The Implementation and Challenges of Using Role-Play to Improve Chinese University Students’ English Communicative Competence

English is taught as a compulsory course in Chinese universities and it is necessary to find feasible ways to improve students’ English ability to communicate. Role-play is believed as an effective way to enhance second language learners’ communicative competence and it is also supposed to be able to be effective in Chinese universities EFL classes. The real effect, however, relies on the actual implementation in classes. If the teachers inappropriately adopt role-play, it is hard to be useful. Due to the lack of research paying attention to this issue, it is not clear how Chinese university English teachers use role-play and whether it is effective. For this reason, a case study was conducted, with a Chinese university EFL teacher taking part in the interviews. The results show that participants attempt to maximise the effectiveness of role-play to improve students’ English communicative competence by practising comprehensive language skills, linguistic knowledge, and workplace topics. Nonetheless, some issues, namely script preparation, students’ proficiency difference and class size are identified. These may challenge the actual effectiveness of role-play. Possible solutions, including selecting role-play according to students’ ability, considering individuals’ condition, mixed-ability grouping and small class size, are suggested thereafter.


Introduction
Since English has become the dominant language of global communication, it is taught compulsorily to students of all majors in Chinese universities (Ministry of Education (MOE), 2017). For the sake of fluent communication in students' future career, the English as a Foreign Language (EFL) courses in universities are expected to enable students to achieve adequate English communicative capacity. Nevertheless, students' capacity is not always as satisfactory as expected (Gan, 2013;Li, 2018). It is necessary for educators and teachers to identify effective ways to enhance students' communication competence.
Chinese national curriculum emphasises students' communicative competence in English. It is required by MOE (2017) that university students must be able to understand and conduct a certain length of everyday conversation, such as personal experiences and daily topics. The further requirement is that students '能用英语就一般性话题进行较流 利的会话、进行书面表达' (can communicate both orally and in writing about common topics) and the highest one is students should '能用英语 在专业 领域 流利进 行 对话、 能就 专业话 题 撰写报 告或 论文 '(be capable of communicating professionally and fluently in the workplace) (MOE, 2017, p. 5). However, the facts are not like what is required by MOE.
Chinese university students encounter four problems when communicating in English. The first one is their inadequate speaking and listening abilities. Zhu, Liu, and Zhu (2017) point out that students just listen to the teacher quietly and seldom speak English in most university EFL classrooms (Zhu et al., 2017). Ironically, although they are listening most of the time, their English listening skill is still deficient (Zhu et al., 2017). Students cannot communicate orally in a fluent manner without adequate listening and speaking skills. Second, Li (2018) observes that writing is ignored in university EFL classrooms, meaning students can hardly write adequate English, which hampers students' written communication. Third, students are blamed to be deficient in linguistic knowledge. In the studies of Gan (2013) and Zhao (2016), the majority of Chinese university students report difficulties in grammar, vocabulary, coherence and other linguistic knowledge when speaking English. Zhu et al. (2017) further point out that the written test also frightens students due to their lack of linguistic knowledge. Unsatisfactory speaking and writing skills again mean low communicative competence. Finally, the lack of familiar topics worries students. They may not know what to discuss when communicating with people from other countries (Li, 2018). Fan. Exchanges 2020 8(1), pp. [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18] Role-play is defined as an activity involving participants in acting out new roles to learn and practise new knowledge and skills, including linguistic knowledge and social manners (Harper-Whalen & Morris, 2005). There are three types of role-play in second language classrooms for learners of different levels. The first one is the full-scripted role-play in which participants are provided with a detailed script telling them what to say exactly (Harper-Whalen & Morris, 2005). It is suitable for low-level learners to understand and memorise the linguistic knowledge (Krebt, 2017). In the second type, semi-scripted role-play, guidance about the role and what to say is presented to intermediate learners, which offers them support as well as the requirement for further knowledge and skills, and thus they can attempt to move on to the higher level (Cho, 2015;Harper-Whalen & Morris, 2005). The third type, non-scripted role-play, only provides information of context to high-level learners so that they have to create original conversation, thereby fully practising the target language Role-play is proposed to have a significant impact on second language acquisition in the following ways. For one thing, students have the opportunities to practise their comprehensive language skills in the roleplay if the script preparation stage is added (Carson, 2012;Edstrom, 2013;Lin, 2009;Tran, 2016). Students need to read and write the scripts when preparing for the role-play and then listen and speak during the performance, which trains their overall language skills (Tran, 2016). Additionally, several studies support the position that role-play familiarises learners with plentiful linguistic knowledge, including vocabulary, grammar (e.g. Cho, 2015; Lin, 2009). To perform a role-play, learners have to review and use the linguistic knowledge they have learnt previously (Lin, 2009). If there is no complete script, they need to acquire new knowledge in order to compose the script. In the research of Cho (2015), both semi-scripted and non-scripted role-play have a positive impact on students' grammar and vocabulary improvement. Other than that, role-play provides learners with various contexts where they would be trained in targeted situations (Chaitanya & Ramana, 2013;Cho, 2015). A good example is the research of Robinson, Harvey, and Tseng (2016) where the authentic contexts enable learners to acquire and practise corresponding knowledge and manners, which is beneficial for their communicative competence.

