Learner Perceptions of L2 Pronunciation Instruction: A Comparative Study

Abstract

This study aims to better understand Vietnamese learners' views of potential learning problems and the impact of teacher-related factors on English pronunciation learning. It examines how more successful and less successful learners perceive pronunciation instruction and the problems they face when learning the L2 phonological system. In this mixed-methods study, 48 first-year English major students (26 strong and 22 weak) at a Vietnamese university were first polled about their learning difficulties and their perceptions of the pronunciation instruction available to them. Four strong and four weak students (chosen using both human raters and a computer-aided rating scheme) then partook in semi-structured interviews. The findings reveal that both groups differ in terms of the difficulties they face, their learning goals and language models, and their assessment of the teaching content and techniques. Several important implications were made regarding the learning goals, the status of non-native teachers, and the discrepancy between learner perception and teacher cognition of pronunciation instruction.

Introduction

It is beyond dispute that developing good pronunciation is crucial to successful L2 learning (Dickerson, 2019; Sugimoto & Uchida, 2018; Yates, 2017) 1 , 2 , 3 . However, until recently, compared with other fields of second language acquisition (SLA), not much has been understood about how L2 pronunciation can be taught and learnt effectively, and more importantly, learners’ perspectives on L2 pronunciation related issues still have considerably low visibility in research.

The current research aimed to identify what stronger and weaker Vietnamese learners perceive to be their learning difficulties as well as to understand how these two groups of learners evaluate English pronunciation instruction. The instruction investigated in the study involves the learning goals, the language models available to the students, the teaching focus, and the techniques used by the teacher. In a context where the learners are at the same age, speak the same L1, possess a relatively similar L2 proficiency, and receive the same instruction, there must be some other factors that might contribute to the different levels of success in English pronunciation learning. This insightful understanding is expected to bring pronunciation instruction closer to learners’ needs.

Research questions

To achieve the aforementioned objectives, the study addresses the following two research questions:

1. What do strong and weak learners view as their difficulties in learning English pronunciation?

2. How do these two groups of learners perceive the pronunciation instruction provided to them in class?

Literature review

Potential difficulties in learning English pronunciation

According to Isaacs and Trofimovich (2017), pronunciation encompases segmental features (individual sounds) and supra-segmental features (stress, rhythm and intonation). Gilakjani and Ahmadi (2011) 4 claim that many second language learners have major difficulties with pronunciation even after a long time of learning the language. Researchers and teachers have attempted to predict and analyse areas of difficulty utilizing contrastive analysis or error analysis so that appropriate remedies can be made, and learning can be facilitated. From a relatively old-fashioned perspective, Hockett (1950) acknowledges two sources of learning difficulty: the habits of pronouncing L1 sounds and the habits of hearing. The former, to some extent, reflects the role of L1 transfer while the latter recognizes the importance of listening skill – perception - in L2 pronunciation learning.

More recent researchers have identified other areas where learners may encounter problems. Cenoz and Lecumberri (1999) claim that learners also make errors when they apply communication strategies such as overgeneralization or approximation. Gilakjani and Ahmadi (2011) 4 views the issue from a cognitive perspective, explaining that L2 learners have problems because they need to reconceptualize the patterns they have internalized for the L1 sound system, rearranging them or even forming new categories for the L2 system.

As regards which component of the phonological system – segmentals or suprasegmentals - may cause more trouble for learners, Derwing and Rossiter (2002) 5 claim that little research has been dedicated to finding out what learners perceive to be difficult in learning or what they believe to be the best ways to overcome the hindrances. To fill the gap, they interviewed 100 students about the areas of difficulty in learning English pronunciation and found that the vast majority of the problems identified by the respondents were segmental.

