Burke, Ms Helen Thursby, Ms Helen Hopcroft, and Mr Nicholas Collier for their suggestions, interest, language assistance and constructive feedback on my very first manuscript My additional thanks goes to Ms Ruth McHugh for helping me with the tedious and long running job of reading proof

I would like to express my heartfelt thanks to Dr Ho Thanh My Phuong As my former supervisor, she has taught me and motivated me more than I could ever give her credit for here

I would also like to take this opportunity to thank my former instructor, Dr Vu Thi Phuong Anh She inspired me with research ideas about language testing and assessment She has shown me, by her example, what a good scientist (and person) should be

My deepest respect and appreciation go to Heads of the EFL Departments at the tertiary institutions in Vietnam for their generosity giving consent to my data collection at their EFL classes I would like to thank the EFL teacher and student participants in

Vietnam Their information and ideas constituted the core of my study Many of the participants set aside considerable amounts of their time to provide me with a profound understanding about their experiences and perceptions of oral assessment The EFL experts’ insights enriched my research results and contributed enormously to the eventual conclusions from the investigation I am greatly appreciative of their enthusiastic

cooperation in my study

I wish to thank my family for their unconditional love and encouragement during

my time of studying away from home They have been and still are ever ready to assist me

in my various endeavours

Finally, there are my lovely schoolmates here in Callaghan campus We were not only able to support each other by deliberating over our study problems and findings, but also happily by talking about things other than just our papers I cannot forget my friends from Vietnam and the US Their messages and emails gave me motivational strength to complete my study

Newcastle, 30 October 2018

Nam Lam

TABLE OF CONTENTS

STATEMENT OF ORIGINALITY ii

ACKNOWLEDGEMENTS iii

TABLE OF CONTENTS v

LIST OF ABBREVIATIONS x

LIST OF TABLES xi

LIST OF FIGURES xiv

LIST OF APPENDICES xv

GLOSSARY xvii

ABSTRACT xviii

CHAPTER ONE: INTRODUCTION 21

1.1 Background of the study 21

1.1.1 A brief history of testing L2 speaking 22

1.1.2 Trends in language testing 26

1.1.3 Context of the research 29

1.2 Research questions 44

1.3 Significance of the research 47

1.4 Organisation of the thesis 49

CHAPTER TWO: LITERATURE REVIEW 51

2.1 Introduction 51

2.2 Key issues in designing spoken language tests 51

2.1.1 Construct validity in oral language testing 56

2.1.2 Content aspect of construct validity 59

2.1.3 Reliability 60

2.3 Conceptual framework for validating speaking tests 63

2.4 Formats of speaking tests 65

2.5 Technological applications in oral assessment 68

2.6 Factors affecting test validity and reliability 69

2.6.1 Assessment criteria 69

2.6.2 Rating scales 70

2.6.3 Test tasks 71

2.7 Washback of oral language assessment 74

2.8 Summary 76

CHAPTER THREE: METHODOLOGY 78

3.1 Introduction 78

3.2 Rationale for the research design 78

3.2.1 Adoption of a mixed methods approach 78

3.2.2 Using a convergent design 80

3.3 Research setting 81

3.3.1 Research sites 82

3.3.2 Research participants 83

3.4 Data sources 87

3.4.1 Test room observation 89

3.4.2 Questionnaire surveys 90

3.4.3 Interviews 95

3.4.4 Expert judgements 98

3.4.5 Documents 99

3.5 Data collection procedures 100

3.6 Methods of data analysis 105

3.6.1 Quantitative data processing 105

3.6.2 Qualitative data processing 106

3.7 Presenting data analysis 112

3.8 Research ethics and compliance 113

3.9 Assuring research quality 115

3.10 Summary 117

CHAPTER FOUR: RESULTS: TEST TAKER CHARACTERISTICS AND TEST ADMINISTRATION 119

4.1 Introduction 119

4.2 Test taker characteristics 120

4.3 Speaking test administration across institutions 135

4.4 Candidates’ and raters’ perceptions of the oral test administration 145

4.4.1 Candidates’ perceptions of the oral test administration 145

4.4.2 Test raters’ perceptions of the oral test administration 150

4.5 Summary 155

CHAPTER FIVE: RESULTS: CONTENT RELEVANCE OF SPEAKING TEST QUESTIONS 157

5.1 Introduction 157

5.2 Defining test constructs 158

5.3 Designing the content judgement protocol 161

5.4 Selecting approaches to data analysis 164

5.5 Relevance of test contents 165

5.5.1 EFL experts’ judgements on test content relevance 166

5.5.2 Linking expert opinions with other data sources 180

5.6 Summary 190

CHAPTER SIX: RESULTS: SPEAKING TEST TASKS 192

6.1 Introduction 192

6.2 A comparative analysis of speaking test tasks across institutions 193

6.2.1 Response format 193

6.2.2 Task purposes 206

6.2.3 Time constraints 225

6.2.4 Channels of communication 230

6.3 Raters’ and candidates’ perceptions of the test tasks 235

6.3.1 Teachers’ perceptions of the test tasks 235

6.3.2 Candidates’ perceptions of the test tasks 238

6.4 Summary 240

CHAPTER SEVEN: RESULTS: RATER CHARACTERISTICS AND RATING ORAL SKILLS 242

7.1 Introduction 242

7.2 Test rater characteristics 243

7.3 Rating and scoring 246

7.3.1 Oral assessment criteria 247

7.3.2 Rating scales 248

7.3.3 Oral rating process 251

7.4 Raters’ consistency in oral rating 256

7.4.1 Scoring methods 256

7.4.2 Aspects of rating in oral assessment 258

7.4.3 Giving bonus points 261

7.4.4 Familiarity between the rater and candidates 263

7.5 Test raters’ and candidates’ perceptions about the practice of rating and scoring 266

7.5.1 Test raters’ perceptions of the rating process 266

7.5.2 Candidates’ perceptions of the rating and scoring 269

7.6 Test score analysis 272

7.6.1 Distribution of test scores 273

7.6.2 Inter-rater reliability in direct scoring between pairs of raters 275

7.6.3 Inter-rater reliability in semi-direct scoring between pairs of raters 278

7.7 Summary 283

CHAPTER EIGHT: RESULTS: IMPACT OF ORAL TESTING 285

8.1 Introduction 285

8.2 Impact of the oral test from candidates’ perspectives 286

8.2.1 Impact of test scores 287

8.2.2 Learning activities candidates found useful for the oral test 290

8.2.3 Candidates’ perceptions of the test impact on EFL learning 292

8.3 Impact of the oral test on teaching from teacher raters’ perspectives 298

8.3.1 Major desired changes in teaching speaking skills 298

8.3.2 Teaching activities to prepare learners for the oral test 300

8.3.3 Teachers’ perceptions of the test impacts on teaching EFL speaking skills 304

8.3.4 Implementing a new method of assessing speaking skills 314

8.4 Summary 316

CHAPTER NINE: SUMMARY AND CONCLUSIONS 317

9.1 Introduction 317

across raters…… ….…

9.2 Summary of research results 318

9.2.1 Issues in test administration affecting test fairness and candidates’ speaking performance 319

9.2.2 Test content relevance and inequality in test questions’ degree of difficulty 323

9.2.3 Diversity in test tasks elicited different speech patterns for assessment 324

9.2.4 Inconsistency in rating and scoring spoken language performance 326

9.2.5 Impact of oral testing on EFL teaching and learning 327

9.3 Limitations of the study and recommendations for future research 335

9.4 Implications 340

9.4.1 Implications for speaking test administrators 340

9.4.2 Implications for speaking test designers 342

9.4.3 Implications for oral test raters and scorers 343

9.4.4 Implications for oral test takers 344

9.4.5 Implications for educational policy makers 345

9.4.6 Implications for language tesing researchers 347

9.5 Conclusions 348

REFERENCES 352

APPENDICES 385

English as a foreign language English language teaching English as a second language English for specific purposes Educational Testing Service International English Language Testing System First language