The Aim of This Study
Considering the above, then, theoretically, role-plays should be effective in improving Chinese university students' English communicative capacity. First, the students need English listening, writing and speaking training, and role-play provides the opportunities for them (Tran, 2016;Zhu et al., 2017). Second, insufficient linguistic knowledge is viewed as a vital problem of Chinese university students' inadequate communicative ability (Gan, 2013;Zhu et al., 2017). Role-play can familiarise students with a lot of linguistic knowledge (Cho, 2015). Third, familiar topics are demanded for communication, and role-play is able to create contexts with plentiful topics (Cho, 2015). Notwithstanding, it is hard to say outright if role-play is effective in real classes in China, because the outcomes, to some extent, rely on the design and implementation of teachers, rather than on the perceived benefits of role plays. Considering the importance of teachers' use of role-play, some researchers explore its actual use in second language classrooms, but the number of such studies is comparatively low (see Table 1). The majority of research gives much attention to the outcome of role-play rather than its actual implementation. Having searched ProQuest and SAGE journals, the author found there were merely three theses investigating how teachers use role-play in primary and secondary EFL classes and almost none concentrated on teachers' use in universities. In addition, there is a limited amount of empirical data in the context of Chinese universities. Although there are abundant conceptual discussions about adopting role-play in Chinese university EFL classes, few empirical studies are collected in CNKI (China Academic Journal Database). Thus, it remains unclear how teachers use role-play in this context, not to mention what impacts the results of role-play. This article, therefore, aims to explore the following questions: • How is role-play used in the Chinese university EFL classes?
• What challenges the use of role-play in Chinese university EFL classes?
• How can role-play be used effectively in Chinese university EFL classes?

Method
A qualitative case study was adopted in this research as the case study focuses on an individual unit and enables multiple variables to be considered (Hancock, 2006;Yin, 2014). A male EFL teacher from a Chinese university took part in this case study. The documents, including the national curriculum released by MOE (2017) and the university's syllabus, were reviewed first for a better understanding of the contextual use of role-play because teaching had to align with the requirements of MOE and the university. Afterwards, an online semi-structured interview with the participant was conducted. During the interview, he was asked to recall and discuss the role-play activities he had conducted before. The reconstruction of events is one of the advantages of the interview (Bryman, 2016). The teacher's descriptions and explanations for the use of role-play reflected his perspectives and considerations. In the semistructured interview, the participant was allowed and welcomed to discuss much more than the questions listed. The researcher had to leave the space for unexpected information.
The question framework involved three aspects (see Appendix). The first aspect was the design of role-play, including the topics and types of roleplay as well as the types of lesson in which role-play was used and so forth. Tran (2016) and Lin (2009) investigated similar themes in different contexts, and their findings reflect the teachers' understandings and concerns of role-play, which is helpful for my research. The second aspect was the implementation of role-play, including procedure, classroom management, and dealing with students' linguistic errors and so on. These were what teachers might encounter when implementing an activity (Harmer, 2015). It is inevitable for a teacher to consider and address these issues, so they were included in the question list. The third aspect was other factors which probably affected the adoption of role-play, such as class size and students' English levels.
After data collection, the data set was sorted and analysed carefully. Yin (2014) regards the criteria for interpreting findings as an imperative part of a research design. In the interest of the best use of the data generated, several strategies were adopted for data arrangement and analysis.
The first strategy was cross-sectional indexing, using categories to establish a common index (Mason, 2002). The data were sorted into several themes. The categories were created according to the interview question list (see Appendix). Microsoft Office Excel was used for the data indexing and analysis. Computer Aided Qualitative Data Analysis Software (CAQDAS) is helpful for the researcher with more than a few texts and documents to index (Mason, 2002). In the current case study, there was a great deal of data in the form of texts which were disorderly because the participants discussed freely in the semi-structured interview. Thus, computer software was used to classify the data.
The second strategy was triangulation. It uses the data collected from multiple sources (Yin, 2014). Some evidence collected from diverse sources differed or overlapped, and it was interwoven and then interpreted. Findings based on several sources of evidence are more persuasive than those drawn from a single source (Hancock, 2006). When the data from different sources accorded, a conclusion can be confirmed. However, when the data was not compatible, it needed to be examined carefully.
The third strategy was case description combined with rival explanations. The case description strategy is to analyse the data according to a descriptive framework (Yin, 2014). The question list in the interview presented a framework, which ranged from the teachers' design of roleplay to the implementation, in order to describe the real use of role-play in the Chinese university. The descriptions involved alternative descriptions, namely rival explanations (Yin, 2014). During data collection, incompatible information arose, and it needed to be compared and discussed in depth. The alternatives, as well as the possible reasons for them, were identified and analysed in the case study because they were able to provide valuable information for the research. The differences were analysed and the reasons why these happened were taken into consideration. In a sense, this strategy helped in preventing the researcher's bias and stereotyping.