Some researchers have studied potential areas of difficulty for specific groups of learners. For example, several studies have been conducted on common problems Vietnamese learners of English face in learning pronunciation. Those include omission of sounds (Ha, 2005) 6 , shortening of sounds and distinction of long and short vowels (Nguyen, 1998) 7 . Tran (2019) 8 reported the same error types in her study on EFL students at a university – omission of final sounds and mispronunciation of vowels. She also saw her students struggling with consonant clusters – one of the most common error types found among Vietnamese learners. Sharing the same research interest, Tran and Nguyen (2022) 9 employed a pronunciation test to investigate how 39 university EFL learners pronounced this feature. The results showed that the types of error depended on the types of clusters and there was a tendency to simplify the complex clusters by deleting one or more consonants in the group.

Learner perceptions of L2 pronunciation instruction

In his study, Alghazo (2015) 10 reported that the students were dissatisfied with both the amount of instruction given and the balance of features covered. Although these students were at a low proficiency level (under the intermediate level, as mentioned by the author), they seemed to know clearly what worked and what did not work for them in terms of course design, teaching styles and language of instruction. Hence, their perspectives are indeed helpful in determining the teaching approach.

In 2004 11 , Pardo conducted a comparison between teachers’ and learners’ attitudes towards the impact of pronunciation teaching and found that while many teachers were unsure of the effectiveness of instruction, the learners tended to consider it very beneficial. To add to this, Henrichsen and Stephens (2015) 12 reported that even though there was a lack of progress in their performance, the learners still found instruction beneficial in terms of the increased awareness, heightened confidence, improved listening skills, and gains in pronunciation learning strategies. They appreciated the benefits of instruction that are likely to extend beyond the end of the course: their confidence, self-awareness, motivation, and strategies for continuing improvement. These are the key to success in L2 pronunciation learning in the long term.

Teacher-related factors and their effects on L2 pronunciation learning

Learning goals

According to Crystal (2012) 13 , approximately one-fourth of the world’s population can use English with only a small proportion of them being native speakers. Ketabi (2015) 14 , Levis (2005) 15 , Moghaddam (2012) 16 and Setter (2008) 17 claimed that in an age when English functions as the basic channel of international communication, native-like pronunciation seems to be unrealistic, unnecessary, and undesirable. Therefore, the current goal in pronunciation instruction should be “intelligibility”, or “acceptable pronunciation” (Gilakjani, 2012) 18 . Murphy (2014) 19 even added that it is unfair and unethical for teachers to make their learners believe that they will ever be able to achieve such a goal.

Pronunciation models

There are several reasons why native speakers should not be considered as the only models for pronunciation teaching: the need of practical knowledge of both L1 and L2 phonetics (Walker, 2001) 20 , the intelligible nature of many English varieties (Jenkins, 2000) 21 , the expeted preparation to teach students at various language levels (Moszynska, 2007, as cited in Setter, 2008) 17 , and the popularity of non-native English language teachers (Miller, 2009) 22 .

There are also good reasons why non-native teachers of English should be included as models for pronunciation instruction: the presence of more aspirational, accessible and relevant models to learners’ needs (Murphy, 2014) 19 and the ability to support learners using their knowledge of both L1 and L2 phonological systems and their own experience in learning (Moghaddam et al., 2012) 16 . Recently, Levis, Sonsaat, Link, and Barriuso (2016) 23 conducted a study on how native and nonnative teachers affect L2 learners’ performance. The results postulated that instruction on pronunciation skills is more dependent on knowledgeable teaching practices than on nativeness.

Teaching focus

Until recently, there has been a long-standing debate over which should be taught in the pronunciation class, segmentals or suprasegmentals. In fact, findings from research on this controversy are divided, with some supporting the teaching of segmentals, while others advocating instruction on suprasegmantals. Since the beginning of the new millennium, there has been a more balanced view of the issue (Ketabi & Saeb, 2015) 14 when it is acknowledged that both segmental and suprasegmental features can harm intelligibility. The question now is no longer whether to teach segmentals or suprasegmentals, rather, what features to teach so that learners can communicate effectively (Ketabi & Saeb, 2015; Levis, 2005; Moghaddam et al., 2012) 15 , 14 , 16 .

