Introduction

It has been well noted in Teaching English to Other Languages (TESOL) and related fields that willingness to communicate (WTC) plays a critical role in learning English as a second/foreign (L2/EFL) language [1]. While it is not difficult to understand how the newcomer status of beginner L2/EFL learners inhibits their WTC, it remains an intriguing issue why many advanced L2/EFL learners, such as postgraduates who have been studying English for several years, still exhibit unwillingness to communicate in English [2, 3]. Although the number of postgraduate students has increased tremendously around the world in recent decades, little attention has been paid to postgraduate students’ WTC in English and we know little about the factors that may shape postgraduate students’ experiences, beliefs, and practices regarding their WTC. Postgraduate students are different from university/college and secondary students due to “their English proficiency levels, learning experiences…ages, the nature of their study programs, personal beliefs, and other contextual factors” ( [4], p. 481), which may have a direct/indirect impact on their WTC in classrooms. An inquiry into postgraduate students’ WTC would advance the theoretical and pedagogical discussions on advanced learners’ WTC in English-mediated academic learning, thus shedding light on effective pedagogical supports for more effective English communication at the master’s level.

This paper intends to contribute to the research on WTC by exploring how Chinese master’s students’ WTC can be shaped by their perceptions of individual, contextual, and societal factors in content classroom contexts where English is used as the medium of instruction.

Willingness to Communicate: Definitions and Influencing Factors

In the 1990s, WTC started receiving Second Language Acquisition (SLA) researchers’ attention. MacIntyre and Charos [5] integrated SLA with communication and proposed that learners’ L2 WTC was not a simple transfer from the L1. They claimed that L2 WTC showed more situated and complicated features. They defined L2 WTC as a learner’s “readiness to enter into discourse at a particular time with a specific person or persons, using a L2” (p. 547). This initiated a situational view of WTC.

Since situational WTC was introduced, researchers have further explored its complexity. L2 WTC has been shown to relate to a range of demographical factors, socio-psychological, and contextual factors. The basic demographic factors include age, gender, and personality. Any change of WTC due to age and gender is closely bound with class orientation and the learning environment [6]. In the Chinese university setting, unlike in secondary school, English teaching becomes more communication oriented and university students that are psychologically immature tend to communicate less frequently [2]. Moreover, communication is strongly affected by personality traits. MacIntyre et al. [1] conceptualized five personality traits of WTC: extraversion, agreeableness, conscientiousness, emotional stability, and openness to new experiences. Later on, Wen and Clément [7] added risk-taking and tolerance of ambiguity for Chinese learners of English.

Socio-psychological and cognitive factors include communication apprehension, communicative competence, and self-confidence. Communication apprehension (CA) refers to the anxiety experienced during the process of communication. It was shown that a higher level of CA would lead to a lower level of WTC [8,9,10]. The relation between anxiety and interlocutors’ potential criticism, including teachers’ and peers’, is rather strong [11, 12]. Communicative competence is usually used as a measure of L2 proficiency. McCroskey and Richmond [13]) stated that self-perceived communicative competence (SPCC) was the self-evaluation of an individual’s ability to communicate properly in a given context. A higher level of perceived communicative competence combined with a lower level of anxiety will lead to greater WTC, which in turn generates more frequent communication in the L2 [14]. Self-confidence boosts learners’ willingness to speak in front of peers in a classroom—it moderates WTC by reducing anxiety and stirring up motivation to communicate [15, 16]. Motivation can also influence WTC indirectly through confidence [2]. Syed [11] and Syed and Kuzborska [12] found perceived opportunity was the most influential psychological factor and they also discussed the influence of emotion, perceived appearance, and pre-occupation.

The contextual factors include communication topic, opportunities of L2 communication, interactional context, teacher-related variables, and cultural value. Pawlak et al. [17] found that changes in learners’ WTC while engaged in conversation were influenced by the topics under discussion. Other researchers have found learners’ familiarity with topics and whether learners have any say in the selection of discussed topics could promote WTC during class time [18, 19]. For instance, [20] pointed out that content and language integrated learning (CLIL) has a potential of promoting students’ WTC.