Second/Foreign Language Ministry of Education and Training (Vietnam) Oral Proficiency Interview

School-based assessment Spoken English

Statistical Package for Social Sciences Teaching English to Speakers of Other Languages Test of English as a Foreign Language

Test of English for International Communication University of Cambridge Local Examinations Syndicate

LIST OF TABLES

Table 2.1 Purposes of language assessment 53

Table 2.2 Contrasting categories of L2 assessment 54

Table 2.3 Kinds of language tests 55

Table 2.4 Categories of speaking assessment tasks 73

Table 3.1 Sampling of participants and research instruments used in the study 87

Table 3.2 Summary of key points covered in the test room observation scheme 90

Table 3.3 Summary of contents covered in the questionnaire for EFL teachers 92

Table 3.4 Summary of contents covered in the questionnaire for EFL students 95

Table 3.5 Summary of the EFL teacher interview protocol 96

Table 3.6 Summary of the EFL student interview protocol 97

Table 3.7 Information about EFL experts 99

Table 3.8 Data collectection methods in a three-stage process 101

Table 3.9 Data collection before the test 102

Table 3.10 Data collection during the test 103

Table 3.11 Data collection after the test 104

Table 4.1 Test takers’ gender profile 123

Table 4.2 Speaking test anxiety across institutions 127

Table 4.3 Candidates’ oral test anxiety in relation to general self-evaluation of English129 Table 4.4 Test takers’ English education prior to the oral test 130

Table 4.5 Test takers’ profile of English-speaking class attendance 133

Table 4.6 Candidates’ oral test anxiety in relation to class attendance 134

Table 4.7 Comparing test administration methods across institutions 136

Table 4.8 Advantages and disadvantages of direct assessment 139

Table 4.9 Raters’ opinions for and against audio-recording oral tests 141

Table 4.10 Examiners’ performance during an oral test session 143

Table 4.11 Means and standard deviations (SD) for candidates’ perceptions of the test administration 146

Table 4.12 Candidates’ comments for and against of computer-assisted oral testing 150

Table 4.13 Means and standard deviations (SD) for raters’ perceptions of the test administration 152

Table 4.14 Teacher raters’ comments for and against computer-assisted oral testing 155

Table 5.1 Comparison of the English-speaking courses across the institutions 159

Table 5.2 Summary of test questions used for oral assessment across the institutions 163

Table 5.3 Information collected from EFL experts’ content judgement protocol 166

Table 5.4 Content validity index of test items (I-CVI) across institutions 167

Table 5.5 Examples of test items with I-CVIs of 1.0 and 83 168

Table 5.6 The average of the I-CVIs for all items on the scale 169

Table 5.7 Examples of items rated ‘Relevant/Highly relevant’ and positive comments by all the experts 171

Table 5.8 Examples of problematic items and experts’ suggestions for revision 172

Table 5.9 Examples of confusing test questions with experts’ comments and suggestions 173

Table 5.10 Question lengths in numbers of total words (TW) and content words (CW) across different task types adopted at the institutions 176

Table 5.11 Examples of picture-cued questions with experts’ comments 179

Table 6.1 Response formats used in oral assessment across institutions 194

Table 6.2 Description of underlying purposes of oral test tasks 207

Table 6.3 Summary of characteristics of the test tasks across institutions 208

Table 6.4 Rubric words and phrases indicating task purposes 209

Table 6.5 Elicitation question types used for interview tasks 220

Table 6.6 Time constraints in minutes for speaking tasks across institutions 226

Table 6.7 Channels of communication in EFL oral assessment across institutions 231

Table 6.8 Means and standard deviations (SD) for EFL teachers’ perceptions of test tasks 236

Table 6.9 Means and standard deviations (SD) for EFL students’ perceptions of test tasks 238

Table 7.1 Age profile of EFL teacher participants in the survey 244

Table 7.2 Assessment weighting of the Listening and Speaking components 246

Table 7.3 Assessment criteria across institutions 247

Table 7.4 Features to be assessed and scored in a rating scale 249

Table 7.5 Example of a holistic rating scale for the interlocutor 250

Table 7.6 Purposes of the oral test for EFL majors 253

Table 7.7 Raters’ tendency of giving bonus points 262

Table 7.8 Oral rating affected by candidates’ performance in class 265

Table 7.9 Means and standard deviations (SD) for raters’ perceptions of the oral rating and scoring 268

Table 7.10 Means and standard deviations (SD) for candidates’ perceptions of the oral rating and scoring 270

Table 7.11 Candidates’ opinions for and against the number of raters in each test room272 Table 7.12 Descriptive statistics of oral test scores across the institutions 273

Table 7.13 Comparing categories of test scores across institutions 274

Table 7.14 Correlation of test scores by University A’s raters 279

Table 7.15 Descriptive statistics of University A’s raters’ scores in the second time of scoring 280

Table 7.16 Correlation of test scores by University B’s raters 281

Table 7.17 Descriptive statistics of University B’s raters’ scores in the second time of scoring 281

Table 7.18 Correlation of test scores by University C’s raters 282

Table 7.19 Descriptive statistics of University C’s raters’ scores in the second time of scoring 283

Table 8.1 Means and standard deviations (SD) for EFL students’ perceptions of the test’s impact on learning 292

Table 8.2 Candidates’ desired strategies to improve English speaking skills 295

Table 8.3 Means and standard deviations (SD) for EFL teachers’ perceptions of the test’s impact on teaching and learning speaking skills 304

Table 8.4 Teachers’ changes and adjustments as a consequence of the oral test 311

LIST OF FIGURES

Figure 2.1 Model of spoken language ability in the Cambridge Assessment 57

Figure 2.2 Construct validity of test score interpretation 58

Figure 2.3 Socio-cognitive framework for validating speaking tests (Weir, 2005, p 46) 64 Figure 2.4 Interrelationship between assessment, teaching and learning 74

Figure 3.1 Example of a paragraph in the First Coding Cycle 110

Figure 3.2 Framework for validating speaking tests (adapted from Weir, 2005) 112

Figure 4.1 Age groups of test takers taking the oral test 122

Figure 4.2 Candidate self-evaluation of English proficiency in terms of general performance, accuracy, and fluency 124

Figure 4.3 Candidates’ experience with oral testing 132

Figure 4.4 Seat arrangements for oral assessment at different institutions 138

Figure 5.1 Content validity index (CVI) 165

Figure 7.1 EFL teachers using spoken English in Speaking classes 244

Figure 7.2 Challenges for raters in oral rating 254

Figure 7.3 Aspects in candidates' oral performance the rater paid attention to 259

Figure 7.4 Inter-rater agreement in oral test scoring across the institutions 277