Teaching techniques

Pronunciation teaching techniques can be classified into more traditional categories (Celce-Murcia et al., 2010) 24 and more innovative ones (Rogerson-Revell, 2011) 25 , both of which can be used for teaching different aspects of pronunciation such as sounds, syllables, rhythm, connected speech, and intonation. Lear (2011) 26 admitted that “there is a significant disparity between learner and teacher beliefs about the use of language learning activities” (p.131), but while a large body of research has been done from the point of view of the teachers, learners have rarely been asked for their opinions about what they find useful or what they often use after class for further practice. For example, the Pronunciation in Second Language Learning & Teaching Annual Conferences have taken place since 2009, producing nearly 150 articles published in the conference proceedings (Levis et al., 2016) 27 . Among those, only about a dozen were dedicated to the learner’s perspective on L2 pronunciation instruction.

Methodology

Design

A mixed-methods research design was employed, integrating questionnaire surveys and semi-structured interviews. In the quantitative phase, 48 first-year English major students (22 strong learners and 26 weak ones) at a university in Vietnam were surveyed for their perspectives on pronunciation learning problems and the role of instruction. Then, in the qualitative phase, four successful learners of pronunciation and four others who were struggling in their study were carefully selected to participate in the semi-structured interviews.

Participants

The population of the study included 167 first-year English majors at a university in Vietnam. At the time of the study, they were enrolled in a compulsory pronunciation course. In the quantitative phase, the researcher employed intensity sampling (Ary et al., 2014; Teddlie & Tashakkori, 2009) 28 , 29 to recruit the participants. Based on the results of the pronunciation course’s mid-term test supplied by the faculty, 22 students with the highest scores (≥8.0/10) and 26 others with the lowest scores (<5.0/10) were asked to complete a questionnaire. The numbers of strong and weak students were unequal as a result of the students’ performances on the mid-term test. Then, criterion sampling was utilized to select the interviewees for the qualitative phase. Invitations were sent to all 48 learners, but only 24 of them (15 strong and 9 weak) agreed to join. More strong students were willing to continue their participation in the study than the weak ones, possibly because they found it more comfortable talking about their learning progress and achievements. They were asked to record their voices while working on some pronunciation tasks and the recordings were evaluated by both a computer-aided system and human raters. The procedure is described below.

First, the participants’ pronunciation was assessed through a computer-aided rating system using both ASR technology and acoustic analysis. They were recorded reading aloud a diagnostic passage (Prator & Robinett, 1985) 30 and 12 sentences (For a copy of these materials, see the Appendix 1). The passage was used mainly for the assessment of segmental features. The recordings were filtered to remove the noise and then played to Dictation – Online Speech Recognition (https://dictation.io/), a computer application that internally uses the built-in speech recognition engine of Google Chrome to transform one’s voice into digital text. To assess the students’ performances on supra-segmentals, the pitch contours of the recorded 12 sentences were analysed using PRAAT, a computer software package for the scientific analysis of speech in phonetics. These pitch contours were then rated against those of native models. Moreover, to ensure the reliability of the scores given, this part was marked by two raters and the results were discussed before the final scores could be decided.

Second, the participants were asked to respond to some questions (See Appendix 2 for examples of the questions). After that, their recordings were rated by two native speaker (NS) and one non-native speaker (NNS) teachers of English. Inter-rater consistency was measured using Cohen’s Kappa coefficient. The average pairwise percent agreement for the 24 participants’ performance was 83.3% while the average pairwise Cohen’s Kappa was .61, which is considered to be substantial (Landis & Koch, 1977) 31 .