Opportunities for L2 communication also have a role to play in learners’ WTC. Baker and MacIntyre [21] were interested in whether the experience of L2 immersion would have an effect on learners’ WTC; results of their study found immersion students did indeed possess a higher level of WTC that was directly attributed to the increased opportunities for L2 interactions, leading them to claim that interactional context also has the potential to shape L2 WTC. Familiarity with interlocutors, group size, and interlocutors’ participation [17, 18, 22] in different settings all have been shown to influence WTC. A study conducted in Pakistan [23], for example, showed that undergraduates tended to speak English with friends and acquaintances rather than with strangers, and they were more likely to initiate communication in private instead of large group or public settings. Yu [24] also found learners with a low WTC were more reluctant to talk when paired up with peers with a higher WTC. Kang [25] pointed out that the English fluency of non-native interlocutors in class has the potential of influencing speakers’ security and thus their WTC. MacIntyre et al. [26] showed that social support including from parents, siblings, and friends all have the potential of promoting WTC both inside and outside the classroom.

Teachers play an important role in helping learners to develop their WTC [27, 28]. A study [11] conducted in Pakistan showed that teachers’ personality and behaviors, especially familiar, friendly, encouraging, and respectful behaviors, positively influenced postgraduate students’ L2 WTC. Wang and Grover [29] explained that teacher immediacy plays an important role in determining students’ situational WTC. For example, the strategies employed by teachers—discipline strategies or recognition/reward strategies—were found to make significant contributions to L2 WTC inside the classroom [30]. Additionally, motivational strategies like positive feedback, affirmative replies, and praise can establish rapport with students, thereby increasing their confidence and WTC in English [22]. Instructional methods employed by teachers also influence L2 WTC; students participate more actively in collaborative contexts compared to teacher-led contexts [6]. Teachers allowing for a proper amount of wait time after asking questions, arrangement of proper topics, and execution of proper tasks can raise students’ interest [18, 25, 31] while delayed error correction [22] has also been positively attributed to students’ WTC inside classroom contexts.

The final factor affecting WTC—cultural value—has been limited to Asian contexts. Implicit ways of communication, introvert personality types, and poor early English learning experiences as a result of cultural background—mostly Chinese or Japanese learning contexts—have been shown to provoke anxiety during class when teachers raise questions or execute group activities [7, 32]. As noted by Wen and Clément [7]), Chinese students are generally considered low risk-takers and being less tolerant of ambiguity. The culmination of these anxious situations has a strong effect on learners’ WTC during class. In addition, Syed [11] stated that the respect for the authority of the teacher accounted for avoidance of interrupting in Pakistan, a non-western ESL setting.

The lion’s share of the WTC literature contains the analysis of quantitative data obtained by researchers administering questionnaires to participants. Although such an approach to data collection may be considered suitable for measuring trait-like features that could increase or decrease one’s WTC in a foreign language, such an approach may, however, be unable to deeply investigate WTC from a situational view. The current study addresses this need through the use of qualitative techniques. There is also the need for researchers to consider the uniqueness of varying modes of education as well as the content knowledge taught. Previous studies have focused mostly on high school and undergraduate students’ learning of the English language as a subject (i.e., EFL) and those that have participated in previous studies were mainly undergraduate English and non-English majors that were studying English language [2, 28, 33, 34]. Syed and colleagues (2016, 2019, 2020) studied postgraduate students’ L2 WTC and they found that variations in L2 WTC of postgraduate students were shaped by situational variables such as interlocutors, topic, and perceived opportunity. They further noted that higher L2 proficiency did not exert a stronger influence on L2 WTC due to the exclusive focus on content rather than on the language per se. Such findings about MBA students in a Pakistan university suggest that postgraduate students’ WTC in L2/EFL classrooms may be shaped by some unique factors (e.g., course content, cultural orientation, and reliance on L1) and longer experience in English learning may not necessarily yield a stronger WTC. Different from the study conducted in Pakistan where English works as an official language used in government, law court, and higher education [11] and frequent use of code switching between the mother tongue and English happened (even between teachers and students), English is not an official language in the Macau SAR but instead it is a de facto working language alongside Chinese. Moreover, due to historical reasons, Chinese and Portuguese are the official languages in the Macau SAR with English being used as a language for work and education. Furthermore, the current study involves students from mainland China where English is considered a foreign language that are studying EMI courses within a multilingual region (Macau SAR) where English is used as a working language. Along this line of inquiry, the current study recruited master’s students from a wider range of courses in a university with a purpose to enrich the understanding of the factors that may shape postgraduate students’ WTC in English-mediated academic learning in higher education.