Figure 8.1 Impact of test scores on candidates 287

Figure 8.2 Useful learning activities for candidates before the test 291

Figure 8.3 Major desired changes in teaching speaking skills 299

Figure 8.4 Factors affecting teaching oral skills 301

Figure 8.5 Useful activities to prepare students for the oral test 303

Figure 9.1 Example of a wrap-up activity for a speaking skills lesson 331

LIST OF APPENDICES

Appendix A: Ethics approval documents ……… 385

A.1 HREC Approval (17/11/2015)

A.2 Information Statement for (a) Head of EFL Faculty

(b) EFL teachers (c) EFL students (*) A.3 Consent Form for (a) Head of EFL Faculty

(b) EFL teachers (c) EFL students (*) A.4 HREC Expedited Approval of a protocol variation (22/02/2017)

A.5 Information Statement for (a) EFL teacher raters

(b) EFL experts A.6 Consent Form for (a) EFL teacher raters

(b) EFL experts A.7 Verification of translated documents

Appendix B: Data collection instruments ……….……… 422

B.1 Observation Protocol

B.2 Questionnaire for (a) EFL teachers

(b) EFL students (*) B.3 Interview Protocol for (a) EFL teachers (*)

(b) EFL students (*) B.4 Documents for EFL experts

(a) University A’s Test Content Judgement Protocol

(b) University B’s Test Content Judgement Protocol

(c) University C’s Test Content Judgement Protocol

(d) Demographic information about EFL experts

B.5 Focus group discussion protocol for EFL teacher raters (*)

Appendix C: Additional resources for data collection ……….………… 451

C.1 Procedure and speaking notes for initial contact with potential participants

C.2 Instructions for raters’ speech sample recording in test rooms

C.3 Checklist of documents to be collected

C.4 Interview transcription template

Appendix D: Examination documents ……….……… 455

D.1 Assessment criteria, rating scales and scoring sheets at three institutions

D.2 Example of test material for oral examiners (interlocutor outline)

D.3 Speaking skills and discussion strategies introduced in course books

D.4 Scoring sheets for semi-direct assessment

Appendix E: Data management ……….……….……… 463

E.1 Coding scheme of participants and data sources

E.2 Managing recorded speech samples

E.3 Multiple data sources used to examine key aspects in speaking test validation

E.4 Tabulated qualitative data

Appendix F: Examples of preliminary data processing ……… 471

F.1 Coding a rater interview transcript in two cycles

F.2 Rough transcript of a speech sample

F.3 Additional excerpts of of task-based oral performances

F.4 Occurrence of language functions in speech samples

Appendix G: Sample rating scales for assessing speaking skills ……… 479

G.1 Sample rating scales from the literature: global and analytic

G.2 Common Reference Levels proposed by the CEFR

G.3 The CEFR-based English Competence Framework adopted in Vietnam (*)

Appendix H: Flowchart of procedures for data collection and data analysis ….……… 486 Appendix I: Transcription notation symbols ……….……….………… 487 Appendix J: List of tertiary institutions in HCMC (Vietnam) ……….……… 488 Appendix K: Original quotes in Vietnamese ……….……….………… 489

K.1 Quotes from the Vietnamese press and literature

K.2 Quotes from Vietnamese interviews with research participants

_

(*) a Vietnamese translation included

GLOSSARY

The following terms will be used as key words in this research study:

• achievement test: a test which aims to measure what has been learnt within a

course of instruction or up to a given time

• construct validity: the degree to which a test can measure what it claims to be

measuring

• content validity: the extent to which the test approximately provides samples

from the domain of knowledge and skills relevant to performance according to the preset criterion

• inter-rater reliability: the extent to which two or more examiners would

agree with each other on the scores awarded to a single learner

• rating scale (also scoring rubric): an ordered set of descriptions of typical

performances in terms of their quality, used by raters in the rating process

• reliability: consistency of measurement across individuals by a test

• reliability coefficient: a statistic on a scale between -1 and +1, expressing the

extent to which individuals have been measured consistently by a test

• testing: the collection of any data that can be used to assess the language

abilities of respondents

• assessment: the inclusion of more variables when attempting to determine

language proficiency and achievement

• washback: the effect of a test on the teaching and learning leading up to it

ABSTRACT

English language mastery plays a crucial role in global communication, trading, and cultural exchange Vietnam has set English as a strategic goal of the national education system for boosting the process of regional and international integration Vietnamese education has been making a great effort to enhance the effectiveness of English as a foreign language (EFL) instruction with a focus on communicative competence,

particularly listening and speaking skills My study aimed to examine the operational practices of oral assessment as an inseparable component in relation to L2 teaching and learning in Vietnam

Participants in my study project were EFL majors (candidates) and teachers

(examiners) involved in testing English speaking skills at three universities in South

Vietnam My data collection for this empirical research started in late 2015 and continued

to early 2016 Data sources included oral test observation, questionnaire surveys,

interviews, testing documents, test scores, and EFL content experts’ judgements

The results highlight a methodological diversity in oral assessment across the

institutions in terms of test administration, task design, language content, the process of rating, and the impact of testing speaking Test scores served more to complete a required unit of study rather than to make reliable inferences about learners’ oral production ability and provide useful feedback for future improvement Interview tasks did not reflect

characteristics of natural conversations when the interlocutor (also examiner) played a predominant role in the assessment context Several oral test questions required candidates’ theoretical knowledge in order to be answered These factors had the potential to increase students’ test anxiety and hindered their best performance Discussion tasks enabled a wide variety of speech functions to be produced and provided opportunities to manage verbal interaction between paired candidates Interactive speaking revealed students’ weaknesses

in a co-constructed oral performance, and a tendency for individual turn-taking to talk about the assigned topic

My study suggests implications for various stakeholders who could assist to improve the quality of oral testing and assessment in educational contexts Rater training and double rating are necessary for oral assessment to eliminate unavoidable measurement errors by human raters My research results indicate a need for more clearly defined assessment criteria and descriptors in the rating scales to obtain higher consistency in assessing spoken English abilities The recent promulgation of the CEFR-based evaluation framework for L2 proficiency in Vietnam has brought both opportunities and challenges for those who are concerned about enhancing and standardising the national quality of EFL education

Chapter One

INTRODUCTION

1.1 Background of the study

Oral language testing and assessment for educational purposes play an inseparable part in Communicative Language Teaching (CLT) as they can provide valuable evidence

regarding pedagogical effectiveness, learners’ progress and achievement in their ability to communicate with others Teachers and education administrators can make data-based decisions to implement curriculum adjustments and/or teaching methods to increase

learners’ success in language study Language assessment is popularly used to gather information for making decisions that have definite influences on stakeholders –

individuals, programmes, organisations, and societies (Bachman & Palmer, 2010)

Contemporary language assessment is based on theories of learning and cognition,

attaching importance to authentic skills and abilities necessary for learners’ future success (Cheng, 2005) Particularly in this era of global integration, when English has become an important means for global communication, cooperation and development, many

theoretical and practical issues have been emerging in research about measuring English as

a foreign or second language (EFL/ESL) competence not only in high-stakes examinations but also in regular school-based assessment (SBA) I provide a brief historical overview of foreign or second language (L2) assessment for better understanding of the advancements

in the use of various measurement methods and current trends in language testing

concomitant with innovations in applied linguistics and EFL/ESL teaching methodology

My review focuses on L2 assessment in “the mother tongue English countries” (Jenkins,