Findings
In this part, the national curriculum and university syllabus are reviewed first in order to provide relevant details of the context in which the participant adopted role-play. After that, the interview outcomes are presented.

Documents Review
The results of documents review are indexed into three categories. The identification of the categories refers to the themes in the interview results. Some of the themes are not mentioned in the documents while some are relevant, and the related information is included in this section. In addition, some themes are hard to split in the documents, and thus they are combined into one category.
The first one is various topics and requirements for students of different levels. MOE (2017) indicates three main topics, including general English, professional English and intercultural communication, and three levels of requirements, namely basic, advanced and proficient requirements. The basic requirement is applied to '英语高考成绩合格的学生' (low-level students who just get a passing grade in English in university examinations) (ibid, 2), and daily communication is the main topic for them. Vocational and cultural knowledge is introduced to intermediate and high-level students, to whom advanced and proficient requirements are applied respectively. These students '大学入学时英语已达到较好水平' (get higher scores in university examinations) (ibid, 2), and more difficult content is suitable for them to learn. Daily, vocational and cultural knowledge can be the topics of role-play in actual use.
The second category is class size control. MOE (2017)  curriculum resources for English teaching and learning) (ibid, 12). As per the university syllabus, teachers should make the most of the Internet to select and create excellent and up-to-date teaching materials. Therefore, they need to select the materials carefully and can employ some electronic devices or applications in their classes.

Semi-Structured Interview
In this section, interview findings are presented in line with the nine themes identified during the interviews (Table 2). These themes are identified based on the question framework (Appendix) with the crosssectional indexing strategy. Related categories in the documents review are also displayed in Table 2 to show the connection between them.
When designing role-play, the participant considered that the type of lesson where he used role-play most is the University English, which is a kind of lesson for the students who do not major in English ( sometimes confronted with, such as airport and hotel conversations, were suitable. As for high-level students, they were required to involve their occupational knowledge, such as the engineering terminologies, in the role-play. In addition, multiple types of role-play were adopted according to students' levels ( Table 2, item 3). Semi-scripted role-play was employed to the students who could only adapt the passages from the textbook or imitate the video of Family Album USA. As for the students who found this easy, the teacher applied non-scripted role-play to them, so they had to use their own words to create original dialogues. He believed: 'You cannot learn by rote. Only when you speak, the language belongs to you. The real confidence comes from speaking, not from scores.' As such, he encouraged students to speak themselves.

Topics
• At the beginning, most topics are shopping and travelling, then the airport and hotel. After that, when students are graduating, they will imitate the boss and managers.
• Various topics and requirements for different levels of students 3 Role-play types • They turn the sentences on the textbook into a dialogue. • For example, in a mock interview, they have to combine the vocational knowledge with English.

Procedure
• They need to prepare the script first.
• I usually ask them to record the roleplay. Then watch the recording later.