Method

The current study was guided by the following research questions:

  1. (1)

    What factors do master’s students perceive as contributing to their WTC in the Chinese EFL context?

  2. (2)

    How do these factors influence master’s students’ WTC in English in the Chinese graduate school classroom?

Context of the Study

The present study was conducted in Macau. The mediums of instruction at universities in Macau are English, Cantonese, Mandarin, and Portuguese. The current study focused on an MA English Studies program at one public university in Macau where the medium of instruction was English. Students in this program are expected to develop their language skills, especially academic writing and presentation skills. More specifically, the current study focused on students from mainland China, whose undergraduate majors were also in English.

Participants

The participants in the study were selected using purposeful snowball sampling. First-year MA students were chosen because most content courses were taught during the first year of the program. One MA student was invited and then through the recommendation of this student, four additional first-year MA students were invited. Four (male = 1, female = 3) of the five invited participants, all with undergraduate degrees in English, were willing to participate. All participants had been learning EFL for over ten years. As all the participants came from mainland China, with an upper-intermediate proficiency in English. In addition, the participants all self-reported “good” or better English-speaking competence. Table 1 presents information about the participants. Pseudonyms were used to mask their identities.

Table 1 Participant profiles
Full size table

Chris and Lydia came from north-eastern China while Dennis and Wendy came from southern China. All the participants spoke Mandarin while Dennis and Wendy also spoke Cantonese. The four participants stated that their English marks were always higher than other subjects in both junior and senior high school. Moreover, their English marks were always higher than their peers. Dennis, Lydia, and Wendy started their formal English learning in primary school. All the participants received private English tutoring during primary years.

Chris and Dennis considered themselves as outgoing while Lydia and Wendy considered themselves as more reserved. Dennis described himself as humorous, claiming he was active to communicate with others in English on any occasion. Lydia stated that she was not an active speaker. While they all claimed the ability to discuss general topics in English without difficulties, Dennis stated he had enough confidence to speak in English about any subject. Regardless, they all claimed that they were talkative with friends in private. Participants were observed in five courses; not all the participants took the same courses. Table 2 provides information on the courses taken by each participant.

Table 2 Observed courses
Full size table

Data Collection

Data collection was conducted over a period of six discontinuous weeks (see Table 3). During the first week of data collection, in-depth interviews were individually conducted with each of the four participants. In addition, classroom observations were conducted each week in which the first author followed the observation protocol to take field notes on the participants’ communicative behaviors. Then post-class group interviews were carried out to gather further information on the participants’ behaviors for 8 out of 12 class sessions observed.Footnote 1 All questions asked during post-class group interviews were based the first author’s field notes pertaining to the participants’ WTC behaviors across different classroom contexts.

Table 3 Procedures of data collection
Full size table

Data Analysis

Data analysis was an iterative and recursive process, which involved reading, coding, rereading, and revising the codes [35]. For the first reading, the researchers read the interview transcripts repeatedly and immediately jotted down ideas, comments, and reflexive notes in the margins that could be used as possible themes.

To answer the first research question, open coding [36] was utilized, which consisted of three steps: conceptualizing, defining categories, and developing categories. For instance, Lydia mentioned that she preferred listening to teachers instead of talking too much in the classroom. She reported that she got used to this way of learning at a young age. This was coded as “learning habit” during the first round of coding. During the second round of coding, the researchers combined what Wendy said about her speaking and answering strategies (e.g., that she would wait for others’ answers and try to develop them) to this “learning habit” code from the first round of coding. Then, after reviewing previous studies on learning styles and learning habits, the researchers finally ended up with the code “learning style.”