2006, p 160), such as the United Kingdom, the United States, and Australia, where English

is the first and official language, and from which language testing was originated and was developed

1.1.1 A brief history of testing L2 speaking

Language assessment, a branch of Applied Linguistics, was put into practice more than two centuries ago Testing L2 speaking skills is the youngest subfield of language assessment (Alderson, 1991) EFL/ESL public examinations date back to the second half of the 19th century when the University of Cambridge sent examination papers overseas for the first time in 1863 (Giri, 2010) It was not until the first decades of the 20th century that

remarkable innovations in language tests were seen in Britain and subsequently in the United States (O’Sullivan, 2012) The University of Cambridge Local Examinations

Syndicate (UCLES) first introduced the Cambridge Proficiency Examination (CPE) in

1913 The examination was administered to test the English language performance of people wishing to pursue an education in Great Britain However, the testing of spoken English for foreigners did not become an area of interest until World War II, 1939 (Fulcher, 1997b)

Testing speaking before 1939

The beginning of language testing research in the United States witnessed an excessive concern regarding how to achieve reliable scores from L2 tests Measuring oral language skills could not produce consistent scores because the rating process depended on a scorer who could be influenced by many uncontrolled factors (Heyden, 1920) Oral tests for large numbers of candidates were considered to be impractical

In 1913, a committee of the Modern Language Association recommended the inclusion of an aural component in U.S university entrance examinations (Spolsky, 1995) Many institutions adopted or adapted the recommended format consisting of a 10-minute dictation section, and a written response section requiring candidates to write down answers

to aural questions delivered by the examiner (Decker, 1925; Lundeberg, 1929) Candidates were not actually required to speak in the test, but write down the phonetic script indicating how written words were pronounced Oral tests took the form of testing pronunciation knowledge rather than communicative performance Concerns about reliability was such a major issue that “few tests in use at that time sought to assess the spoken language, even though elements within the profession had been aware of this deficiency for a long time” (Barnwell, 1996, p 18)

In 1930, the College Entrance Examination Board introduced the first true speaking test component Before that, language testing practioners solely concentrated on multiple-choice tests as objective measures of L2 proficiency rather than the complexity of testing speaking This test was designed to meet the requirement of the U.S government that institutions had to have a clear indication of students’ English language ability prior to admission (Spolsky, 1995, p 55) Its Speaking component constituted one part of a five-section test: Reading I, Reading II, Dictation, Speaking, and Essay writing For the oral section, the candidate was engaged in a topic-based conversation with the examiner Multi-aspect assessment criteria – fluency, responsiveness, rapidity, articulation, enunciation, command of construction, use of connectives, vocabulary and idioms – were rated on a 3-point scale of: proficient, satisfactory, and unsatisfactory Test takers’ shyness was recorded

as an extra consideration in test score interpretation (Fulcher, 2003, p 3)

The story of L2 testing in the United Kingdom was slightly different from that in the United States Testing of English for foreigners by the University of Cambridge started

as early as the mid-19th century By 1898, the University of Cambridge was in charge of 36 colonial centres for overseas examinations with more than 1,200 candidates (Giri, 2010) In

1913, the Certificate of Proficiency in English examination introduced by the University of Cambridge Local Examinations Syndicate (UCLES) included a sub-test of spoken English (Roach, 1945) in which there was a 30-minute section for Dictation and another 30-minute section for Reading and Conversation, which was graded only for pronunciation The main purpose of the British tests was to support the training programme and encourage effective language education in foreign schools (Brereton, 1944) The Cambridge examinations were

“innovative and almost non-academic” in their objectives, focusing on language use rather than language knowledge (Giri, 2010) Unlike in the United States, assessing speaking skills in the United Kingdom was not much concerned with consistency or measurement theory (Fulcher, 2003) The College Board's examination put more emphasis on accuracy in language evaluation with true/false questions and criterion-referenced grading, whereas the Cambridge examination concentrated more on the local syllabus with English literature and its absolute reliance on subjective grading (Fulcher, 2003; Taylor, 2011)

Testing speaking in the war years

The Army Specialized Training Program (ASTP) was launched in 1942 to address the problem that many American service personnel did not have sufficient foreign language skills required for carrying out their duties in World War II This was the first language instruction programme to equip trainees with colloquial spoken language and knowledge of the context in which the language was to be used (Angiolillo, 1947, p 32) The assessment

of a trainee’s success upon completing the programme shifted from grammar knowledge to ability to use the language The test included three tasks: securing services, asking for information, and giving information Well-trained interlocutors and raters were required to administer the test in friendly and informal settings

In 1956, the Foreign Language Institute (FSI) undertook the responsibility to assess language proficiency for all human resources engaged in external service One considerable challenge in testing was the officers’ age variation, which was reported to influence raters’ evaluation Two years later, the EFI testing board supplied raters with the 6-point scale and

a 5-factor checklist for assessment including accent, fluency, grammar, vocabulary, and comprehension This new testing procedure was later adopted by the Central Intelligence Agency (CIA), the Defense Language Institute, and the Peace Corps during the 1960s (Fulcher, 2003)

Testing speaking in the post-war years

The Test of English as a Foreign Language (TOEFL), which was developed by

the Educational Testing Service (ETS) in the United States in the early 1960s and remained unchanged for 40 years, marked “a major innovation in the standardization of English language tests” (O’Sullivan, 2012, p 13) The TOEFL Test of Spoken English (TSE) is optional for candidates and “most commonly intended only for those applying for graduate teaching assistant positions” (Dalby, Rubenstone, & Weir, 1996, p 38) The official

speaking component was not included in the TOEFL until late 2005 when the ETS

introduced a new format - the TOEFL Internet-based Test (iBT) The TOEFL scores

provided evidence of English language ability for non-native speakers to be considered for admission not only to colleges and universities in the United States but also to tertiary institutions many other English-speaking countries such as the United Kingdom and

Australia

By the early 1980s, test developers from the English Language Testing Service (ELTS) in the United Kingdom began to explore the potential of designing tests of

language for specific purposes The ELTS soon became the International English Language Testing System (IELTS) in 1989 The IELTS is a standardised test of English language proficiency designed for non-native English language speakers worldwide At present, the IELTS is jointly managed by the British Council, International Development Programme (IDB): IELTS Australia, and Cambridge English Language Assessment The IELTS test covers all four language skills in four separate tests: Listening, Speaking, Reading, and Writing The Speaking test is a face-to-face interview between the test taker and an

examiner A speaking test session includes three parts: introduction and interview (4–5 minutes), long turn (3–4 minutes), and discussions (4–5 minutes) Speaking tests are

included in both IELTS Academic and IELTS General Training formats These two types

of the IELTS test serve different purposes: the first “is for people applying for higher education or professional registration in an English speaking environment”, and the latter

“is for those who are going to English speaking countries for secondary education, work experience or training programs” (International English Language Testing System, 2018a)

Both are designed to cover the full range of ability from non-user to expert user

Language testing in Australia had its first steps in the early years of the last century The 1980s and 1990s were “characterized by a striking growth in the application of

language testing – frequently in contexts governed by macropolitical pressures”

(Hawthorne, 1997, p 248), and marked with the establishment of several institutes

specialising in ESL/EFL teaching and research For example, Australian Council for

Educational Research (ACER), established in 1930, has grown into one of the world’s leading bodies in educational research In 2012, ACER launched the PAT series