Classroom management
• You just ask them to comment on other's role-play then they'll listen earnestly.
6 Differentiated instruction within classrooms • Higher expectation, higher achievement • If they don't want to perform in the class, they can submit the video of roleplay.
• Various topics and requirements for different levels of students 7 Dealing with linguistic errors • Before the role-play, they need to give me the scripts…I'll check the language beforehand. 8 Class size • The largest number of students was 60 in one class. • Only 1-2 groups per week. Each group has 2 chances at most in a semester • Class size control Fan. Exchanges 2020 8(1), pp. 1-18

Technology
• I ask them to imitate the Family Album US. • They will add background music, special effects to the video.

• Selection and application of materials and technology
During implementation, the participant extended the role-play procedure by adding preliminary and subsequent stages ( Table 2, item 4). Before the class, the students needed to compose a script. When preparing, the students communicated and cooperated with each other, which developed their teamwork awareness and ability as required by MOE. During the role-play, the entire process was recorded, and afterwards the students reflected on their performance according to the recording. Second, the teacher managed the classroom by engaging the audience ( Table 2, item 5). When some students were doing the role-play, the others were required to comment on the performance later, which means they had to watch carefully. Third, in terms of differentiated instruction within classrooms, the teacher attempted to involve all students and provided help corresponding to their condition (Table 2, item 6). According to the participant, although the students at different levels were divided into Class I and Class II, differences still existed within the same class. He said: 'If you are too strict to low-level students, they may be too upset to do the role-play.' For those not confident in performing publicly, he encouraged them to make a role-play video which could be as entertaining or informative as they pleased. But high expectations were given to highlevel students. Fourth, the teacher emphasised that he would not interrupt students when linguistic errors occur during role-play (Table 2, item 7). He believed role-play was a complete performance, and it should not be interfered with. He chose to check the scripts before the role-play. In this way, the students memorised and practised the correct sentences.
When asked what may affect the effectiveness of role-play, the participant complained that there were too many students in a class (Table 2, item 8).
As mentioned before, the students in the University English were not English majors, which means a larger class size. The teacher explained that one English class normally had 50-60 students. English majors would be divided into two classes, no more than 30 students for each. But non-English majors in University English had to crowd into one class. He grouped students and required them to do role-play in turn, but each group had no more than 2 times to perform in one semester.
An unexpected finding was that the participant used multimedia technology to a great extent ( was selected as an additional teaching material by the teacher rather than the university. When making a video, the students were welcomed to use visual and audio effects. The participant said that creativity was emphasised by MOE and role-play was a good chance to inspire students. Video effects were interesting to students so they will take part in the roleplay more creatively.

Discussion
Research findings are analysed in this part. The use of role-play is discussed first, followed by the challenges identified in the interview. Subsequently, suggestions are given to deal with these challenges.