To answer the second research question, following open coding, axial (or theoretical) coding was used to “relate the different categories to each other” ( [37], p. 112). The interrelationship of all categories was explored in the process where students’ WTC behaviors were treated as a central phenomenon. Causal condition [38] of situational factors was discussed and then the researcher differentiated between main themes and subordinate themes. The new categories of information emerged when exploring their relationship and descriptive phrases were used to portray the framework.

Results and Discussion

Main Factors Contributing to Master’s Students’ WTC

The factors found contributing to the students’ WTC were categorized into three dimensions: individual, contextual, as well as socio-cultural and educational. These factors perceived by master’s students as contributing to their WTC are described and interpreted below under the three dimensions.

Individual Factors

In this section, individual factors including attitude, motivation, expectations, self-confidence, affective change in the classroom, and learning style are discussed. Similarities and differences in the attitude and motivation of the four participants emerged from the data. They shared the belief that they wanted to further their English language studies and accumulate professional and academic knowledge to master particular English skills. The interview data revealed their attitudes toward English language learning influenced their WTC behavior and reflected their performance in class. They all regarded English as a communication tool to be used to perform future job functions as well as to interact with others socially.

Among all the participants, Dennis was the most active during classroom communicative activities. He believed that English is a practical work skill and a higher English proficiency equalled stronger competitiveness in the job market. He said:

I am comparatively active and I always hold the attitude that I want to learn more. What motivates me to communicate in class…The first is my eagerness for knowledge. The second is seizing more opportunities to create a good impression on teachers. The third is for a future job because you need expressiveness. (Dennis, November 18, 2016)

Wendy said that the master’s courses were full of professionalism and technicality which placed heavy pressure on her. Lydia said she anticipated that the master’s courses were going to be difficult; however, there were a few differences between the reality and her expectation. Much new knowledge about academic issues and theories were poured into classes, which she felt difficult and time-consuming to absorb. Dennis also commented:

I expected the master’s curriculum to be practical and interest-oriented. After looking at the course titles, I supposed that I would love those courses. But when I took the courses, I found that they were not what I imagined. (Dennis, November 18, 2016)

Peng [33] indicates that an individual’s learning attitude, motivation, and expectation impacts a learner’s WTC. Students’ internal attitudes and motivations drive their communicative participation in class. Lydia believed that participation did not equate frequent interaction. Sometimes, it seemed that she resisted to communicate with teachers and peers and instead only listened in class. Among all the participants, she was the most silent in class. As she said:

I listen to what they say carefully, and I think about the issues carefully. I have absorbed what has been taught. I think this is a kind of participation. (Lydia, November 25, 2016)

The participants faced difficulties due to their slow absorption of theoretical academic terms delivered with English as the medium of instruction. They contrasted this situation to their undergraduate studies where introductory linguistics and general English literary works were often explained in Chinese. They felt it was difficult to get involved in class communication with the teachers as they felt they were still developing content knowledge. Take Wendy’s feelings for example:

Now I need to study a certain field like linguistics or literature. If they are taught in Chinese, it may be acceptable. But full English teaching is hard for me to adapt to. Even if those subjects are taught in Chinese, the content might still be difficult to learn. It is much harder in English because I don’t have much background knowledge. (Wendy, November 17, 2016)

Learning interest [33] and topic interest [25, 31] have been shown to affect learners’ WTC. A favorable attitude toward a course or a topic discussed in a course will result in an increase in students’ WTC. Learners will judge the utility and benefits of the topic, and then initiate a conversation or discussion. In one particular class, the researcher field notes revealed an instance of a sudden heated discussion surrounding the concept of “stance.” The post-class group interview transcript relay’s each participant’s thoughts:

Researcher: When stance was mentioned, why was everybody so interested and began asking the teacher questions?

Dennis: We didn’t know the meaning of the concept.

Chris: I knew the spelling but I didn’t know the specific meaning in this context.