The human raters’ opinions and the results produced by the computer-aided rating system, were compared and then combined. Finally, the students in each group were ranked according to their total scores, and the interviewees were selected based on that order, starting from the highest for the strong learners' group and the lowest for the weak learners' group. Eventually, the number of interviews stopped at eight, with four strong and four weak learners, whose performances remained consistent throughout. The sampling procedure is summarized in Figure 1 .

Figure 1 . Sampling procedure

Instruments

There were three parts in the questionnaire: the first one addressed the difficulties that the surveyed students encountered during the pronunciation course they were attending, the second aimed to investigate their perceptions of the instruction they received during the course, and the last one helped to collect demographic information on the participants.

The semi-structured interviews were used as an instrument for the collection of qualitative data. There were two parts in the protocol, with the first one further exploring the causes of L2 pronunciation learning difficulties and the second one devoted to gaining better insights into the impact of pronunciation instruction on the learning process. (For a copy of the questionnaire, see Appendix 3)

Results

Potential learning difficulties

There are eight items in this subsection of the survey, with four related to segmentals and the other four concerning suprasegmentals. The participants were asked to rate the difficulty level in learning these features on a scale from 1 (Easy) to 5 (Difficult). The results show that both strong and weak learners appeared to encounter the same problems in learning, the biggest three of which are intonation, sentence stress and consonant clusters. See Table 1 below for more detailed statistics.

Table 1 Potential learning difficulties – Means and Standard Deviations

In the interviews, the researcher asked the interviewees to explain why they found these features difficult to learn. Remarkably, both strong and weak learners complained about pronunciation instruction at high school, saying that it was either hardly taught or taught in ineffective ways. For example, Weak Learner 2 gave some detailed description of how she was taught to produce intonation and consonant clusters at high school:

The teacher did not give much practice. If there was some, then she did not correct our intonation. She said just to say it correctly, just repeat it, just say it, as long as it is clear enough to hear, then that’s it.

… In the past, I ... in general, I just listened to however the teacher said. She did not analyze this, like there are 3 sounds, for example. She just said “scream”, then I just repeated after her. (W2)

Effects of pronunciation instruction on learning

Learning goals

The six items in this section of the questionnaire were aimed at determining which of the two goals in learning English pronunciation – nativeness versus intelligibility – was more common among these learners (Q9, Q11, &Q13) and the effect of the teacher on such a goal (Q10, Q12, &Q14). If a participant is uncertain about any item, he or she can opt for Don’t Know ( D/K ) instead of Yes or No . A comparison between the results of the strong and weak learners’ groups has led to two remarkable differences, as shown in Table 2 . First, more learners in the former group than in the latter group aimed at nativeness (Q9: 77.3% vs 61.5% and Q13: 77.3% vs 69.2%). Second, the weak learners tended to be more heavily affected by their teachers than their counterparts in aiming to sound native-like (Q10: 69.2% vs 54.5%).

Table 2 Learning goal

When the two groups of interviewees are compared, two differences, though not very obvious, were seen. First, while all the strong learners insisted on nativeness as their goal, Weak Learner 4 admitted that despite a preference for a native accent, she knew it was impossible to achieve it, and so was only aiming at being understood by other people. Second, the strong learners gave a variety of reasons for their answers, which are quite personal and unique, such as having a good feeling when speaking like native people (S3), or wanting to be like their idols, who speak English with a native-like accent (S4). Whereas, the weak learners mentioned two common reasons: better proficiency (W1, W2) and confidence in communication (W1, W3). One student explained:

I want to be more confident in communication. Back then, I was in middle school trying to communicate with foreigners. I was very afraid (shy), partly because my pronunciation was not good. (W3)

Language models

The participants were asked to state how much they agreed or disagreed with four statements concerning the language model in a pronunciation class. The first two items in the section were intended to find out whether they would like to study with a native (NS) (Q15) or non-native (NNS) model (Q16) while the last two looked into their attitudes towards the pronunciation teachers’ knowledge of both the L1 and L2 (Q17) as well as their shared learning experience (Q18). There is hardly any difference in the responses provided by the two groups of participants. Both groups still valued NS teachers over NNS ones, but they did acknowledge the benefits of studying with a NNS teacher. The majority of the respondents agreed that one of the strengths of non-native teachers is their knowledge of both English and Vietnamese while even higher percentages admitted that non-native teachers can be good models because they can share their learning experience with the students.