(Progressive Achievement Tests) as an online assessment system to provide teachers with objective and norm-referenced information about students’ skills and understanding in their learning and development, including language skills (ACER., n.d.) Founded in 1988, the National Centre for English Language Teaching and Research (NCELTR) has become an integrated centre for research, publishing, teacher education, TESOL/Applied Linguistics resources, and English language programmes, in coordination with Cambridge Exams (NCELTR, 2004) The late 1990s witnessed the Australian Second Language Proficiency

Ratings (ASLPR) become the standard tool adopted widely in Australia for “provid[ing] performance descriptions couched in terms of the practical tasks that learners can carry out and how they carry them out at nine points along the continuum from zero to native-like proficiency” (Ingram, 1996, p 3) This form of descriptive (or criterion-referenced) scale presents three kinds of information about each proficiency level: general description of the language behaviour suitable for the level, (2) examples of what specific language tasks are and how they are performed, and (3) comments briefly explaining concepts and key

features The ASLPR can, therefore, be used for candidates with a diverse range of

employment and educational backgrounds (Giri, 2010; Ingram, 1996)

Australia’s language tests also served as devices of social policy “designed to control the flow of immigrants and refugees and to determine access to education and employment in settings of intergroup competition” (McNamara, 2005, p 351) For

example, the implementation of the White Australia Policy, which ended in 1973 (Antecol, Cobb-Clark, & Trejo, 2004), indicated that “covert policy hinged less on issues of language and much more on race” (Smith-Khan, 2015) Once used for a political purpose, the

language testing was aimed to facilitate immigrants speaking a European language, but to restrict “non-European” people’s migration (Smith-Khan, 2015) Language testers in Europe and America have been involved in considerable debates surrounding the role of language testing in immigration procedures and citizenship policy (McNamara & Eades, 2004)

1.1.2 Trends in language testing

The above historical overview demonstrates that oral testing has achieved significant advances from its beginning up to the present Modelling in L2 testing is not a new

concept Through the development of language testing, we can see that L2 theorists and methodologists have developed and adopted language testing models in accordance with the existing models of language teaching that changed over time to meet socio-political needs in specific periods of history Trends in language testing feature those in language teaching The following section provides a critical perspective on three different trends in language testing based on various ways of defining language ability: the pre-scientific trend, the psychometric trend, and the psycholinguistic-sociolinguistic trend

The pre-scientific trend

Language testing theorists vary in their views of the nature of language and language abilities, which in consequence, has led to variations of language testing trends in different but overlapping periods at times Before the 1960s, language was viewed as a means to know about the target culture and language learning as a way of intellectual training

(McGarrell, 1981) The purpose of language teaching and learning was to achieve

knowledge about the target language (vocabulary, grammatical rules) as objectives of study This “separatist” view (Davies, 1968, p 216) did not make an important

consideration of the relationship between society and language, or the social context in which language was used Language testing was viewed as testing the memorisation of words and grammatical accuracy through writing and translation exercises

In this “intuitive” stage (Madsen, 1983) or also the pre-scientific era (Spolsky, 1978) of language testing history, decisions on teaching and testing chiefly depended on language teachers’ and testers’ personal discretion Subjective judgements were, therefore, inevitable The pre-scientific trend resulted from the traditional grammar-translation

approach in language teaching that required learners to demonstrate understanding of vocabulary meaning learnt by memorisation and to apply rules from grammar lessons taught in a deductive manner

The advantage of this language testing trend is that it has enabled a global

evaluation of learners' L2 ability through their composition and writing activities that require them to synthesise their knowledge of linguistic rules and components to produce syntactically and semantically acceptable language Test tasks such as sentence structure analysis can help learners gain control over formal accuracy when labelling its parts or combining parts into larger units (Ingram, 1985)

The psychometric-structuralist trend

During 1960s, the birth of behaviourist language learning theories emerging out of the work

of Fries (1952) and Skinner's work (1957) led to a psychometric-structuralist trend in language testing, also known as the era of scientific language testing (Madsen, 1983) This trend provided another view of language as a combination of discrete-point patterns that

learning aimed at “acquiring conscious control of the phonological, grammatical, and lexical patterns of a second language, largely through study and analysis of these patterns

as a body of knowledge” (Carroll, 1983) The audio-lingual approach, the oral structural method, the mimicry-memorisation method, etc were typical of L2 pedagogy during this period Language competence was viewed as the ability to handle discrete elements of the language system and develop individual language skills

The discrete-point approach to testing aimed to measure language proficiency by examining learners' grammar and vocabulary knowledge of the discrete items (e.g syntax, correct choice of preposition), and discrete aspects of language skills by treating each of them separately (e.g listening for gist, reading for specific information) Discrete-point language testing became the most widely used in the1960s and 1970s Tests of this type are still popularly practised in several parts of the world today (Giri, 2010)

This approach to testing separate aspects of language received a great deal of

criticism in the L2 teaching community The counter-argument was that language is not only the sum of its parts, but the discrete parts should be mobilised and mutually integrated

to perform particular tasks in particular circumstances (Ingram, 1985) Language testers should pay more attention to the development and measurement of communicative

competence rather than those of linguistic competence (Weir, 1988) This suggestion is in line with beliefs that instead of establishing a learner’s L2 knowledge in terms of skills and elements, testers should attempt to test his/her ability to perform in a specified

sociolinguistic context (Spolsky, 1978; Morrow, 1979)

The psycholinguistic-sociolinguistic trend

L2 learning involves mastering its skills (listening, speaking, reading and writing) and elements (pronunciation, vocabulary, grammar, etc.) Nevertheless, a language is not learnt

as isolated components of a system but in its entirety, i.e language skills and elements are learnt together in their mutual relation This integrative (or holistic) perspective of language and language learning implies that language learning is a unitary process that goes beyond the mastery of individual skills and elements and includes an appropriate organisation of these components in diverse social situations (Giri, 2010)

The psycholinguistic-sociolinguistic trend, also referred to as the

integrative-sociolinguistic trend (McGarrell, 1981), came into being as an opposing trend to the

discrete-point approach to language testing It acknowledges that language ability is

revealed in actual performance rather than the accumulation of discrete language elements Language learning is more than gaining control over a set of structures or usage, so the purpose of language testing should be oriented to accessing communicative competence demonstrated by a candidate's performance in a given social context (Howard, 1980; Weir, 1988)

1.1.3 Context of the research

Vietnam is a Southeast Asian country with a history stretching for over 4,000 years

Vietnamese people have possessed a traditional fondness for learning (Zhao et al., 2011; Le

V M., 2016) The feudatory monarchs’ education in Vietnam through many centuries was dominantly influenced by Chinese Confucianism, from its script system to ideological values (Lam & Albright, 2018; Welch, 2010; MOET, 2016a) The appearance of Western elements in Vietnamese culture can be traced back to Catholic missions established by Western priests in early decades of the 16th century (Fox, 2003; Alpha History, 2018; Viet

Vision Travel, 2007) Today’s Romanised Vietnamese script, known as Quốc Ngữ

(language of the country), was developed from important contribitions by the French Jesuit missionary Alexandre de Rhodes during the 17th century (MacKinnon & Hao, 2014; Pham

K., 2017) The teaching of Quốc Ngữ started in the South of Vietnam at the beginning of