The Real Use of Role-Play
This section analyses the teacher's use of role-play and its outcomes. The national curriculum and unidentified syllabus are also involved to support the analysis.
Role-play was used to improve learners' communicative competence by extending the procedure to provide students with opportunities to practise their comprehensive language skills, including speaking, listening, reading and writing (Table 2, item 4). This is consistent with the findings of Tran (2016) and Chaitanya and Ramana (2013) that role-play is useful in improving multiple language skills. Role-play does contribute to learners' speaking development considerably, but speaking is not the only benefit of role-play (Krebt, 2017). Students read and write the scripts, and then listen and speak when performing (Tran, 2016). When they can use those skills better, they can communicate better as oral communication involves speaking and listening while written communication consists of reading and writing. MOE and the university highlight the significance of communicative competence as well as comprehensive language skills (MOE, 2017). The teachers' extension of role-play met the requirements and would be beneficial for the students' communicative competence.
Students' linguistic knowledge was acquired and corrected by the teacher checking students' script beforehand and asking students to reflect on the video after their performance (Table 2, item 4, 7 and 9). Edstrom (2013) highlights the significance of drafts in role-play because neither skill is thoroughly addressed without linguistic accuracy. Hence, the participant spent time on checking students' drafts to ensure they had obtained the linguistic knowledge well. The use of technology also helps students identify the linguistic errors they have made. Cohan and Honigsfeld (2013) argue that technology not only advances students' literacy development but also their agency. When making the videos, the students were actually reviewing their own performance. In this way, they would find some linguistic mistakes they did not realise before. Tran (2016) claims most students prefer their mistakes to be corrected at the end of role-play, so the fluency of communication is retained. Edstrom (2013) also suggests that follow-up activities can help students retain knowledge and skills. In addition, for the groups which finish their role-play before the others, follow-up activities could engage them, and thus facilitating classroom management (Ladousse, 1987, as cited in Lin, 2009). The participant's management of correction was helpful to reinforce his students' knowledge.
The students practised workplace communication in role-play (Table 2, item 1 and 2). Carson (2012) suggests that role-play enables students to rehearse for real-life situations. The students likely have to use English to communicate in the workplace in their future life. Proficient English communicative ability around professional topics is expected by employers (Yu, 2012). MOE also requires intermediate and high-level students to be capable of communicating professionally and fluently in the workplace (MOE, 2017). The participant's students were not English majors so they might have the terminology which was taught in Chinese but needed to be used in English. By simulating working situations, students can learn the terminology and apply them to deal with specific situations in the workplace (Robinson et al., 2016). The introduction of professional topics in the class enabled students to integrate their vocational and linguistic knowledge and may help them improve their competitiveness in the labour market.
All students were involved in role-play in the following ways. Firstly, semiscripted and non-scripted role-play were selected according to students' levels ( Table 2, item 3). Different types of role-play require different skills and proficiency, and provide different levels of comfort for learners, therefore it is necessary to find an appropriate role-play type for students (Harper-Whalen & Morris, 2005). By this means, it can be avoided that the students cannot fully participate in the role-play because it is too easy or too difficult for them. It is noteworthy that full-scripted role-play was not employed by the participants. A possible reason is that university students had already achieved a certain level of English as they had learnt English mandatorily from Grade 3 in primary schools (MOE, 2001). The participant chose these two types of role-play but abandoned the easiest full-scripted role-play, which means he had considered his students' current proficiency. Secondly, the students who were not performing were also engaged in role-play (Table 2, item 5). Similarly, Lin (2009) suggests the same solution to increase audience participation by asking students to evaluate their peer's performance after watching. In this way, students' attention is maintained (Lin, 2009). Thirdly, differentiated instruction was adopted in the class (Table 2, item 6). According to Tran (2016), weak students might need help at the beginning stage of a role-play, or they Fan. Exchanges 2020 8(1), pp. 1-18 cannot go further in their practice. In the current study, the teacher's support for low-level students could help them conduct role-play more smoothly.

Challenges
The students prepared scripts accurately before the class, whereas they lost the opportunity to practise natural and unexpected conversation ( Table 2, item 4 and 7). They learnt and used various vocabulary and grammar to compose the script, then recited the scripts corrected by the teacher, similar to the finding of Edstrom (2013). She asserts that students obtain nothing due to the lack of focus on linguistic accuracy, and thus their drafts should be reviewed and revised seriously (Edstrom, 2013). By this means, the students can acquire accurate linguistic knowledge well. Tran (2016), by contrast, defends this use of role-play because he believes that advanced preparation reduces students' exposure to natural communication. Cho (2015) contends that if the script is prepared, the language is not produced as in real life. Communication in real life happens naturally and unpredictably, and students should learn how to cope with it during a role-play (Tran, 2016). Indeed, the participant's students already knew what would happen next before the performance. They had no opportunity to learn how to face unexpected communication. The authenticity of their role-play was decreased. Carson (2012) compares the advantages of improvised and prepared role-play. Improvised role-play introduces the unpredictable language use in the real world to learners while prepared role-play helps students to focus on accuracy (Carson, 2012). It is hard to get all the benefits at the same time so the teacher needs to balance the uncertainty of natural communication and linguistic accuracy.
The students in one class had different English proficiency, so it was a challenge for the teacher to give appropriate instruction (Table 2, item 6 and 9). In the study of Lin (2009), improper role-play difficulty resulted in some students being unable to finish a role-play even with extra preparation time, but higher-level students found it too easy, so they started to chat once they finished. In the same role-play, it was hard for some students to prepare for more difficult role-play, whereas it was not easy to engage higher-level students (Lin, 2009). This suggests that the diverse difficulties of role-play should be given to different levels of students. The participant showed a good way. He had different requirements to students according to their English proficiency. High-level students had to meet his strictest requirement. Those who found it hard to complete the role-play were dealt with less strictly, or even allowed to make a video instead of a live show. At least, they can imitate the video of Family Album USA. By this means, all students were engaged and confronted with corresponding difficulties of role-play, which is supposed to be a useful solution for the issue pointed out by Lin (2009). However, there is one thing in doubt. Family Album USA was published in 1991, nearly 30 years ago (Family Album USA, n.d.). It is unclear whether the use of language in this video is still common today. According to the curriculum and syllabus, the teaching materials should be excellent and up to date (MOE, 2017). It would be better if the participant uses a newer material so that low-level students can keep pace with the language popular nowadays.
The drawback of large class size was underlined by the participant ( Table  2, item 8). He said the students in one group had few opportunities to perform because 'the largest number of students was 60'. He suggested that students would have more opportunities in a small class than in a large class. Although MOE requires a small size class (no more than 35) for speaking and writing, both the curriculum and syllabus set no limit of student numbers in comprehensive Class Ⅰ (MOE, 2017). However, because the classes where the participant employed role-play were not speaking or writing classes, the student number was far more than 35. As mentioned by the teachers, it is less controllable to do a role-play in such a large class. This finding is congruent with the conclusions of Tran (2016) that chaos in a crowded class is viewed as the biggest problem in conducting a role-play. On the other hand, the more students there are in one class, the more differences the teacher needs to cope with. This in turn aggravates the challenge of students' English proficiency difference.