Findings from the interviews revealed two differences. First, only the weak learners expressed doubts about the accuracy of the non-native teacher’s pronunciation; the strong learners just reported feeling bored. Second, while most strong learners named a benefit of working with a non-native teacher, two weak ones (W1 and W2) did not and another (W4) only appreciated the possibility of using the L1, which seems to be irrelevant in an L2 pronunciation classroom.

Teaching focus

In this section, the respondents were requested to indicate the amounts of instruction that their teachers provided for the eight pronunciations aspects (Q19 – Q26): vowels, consonants, consonant clusters, final sounds, word stress, sentence stress, intonation, and linking. They rated the amounts based on a scale from 1 ( Little ) to 5 ( A lot ). The results also revealed that the two groups of strong and weak learners largely agreed with each other, with word stress, sentence stress and and vowels reported to receive the greatest amount of instruction (See Table 3 ).

Table 3 Teaching focus

A comparison of the two groups of interviewees uncovers two differences. Firstly, while most strong learners attributed the teacher’s focus on word stress to a lack of understanding of students’ needs, the four weak learners were inconsistent, giving a variety of explanations ranging from the teacher’s not understanding what the students need (W4), or making a decision based on students’ performance (W2 and W3) to teaching what is tested (W1). Secondly, while all the strong learners complained about not receiving the instruction they need, which led to unwilling self-study outside class, only two weak learners shared the same criticism. The other two (W3, and W4) found it acceptable for the teacher to so do, saying that it did not harm their learning.

Teaching techniques

The respondents were asked to rank the techniques, tools, and activities that their teachers used in the classroom according to their usefulness in helping them improve their pronunciation. They were also reminded that if a certain activity/tool was not used in their class, they should choose N/A (Not Applicable). Table 4 below shows the mean scores and the percentages of respondents selecting 4 and 5 combined for all items.

The findings show that both groups perceived minimal pair drills (Q30: M=4.64 & 4.12) and IPA practice (Q33: M=4.36 & 4.42) to be the most useful techniques. In contrast, the least useful one is using clapping and tapping (Q31: M=2.36 & 2.54). A closer look at the mean scores indicate the weaker’s preference for the teacher’s use of concept explanations, visual aids, songs and poems, IPA practice, role play, group/pair work, films and dictation exercises while the stronger seem to favor repetition, minimal pairs, games and Internet materials. Yet, the biggest differences can be found in two items: Q27 and Q30, when the percentages of respondents rating the techniques at 4 and 5 were combined. For one thing, weak learners found the teacher’s explanation of theoretical concepts more valuable than strong learners (57.7% vs 27.2%). For another thing, doing minimal pair drills seemed to be less useful for them than for their strong counterparts (73.1% vs 100%).

Table 4 Teaching techniques

Discussion

Difficulties encountered in learning by strong and weak learners

The results from the survey show that both strong and weak learners find it more difficult to deal with supra-segmental features, especially intonation and sentence stress. This contradicts what Derwing and Rossiter (2002) 5 found in their study. Yet, no conclusion can be made from this comparison. The teachers in the current research might have focused more on teaching supra-segmental features, especially word stress, so their students might have encountered more difficulties learning them due to greater amounts of exposure to the features. Derwing and Rossiter, however, provided no information about the focus of instruction that their subjects received. Therefore, their subjects may have spent more time learning segmentals and thus may have had more problems dealing with them.