1879 (X V Hoang, 2006) The policy of Quốc Ngữ utilisation expanded to the early

20th-centuty North’s education (Tran, 2009) and gradually became a Vietnamese people’

popular means of communication

In almost 100 years under French colonial rule from the mid-19th century until

1954, French education was imported, and French gradually became the official medium of instruction at all levels in Vietnam This period winessed “a mixed education system with French schools, Franco-Vietnamese schools and Confucian feudalist schools and classes existing side by side” (M H Pham, 1991 , p 6) French schools were established “for Europeans and not necessarily for Vietnamese” (Kelly, 1978, p 96) Franco-Vietnamese

schools combined both the French and Vietnamese languages in teaching subjects related to Vietnam such as morals, literature, geography, history, etc These bilingual schools

inspired intellectuals welcome the new Western-style learning, seeing it a good approach to enlighten the people, develop the country, and consequently step

towards to national liberalization (Tran, 2009, p 19)

The employment of French in governmental affairs and official examinations

restricted the influence of Chinese culture and the use of Mandarin Chinese calligraphy (T

G Nguyen, 2006; H T Do, 2006) Colonisation resulted in “the fall of the old Mandarin class and the rise of a new elite of French-speaking Vietnamese administrators” (Wright,

2002, p 229) Meanwhile, the call for maintaining and developing the Vietnamese

language received great support from patriotic Vietnamese intellectuals (e.g Association

for Quốc Ngữ Dissemination, Dong Kinh Nghia Thuc Movement) and local Vietnamese

press (e.g Gia Dinh Newspaper, Dong Duong Magazine, Nam Phong Magazine) not only for the sake of literacy enhancement but also for national destiny (H H Nguyen, 2017; V

S Nguyen, 2016: V P Le, 2014; H A Do, n.d.) On one hand, the Vietnamese

intelligentsia of this period valued French in enriching knowledge about the French

civilisation and new advancements from the West On the other hand, they did not advocate

the adoption of French as the country’s official language, but asserted that Quốc Ngữ was

the only language suitable for Vietnamese people and culture and should be consolidated in the national existence and development (T P H Tran, 2009; V P Le, 2017) Success of

the August Revolution of 1945 in northern Vietnam contributed to making Quốc Ngữ the

national and official language to achieve “a goal of full literacy throughout the population” (Wright, 2002, p 233)

The popularity of Quốc Ngữ, which uses the Latin script, created favourable

conditions for the introduction of English into Vietnam during the French colonial regime The French rulers needed English speakers capable of working in diplomatic affairs and commercial services in Vietnam For this reason, English was only promoted as a required subject in senior high schools, not at lower educational levels (L T Lam, 2011) The educational policy towards English in particular, or any other foreign languages in general, from a historical perspective, “has been a barometer of Vietnam’s relations with other

countries and how the foreign language curriculum has been directly affected by those relations” (Wright, 2002, p 226) in each specific period The 20th century saw “political and economic factors were identified as most important in determining what foreign

languages were to be promoted” (Do, 1996, p i)

1.1.3.1 Historical overview of EFL education in Vietnam

The history of English language education in Vietnam can be divided into two periods: (1) before 1986 and (2) from 1986 up to the present The year 1986 was chosen as a dividing point of time because this year marked the Vietnamese government’s initiation of its

overall economic reform (Đổi Mới), implementing the open-door policy to the world “as a

departure from obsolete dogmatism” which had lasted over the decade after Vietnam became an independent country English has emerged as a dominant foreign language in Vietnam since then

(1) Teaching English in Vietnam before 1986

English in Vietnam in the past experienced many ups and downs English language

teaching (ELT) in Vietnam before 1986 can be subdivided into three periods: (i) from the beginning of the French colonisation of Vietnam up to 1954, (ii) from 1954 to 1975, and (iii) from 1975 to 1986

Before 1954, English education was limited due to the promotion of French during the French colonial period However, English obtained an important role in the secondary language curriculum and became a madatory subject for senior high schools There are almost no extant writings on ELT in Vietnam prior to 1954 What has been kept until today

are some English textbooks designed by French authors (e.g L’anglais Vivant: Classe de sixième, L’anglais Vivant Classes de troisième, 1942), and a few English-Vietnamese

dictionaries compiled by Vietnamese scholars The contents of those teaching materials include pronunciation drills and reading comprehension practice The chief method of teaching English in Vietnam during that time was the grammar-translation method (Hoang, 2010)

From 1954 to 1975, Vietnam was divided into two parts – North and South, as a

consequence of the Geneva Conference signed in Switzerland (1954) between the two

governments in the same country: Democratic Republic of Vietnam (the North) and

Republic of Vietnam (the South) During that period, Russian was studied in North

Vietnam as the main foreign language in the formal educational system to enable direct interactions with the Soviet Union, whose major aid effort was “to help “industrialize” Vietnam to show Southeast Asia the practical benefits of socialist orientation” (Radchenko, 2018) English was the dominant foreign language in South Vietnam only and was taught for direct interactions with the U.S.A ELT in South Vietnam was facilitated with a free supply of English textbooks and teaching resources from the US Government (Tran, 2014) The purpose of foreign language teaching was to equip students with sufficient knowledge and communication skills in English to work with foreign organisations and promote co-operation with advanced capitalist countries (T H Dang, 2004)

After the reunification of the country in 1975, foreign language teaching in Vietnam was characterised by the dominance of Russian over English and French in the national education system (Be & Crabbe, 1999) From 1975 to 1986, learners of Russian received educational aids from the Russian Government Every year, hundreds of Vietnamese

teachers and students were sent to the former Soviet Union for both undergraduate and graduate studies When Russian held a dominating position of foreign language in Vietnam, English was taught in a limited number of classes in just high schools in big cities Tertiary-level education witnessed a decrease in the number of students enrolling for English both as

a discipline and as a selective subject (Hoang, 2010)

(2) Teaching English in Vietnam from 1986 up to the present

The period from 1986 up to the present is characterised by the rapid EFL development and expansion in Vietnam The English learning movement commenced in late 1986 as

Vietnam opened its diplomatic door to the whole world In the context of economic

integration, English became the first option for the majority of language learners A new

7-year series of English textbooks, entitled from English 6 to English 12, was applied to

students from Grade 6 to Grade 12 accordingly Although the textbooks claimed to train EFL students for all of the four language skills, more opportunity was provided to develop reading skills while very few tasks were devoted to listening and speaking skills There was even a significant decrease in the proportion of listening and speaking activities in the

textbooks for the last three levels (Grades 10 to 12) to less than 20% for Speaking, and 0% for Listening skills The unbalance of language skills focus in formal EFL education was subject to “the backwash effect of the structural end-of-level, graduation examination on textbook design and methodology” (Be & Crabbe, 1999, p 136) as the national high school English examination designed for the 12th graders took the form of paper-based

assessments concentrating on reading comprehension, and grammatical and lexical

questions This is the most important examination after 12 years of schooling because

“passing the exam certifies young people as having completed secondary education and paves their way for further education at the university level” (World Bank, 2017)

Vietnam made a notable innovation point in 2010 when the Vietnamese Education Publishing House (VEPH) in collaboration with world leading publishers (e.g MacMillan Education and Pearson Education) designed and developed the 10-year English textbook series (Hoang, 2016), i.e new generations of school children start learning English three years earlier than in the past This reformation in textbook writing for language learners from primary to high school aimed to improve the English teaching and learning quality in Vietnam, and “provide an excellent vehicle for effective and long-lasting change”