Suggestions
It is suggested that teachers could select improvisational or prepared roleplay according to students' ability. Improvisational role-play with error correction after it could be employed to the students who already have some English linguistic knowledge and can communicate to some degree. By doing so, they have access to unexpected real-life communication (Carson, 2012). The linguistic accuracy could be addressed after role-play so the fluency of communication is retained (Tran, 2016). Nevertheless, for the students who are unable to maintain the flow of speech in English, it would be better if the teachers extend role-play with preparatory activities. Mardiningrum (2016) and Weiss (2007) observe the challenges brought by improvisational role-play to low-level students. Tran (2016) also admits that weak students need extra help at the beginning of a roleplay, or they cannot continue. Having prepared before, the learners could understand and practise the linguistic knowledge in the script, then their language would be more accurate during the role-play.
There are two suggestions for students' proficiency difference. Individuals' condition should be taken into consideration by the teachers when they design and implement role-play. The topics, types, preparation time and difficulty should be suitable for their students. Edstrom (2013) proposes that teachers must negotiate with students to identify and establish optimal conditions for teamwork. Several factors, such as personality, language proficiency and prior knowledge, may affect students' motivation and performance (Edstrom, 2013). Therefore, teachers should collaborate with students to find the best ways forward. For example, when the participant's students found it difficult to finish a role-play, he allowed them to take less important roles or to submit a video. They cooperated and found a comfortable way to complete the activity so that the students were able to participate and enjoy it. Furthermore, mixedability grouping could be employed in role-play. Although the students at different levels are already divided into different classes, the English levels of students in the same classroom are still different. In this case, Cornish (2009) suggests mixed-ability grouping as it provides opportunities for peer tutoring and more student choice and negotiation. As discussed above, the students need different levels of difficulty and preparation time, but this is difficult to accomplish in one class. If the teachers group students according to their proficiency, they could select the roles suitable for themselves and help each other finish the role-play.
A small class size is recommended for an effective role-play. As the participants complained, students had fewer opportunities to take part in a role-play in a large class than in a small class, and it was difficult to manage such a large number of students. Moreover, individual students may receive less attention from their teachers in a large class, and differentiated instruction becomes more difficult. Tran (2016) argues that a smaller teacher-student ratio is beneficial for English education, thereby calling on reduced class size. His appeal also applies to the context of Chinese universities where a large class size dominates EFL instruction.

Conclusion
This was a small-scale case study concentrating on a teacher's use of roleplay in Chinese university EFL classes. The findings show that the participant attempted to maximise the effectiveness of role-play to improve students' English communicative competence from multiple aspects, including comprehensive language skills, linguistic knowledge, and workplace communication. This is consistent with the studies which investigate the concerns of Chinese universities students in English Selecting role-play according to students' ability, considering individuals' condition, mixed-ability grouping and small class size were suggested to make role-play more effective.
Further research may involve both teachers and students. In this way, students' comments on their teachers' use of role-play would be collected. The opinions of teachers and students can be compared and triangulated. More than that, it is suggested that future researchers include observation in their studies. It would also help triangulation, and then the effectiveness and ways of employing role-play could be investigated profoundly.
Hongming Fan is a first-year PhD student in the Faculty of Education at Monash University. She also obtained her master's degree in TESOL at Monash University. Her research mainly focuses on the communicative competence in ESL/EFL learning, with a broad interest in language acquisition, higher education, curriculum and leadership.