In addition, the findings from the interviews reveal three major reasons why both groups of respondents find intonation, sentences stress and consonant clusters difficult to learn: the complex nature of these features, the influence of the L1 and, the most important of all, the pronunciation instruction that they received at high school. First, it seems to be true that some features are really difficult for Vietnamese learners to acquire, for example, the fricatives s, z, ʃ, ʒ, θ, and ð and the affricates dʒ and tʃ. This finding is echoed by Ha’s (2005) 6 study, which claims that the absence of the features ʃ, ʒ, θ, ð, dʒ and tʃ in the Vietnamese sound inventory, the misperception of sound aspiration, and the inability to distinguish between aspiration and friction are the causes of their difficulty in learning these sounds.

For the second cause given by the interviewees, Gilakjani and Ahmadi (2011) 4 explained that L2 learners have to reconceptualise the patterns they have internalized for the L1 system. In this case, for example, Vietnamese learners of English need to form new categories for the English sounds θ, ð, dʒ and tʃ, which do not exist in the Vietnamese phonological system. This reconceptualisation is obviously not an easy task for them to perform.

Finally, the majority of the interviewees considered the way English pronunciation was taught in high school as a main cause of their current learning difficulties. What can be recognized from their narratives is a lack of practice and feedback, the use of inappropriate methods, or even the absence of pedagogy (when the teacher was reported to just tell the students “ just to say it correctly, just repeat it, just say it” while teaching intonation). This is, however, not surprising in the context of Vietnam at the moment, when the Ministry of Education and Training reported that only 69% of English teachers nationwide are linguistically qualified (H. Nguyen, 2019) 32 , with many of them struggling with speaking skills in general and pronunciation in particular.

Strong and weak learners perceive pronunciation instruction differently.

It is undeniable that L2 teachers themselves and the instruction they deliver have certain impacts on learners and their learning. The current study has found that these influences are dissimilar for strong and weak learners, which may contribute to the different levels of achievement. Firstly, the survey results show that a native-like accent seems to be more relevant to strong learners than weak ones while the latter group appear to be more affected by their teacher. In other words, the successful learners seem to know what they want, set it as their goal, plan for it, and are motivated enough to work hard towards it. In contrast, the unsuccessful ones may only try to work towards what the teacher sets out for them without knowing whether it is achievable or not and thus can be demotivated once little progress can be seen.

Secondly, the weak learners show a lack of trust in their non-native speaker teacher’s pronunciation, neither do they recognise any benefits from learning with such a teacher. Their stronger counterparts, in contrast, still find it beneficial, in one way or another, to study with a non-native teacher. This might indicate a heavy dependence on the native language model among the less unsuccessful learners. This could result from the belief in nativeness as a proper learning goal and a lack of guidance from the L2 teacher regarding the legitimacy of intelligibility as an alternative goal in learning pronunciation. The successful learners seem to be less reliant on the teacher as a language model.

Thirdly, the interview results show that the weak learners perceive what the teacher focuses on in the classroom as appropriate and reasonable while the strong ones are more critical of what is taught. They actively reflect on their own learning and then expect the teacher to respond more closely to their needs. In other words, for the successful learners, there is an element of choice and relevance (Tominaga, 2009) 33 regarding what should be taught and learnt. In contrast, the less successful learners, once again, appear to be more reliant on the teacher, accepting what is provided without much questioning.

Finally, the findings from the survey indicate that strong and weak learners value the teaching techniques employed by the teacher differently. This demonstrates the disparity between learner percetion and teacher’s cognition of language learning activities (Lear, 2011) 26 . To be more specific, in the current study, what teachers assume to be harder, such as theoretical concepts, is actually preferred by weaker students, while simpler activities, like minimal pair drills might not work for them.