(Hutchinson & Torres, 1994, p 323) There used to be a 3-year English textbook series for students who started learning English from Grade 10 Today, there are two popular systems

of English textbook series in nationwide use: the 7-year English programme for students who learn English from Grade 6 to Grade 12, and the 10-year English programme for those starting from Grade 3 The 10-year English programme – a product of the National Foreign Languages Project – was initiated in large cities of Vietnam in 2010, first in Ha Noi and HCMC This programme aims to achieve comprehensive innovation in EFL teaching and learning throughout the national education system (MOET, 2010; “Distinguishing English programmes”, 2018) The methodological approach adopted for the new programme is communicative language teaching (CLT) As such,

teaching and learning are organised in a diverse communicative environment with several interactive activities (games, songs, story-telling, riddles, drawing, etc.), in the forms of individual work, pair work, and group work (Huy Lan & Lan Anh, 2010)

However, the programme implementation has faced numerous obstacles because of

inadequacy of qualified teaching staff and school facilities (Anh Thu, 2017; Quynh Trang, 2017)

In major cities of Vietnam, secondary students are taught two simultaneous English programmes: one is designed by the MOET (7-year programme), and the other is an

intensive English programme (10-year) by foreign textbook designers to help to prepare students for International English Certificates, e.g Cambridge ESOL examinations It is unnecessary to require students who have learnt English for at least 3 years earlier to start again with those who only started learning English from Grade 6 However, this

unreasonable existence has lasted for years (My Ha, 2016; Hoang Huong, 2018) Inevitable consequences are boring lessons for both teachers and students, and discrepancy in English knowledge and proficiency across students of the same grades

Since 2014, Vietnam has adopted a 6-level evaluation benchmark of English

proficiency based on the Common European Framework of Reference (MOET, 2014) According to this framework (Appendix G.3), primary school students should obtain Level

1 (A1) after 5-years of English training Level 2 (A2) is required for secondary and

vocational training students High school graduates, and non-English major universities are obliged to achieve at least Level 3 (B1) for English proficiency English major college and university graduates are expected to reach Level 4 (B2) and Level 5 (C1) respectively (Thanh Tam, 2016; MOET, 2014)

1.1.3.2 Current situation of Vietnamese EFL education

In the present situation of internationalisation, the English language plays an increasingly important role in many fields such as economics, science, education, etc in Vietnamese society (Hoang V V., 2013; Phung V M., 2015) Like many other Asian countries,

Vietnam considers foreign language teaching and learning, particularly English, to be “a national mission” specified in its educational innovation strategies (Phan L H., 2013, p 162) Teaching and learning English in Vietnam have grown at “an unprecedented speed” with the establishment of numerous foreign language centres (Do H T., 2006) English has become a compulsory subject in the national education system at both secondary (Grades 6-9) and high school (Grades 10-12) At primary level (Grades 1-5), English is taught as an

elective subject (Hoang V V., 2010) depending on specific conditions and needs of each locality There has been a growing need for Vietnamese learners to have their English proficiency assessed and certified for employment, job promotion, graduation

acknowledgement, pursuance of higher education, studying abroad, etc (Qandme, 2015; Nunan, 2003)

My study on oral language testing and assessment was derived from the current situation of English education in Vietnam, which I will discuss with regard to the crucial role of English in socio-economic development, the ongoing challenge of L2 competency testing and certification, and the adaptive application of a common reference framework in foreign language assessment as an ultimate requirement of EFL education in the context of international educational integration

(1) Role of English education in socio-economic development of Vietnam

The initiation of political and economic reforms (‘Đổi Mới’) in 1986 marked a growing

recognition that foreign language learning plays a strategic role in facilitating social change (Bui & Nguyen, 2016; Nguyen, 2012) Since Vietnam joined the World Trade Organisation (WTO) in 2006, English has become an essential foreign language in the national education system (Nguyen N T., 2017; Ton & Pham, 2010) The growth of the English learning movement in Vietnam has been in line with the popularity and wide spread of EFL teaching and learning in South-East Asian countries and in the world (Kam, 2002)

Vietnamese educators assert that enhancing English competence is a step heading in the right direction to help Vietnamese education integrate into world education (Bao Dien

Tu Giao Duc Viet Nam, 2015) This strategy, as part of the National Foreign Languages Project for the period from 2008 to 2020, aims at making substantial changes to foreign language education in Vietnam, especially English (Vietnamese Government, 2008) This project formally addresses the nationwide issues of teacher training and language teaching quality (Nguyen C D., 2017), and more broadly, its goal is to arouse the entire society’s interest and create a universal motivation for English learning

The increased use of English in universities via imported educational programmes and English as a mean of instruction (EMI) were strongly encouraged as part of the

governmental policies to promote global cooperation and Vietnamese education

modernisation (Vu P A., 2018) Vietnam’s integration with the world attracted significant investment from international organisations such as the World Bank (WB) and the United Nations Educational, Scientific, and Cultural Organisation (UNESCO) WB-funded higher education projects fostered the adoption of foreign programmes in English, which

contributed to the standardisation and competitivity of Vietnamese tertiary curricula (Dang

Q A., 2009) Despite the availability of many other foreign languages, (e.g French,

Japanese, Chinese, Korean, etc.) most Vietnamese students choose to learn English because

it offers many more well-paid job opportunities, Internet information access, or study abroad in advanced education in the United States, the United Kingdom, Australia, Canada, etc Recently, many Vietnamese students have chosen to study in Asian countries like Singapore, Thailand, Japan, or South Korea, but English is an important medium of

instruction, and knowing English is a remarkable advantage for intercultural

communication in these countries (Le & Chen, 2018) English is now widely used in

various activities at school, at work, and in daily communication since millions of speaking people from different countries come to Vietnam for travel, work, and education (Thanh Binh, 2015; Le Q K., 2016; Lam & Albright, 2018)

English-(2) Challenges in teaching and testing oral skills in Vietnam

Teaching and learning EFL for the purpose of communication has been an ultimate

necessity from primary schools to universities to support Vietnam’s open-door policy and global integration At no time in the recent history of Vietnam has there been such an apparent shift from the grammar-translation method towards a communicative approach, particularly focusing on the development of oral communication skills (Bui H., 2006; Le S., 2011) In CLT classrooms, “the objective of teaching spoken language is the development

of the ability to interact successfully in that [target] language, and that this involves

comprehension as well as production” (Hughes, 2003, p 113) To evaluate learners’

language acquisition and their ability to use spoken English for communication, oral test development and administration should be placed under the serious consideration of policy makers, test designers, and examiners

At most universities in Vietnam, students’ performances in a test for a unit of study (or module) are assessed using a 10-point grading system The result will be converted into

letter grades, e.g from A through D or F (MOET, 2007) This academic evaluation policy requires more meticulous consideration to achieve the highest precision and objectivity in scoring because numerical test results contribute to making important educational decisions and orienting pedagogical activities (Language Testing Service, 2003) A disparity of 0.5 in test scores, or even 0.1 according to institutional scoring regulations, may result in a

difference to the ‘Pass’ or ‘Fail’ status of a candidate, an obligation of re-enrolment (and consequentially re-payment) for a ‘Fail’ subject, decisions regarding scholarship awards, graduate degree classification, etc