Implications and conclusion

The first implication the researcher would like to make is based on the finding that strong learners tend to be more independent of their teachers. Teachers can generate and facilitate such independent learning in several ways, especially for weaker learners. Teachers should make students aware of the benefits of actively selecting appropriate learning methods and the necessity of continuing study outside the classroom and after the course. For example, they may organize class discussions in which students are asked to suggest different ways they have used or think they can use to learn certain features more effectively or ask students to keep a diary in which they reflect on how certain activities have worked for them. Additionally, Vietnamese learners need a lot of know-how. They need to know how to plan their study based on their own needs, carry out their learning using appropriate strategies, techniques, and tools, monitor their progress, and assess their performance. For instance, teachers can inform their students of the potential of using an ASR dictation program such as Google Voice Typing for assessing their own production, especially segmentals, for getting feedback and for practice outside the classroom as this program “may now rival human listeners particularly for free speech” (McCrocklin et al., 2019, p. 197) 34 .

Another implication is as Vietnamese students may not trust their English pronunciation teachers because of their non-native accents, the teachers need to be reminded to constantly improve their own pronunciation to win their learners’ trust and to foster understanding in communication (Gilakjani, 2012) 18 . In an age when learners have easier access to native accents, it is necessary that the non-native teachers of English have accurate production of both segmental and suprasegmental features and be comfortably intelligible. As a result, they can be confident when talking to their students and their students can also feel confident about learning L2 pronunciation with a non-native speaker teacher.

All in all, this paper hopes to have provided L2 teachers and researchers with useful information on how a specific group of learners are learning an L2 phonological system. More importantly, it has, in one way or another, listened to learners’ voices, exposing their views to teachers, making them reconsider what they are doing in their classes. It might also have brought research work closer to the real classroom, providing more practical ideas for teachers. In a nutshell, it is expected that this research has contributed to the empowerment of L2 learners, helping them to achieve more success in learning L2 pronunciation.

Biodata

Dao, Nguyen Anh Duc has been teaching English for more than twenty years and is currently a lecturer at the School of Languages– International University (Vietnam National University – Ho Chi Minh City), Vietnam. She also works as a local trainer for Cambridge University Press. She earned her Ph.D. from the University of Nottingham. She holds an MA in TESOL Studies from the University of Queensland, and an MBA from Bolton University. Her research interests are teaching methodology, L2 pronunciation teaching and learning, and teacher education.

Appendices

Appendix 1

A diagnostic passage

When a student from another country comes to study in the United States, he has to find out for himself the answers to many questions, and he has many problems to think about. Where should he live? Would it be better if he looked for a private room off campus, or if he stayed in a dormitory? Should he spend all of his time just studying? Shouldn’t he try to take advantage of the many social and cultural activities which are offered? At first it is not easy for him to be casual in dress, informal in manner, and confident in speech. Little by little he learns what kind of clothing is usually worn here to be casually dressed for classes. He also learns to choose the language and customs that are appropriate for informal situations. Finally he begins to feel sure of himself. But let me tell you, my friend, this long-awaited feeling doesn’t develop suddenly, does it? All of this takes will power.

(Prator & Robinett, 1985)

The 12 sentences

(In 2-line dialogues, students will read B’s lines only.)

1. Eat it with some cheese!

2. What do you think?

3. She’s given him some money.

4. Excuse me, I think you’re in my seat.

5. Do you want a super burger or a regular burger?

6. I know your parents live here, but were they born here?

7. A: We’ve won a holiday for two in Jamaica!

B: Brilliant!

8. A: I’ve crashed the car again!

B: Well done!

9. They took his computer, television, video, CD player and all his CDs.

10. A: Do you have a nice flat?

B: Yes, a very nice flat.

11. A: Excuse me, can you help us?

B: Yes?

12. A: OK, well go across the bridge and turn right.

B: Turn right?

Appendix 2

Prompts :

1. Please introduce yourself.

2. Tell me about your family.

3. Tell me about something you love doing in your free time.

4. What did you do on your last holiday?

Appendix 3

Figure 2

Figure 2 . Appendix 3

Acknowledgments

None.

Ethics approval and consent to participate

Not applicable.

Competing interests

The authors declare that they have no competing interests.