Despite many attempts to improve the quality of EFL teaching and learning such as teacher training programmes, international TESOL conferences, teaching material changes, restructuring curriculum, etc the most recent statistics by the MOET found that:

University graduates who can meet the English skills requirements of the employers account only about 49%, an approximate of 18.9% graduates cannot, and 31.8% of them need further training It means that more than half of graduates do not

sufficiently meet English speaking skills requirements (V Le, 2016)

Some Vietnamese university graduates have obtained sufficient credits for their English modules at school, even holding international English certificates (TOEIC,

TOEFL, etc.) with a satisfactory score for graduation acknowledgement, but they cannot speak English fluently (Thanh Ha, 2008; Moc Tra, 2017; Nguyen X Q., 2016) It seems that the question raised over a decade ago regarding the quality of university students’ English education currently remains an unsolved problem: learning English for 10 years, students cannot use the language (Vu, 2007) Four deficiencies associated with the EFL training in Vietnamese tertiary education are (1) lack of attention to the English level of student intake, (2) lack of comprehensive criteria to evaluate learners’ level, (3) lack of concern about the role of testing and assessment in the training process, and (4) lack of consideration into the learner’s role in the training process (Vu, 2007)

EFL testing and assessment in general and spoken English evaluation in particular demonstrate noticeable shortcomings from secondary to tertiary level in Vietnam These deficiencies have direct or indirect impacts on young learners’ English-speaking ability, from their early years of English learnings at lower secondary schools to their successive

years of higher education at university CLT started as a new approach to L2 pedagogy in the 1970s and 1980s (Richards, 2006; Pham H H., 2007; Walia, 2012; Murphy & Baker, 2015) and has been adopted into teaching English in Vietnam since the early 1990s (Ngoc

& Iwashita, 2012) The main goal of CLT is the teaching of communicative competence rather than grammatical competence as in traditional lesson formats (Richards, 2006) The

practice of CLT does not neglect the role of grammar in language learning, but makes a clear distinction “between knowing various grammatical rules and being able to use the rules effectively and appropriately when communicating” (Nunan, 1989, p 12)

Communicative competence emphasised in CLT includes the ability to use the target language for different purposes and “maintain communication despite having limitations in one’s language knowledge (e.g through using different kinds of communication

strategies)” (Richards, 2006, p 3).To achieve those obectives, CLT classrooms are

characterised by “practical hints and communicative activities designed to stimulate

conversation and communication among students” (Banciu & Jireghie, 2012, p 96) such as role-play, interviews, information gap activities, pairwork, groupwork, etc

CLT has been implemented for almost three decades in Vietnamese EFL education since the 1990s (Ngoc & Iwashita, 2012) However, oral testing to assess Vietnamese EFL learners’ communicative competence is not included in the national graduation exams for high school students This is the most important examination that students have to pass after 12 academic years of formal education from primary to the completion of high school level, and to try to obtain the highest scores they can to compete for university admissions The score of the English subject is calculated together with three or four other subjects’ scores to decide whether the student can pursue further education at tertiary level A

difference of 0.25 points in the total score might decide whether a student will be admitted

to a university or not (Dieu Linh, 2018) Nevertheless, the national graduation exam papers

of English for high school students have not been designed in a standardised format but have been altered year after year in the proportion of multiple-choice items and self-

constructed responses

Innovations in foreign language examinations are like getting stuck in a roundabout circle Before 2006, candidates in the university entrance examination took the

English test comprising 20% multiple choice format and 80% creative writing From 2006, the English test is 100% multiple choice format After that, the

multiple-choice format was criticised for not being able to evaluate students’

authentic ability, far away from practical skills, and then the English test was adjusted to 80% multiple choice and 20% for creative self-constructed responses The change was maintained for two years The 2017 exam had 100% multiple choice format again (Thanh Tam & Phuong Hoa, 2016)

English lessons at high schools focus mainly on grammar, vocabulary, and written tests Students virtually do not have listening and speaking practice, or group discussions (Minh Nhat, 2007) Nevertheless, there exists a remarkable gap between methodological guidelines and classroom practices of English teaching as indicated in the following

comment:

although the rhetoric of the Vietnamese Ministry of Education and Training stresses the development of practical communication skills, this is rarely reflected at the classroom level, where the emphasis is on the development of reading

comprehension, vocabulary and structural patterns for the purposes of passing the end-of-school and university entrance examinations into colleges or universities (Hoang V V., 2010, p 16)

Oral assessment at tertiary level is included in the English major training

programmes in the form of mandatory end-of-course tests for the Listening-Speaking modules There is a high demand for spoken language ability from English majors because these students need oral skills in English for studying their major subjects (e.g linguistics, translation-interpretation, language teaching, public speaking, etc.), and for professional practice after graduation (e.g office staff, translators, interpreters, language teachers, etc.) The rating and marking are usually performed by the teacher-in-charge of Listening-

Speaking classes What to be assessed (the content) and how to assess it (the method) are usually predetermined by the English faculty and relatively identical across classes of the same institutions Nevertheless, most Vietnamese universities do not require non-English majors to take school-based speaking tests Only about 65% of non-English majors

understand the importance of learning English Students are not involved in serious study

of English, but learn it merely for form’s sake to obtain a programme completion certificate (Anh Tu, 2013) This limitation would affect opportunities and competitive advantages for career development, income, and higher education End-of-term English tests for non-English majors mainly focused on grammar accuracy, vocabulary range, reading

comprehension, sentence writing or short paragraph writing Ignorance of assessing

speaking skills leads to the possibility that “teachers may simply not teach certain important skills if they are not in the test” (Weir, 2005, p 18) Explicable reasons for institutions not

to administer oral tests are numerous: students’ weakness at speaking as a result of English learning from high schools as presented above; difficulties in managing large classes of 50

or more students (Hoang, 2010); time-consuming administration due to insufficient

facilities for computer-based testing (CBT); teachers’ lack of training for oral assessment (Le Nhi, 2008); avoidance of raters’ subjectivity in scoring and candidates’ complaints about test results when speaking test performances are not audio-recorded, etc

Certification of English proficiency, including Speaking skills, is a real need in today’s Vietnam’s changing society, not only for academic acknowledgement, but also for job applications, career promotion, and higher education admission However, domestic certification of English qualifications is not widely recognised For example, the Vietnam National English certificates corresponding to three levels: Elementary (Level A),

Intermediate (Level B), and Advanced (Level C) applied since 1993, are not reliable any more because extensive test administration and certificate issuance have resulted in

reduction of quality of these national certificates (Nhu Lich & Ha Anh, 2014; Ngoc Ha & Tran Quynh, 2014) Further, they are no longer compatible with the CEFR benchmark that the MOET’s guidelines deploy across institutions (Appendix G.3), and so cannot be

competitive with other international English certificates such as TOEIC, TOEFL, IELTS, etc Another example is the recent emergence of the Vietnam National University HCMC English Proficiency Test (VNU-EPT), which can be regarded as an in-time response to the need of certifying the English competence of undergraduate and graduate students in and outside the Vietnam National University The VNU-EPT proves to be a good substitute for the level A-B-C certificates in that it is low-fee (suitable for most students) and contributes

to the English standardisation of graduate outcomes constructed on the 6-level CEFR Nevertheless, there are strategic challenges making VNU-EFT popular with test users and