Wendy: We didn’t know what theory it was related to.

Chris: This word is the key. If we could not understand it, it would be hard to move on.

Wendy: I wanted to know if I could use the related theory in my own project.

Dennis: Yes, I think it might be useful in my paper. That’s why I engaged in the discussion actively.

(Post-class group interview 1, November 15, 2016)

On another occasion, students were required to give presentations. Wendy was active in raising questions. She said she wanted to confirm her assumptions:

I didn’t watch this movie [Momento] before. So, I was very curious about the plot. I had some assumptions about the arrangement and layout of the plot. I wanted to ask him so I could confirm it. (Wendy, Post-class group interview 2, November 16, 2016)

For the master’s students, their primary concern was unfamiliarity with the topics. Lydia explains this in the following excerpt:

Whether I have confidence depends on the topic. Free chat is my thing. But don’t ask me anything about academic issues in class. (Lydia, November 25, 2016)

The above negative situation hindered Lydia’s self-confidence to the point that she became more unwilling to talk in class. It was justified that knowledge of the topic under discussion could increase or decrease a leaner’s linguistic self-confidence [31]. In this study, the difficulty in comprehension made an impact on the master’s students’ self-confidence. As Wendy commented:

I am not confident in making a statement due to comprehension problems. I feel that I cannot ask some substantial questions when I don’t understand the issues or contents given by teachers at all. Therefore, I would rather not participate in the communicative activities and just listen to the teacher’s instruction in silence. (Wendy, November 17, 2016)

As mentioned before, self-confidence also relates to the overall belief in being able to communicate in the L2 in an adaptive and efficient manner. From the participants’ profiles, they all had faith in their abilities to use English for daily communication. For example, Chris stated:

It is no exaggeration to say that I ranked the first in English from middle school to high school. Though during my undergraduate there were some students who were better than me, I still had a good ranking. I ranked in the top 60 to 80 among 880 students. I think my speaking ability is great and daily oral communication is not a problem for me. (Chris, November 15, 2016)

The master’s students’ affective change in the classroom influenced their situational WTC. This affective fluctuation might have been triggered by different causes such as fear of making mistakes, indifference of the topic under discussion, and sudden appearance of familiar topics. Peng [33] pointed out that anxiety becomes more intense during whole-group discussions. Any one particular emotion is not static where the most recurrent affective factor is anxiety. Master’s students’ anxiety usually arose when they worried about their own performance. As Wendy mentioned:

I am relaxed when I am listening to the teacher in the class. But I am anxious when I can’t understand some points. I am worried that my incomprehension will negatively affect my performance or learning. (Wendy, November 17, 2016)

The master’s students were conscientious and worried about their in-class performance. They wanted to leave a good impression on their teachers. The insecurity appeared when their knowledge gap became obvious. As Wendy noted:

I had expectations that I would participate actively in class, but I have become more and more passive. I didn’t anticipate that my mastery of knowledge could not reach my expectation. I feel frustrated about my performance as a master’s student. (Wendy, November 17, 2016)

The affective fluctuation is also hard to control since it integrates the individual’s own personality and emotional management with the surroundings. From the classroom observation on November 16, 2016, Chris was silent and nervous when other students gave their presentations because later it would be her turn to give a presentation. She was worried about her performance and the teacher’s evaluation. After her presentation, she became much more relaxed and active. When the teacher asked students about a slogan he mentioned at the beginning of the semester, Chris was excited to speak out the sentence “I teach, you learn, we learn together.” In another class, one student reported discourse analysis of a posting on Weibo during her presentation; the participants enjoyed talking about the topic because the contents were Chinese-based, and they felt it was humorous. The excitement and enjoyment initiated their WTC and they shared their opinions with the teacher and the presenter. As Kang [25] noted, responsibility provokes learners to engage in conversation or discussion. The master’s students sometimes felt obliged to convey their opinions when the topic was related to their cultural background. Other times when a long silence occurred, they felt a sense of responsibility to “break the ice.” They shared the following example regarding group interactions:

Researcher: What do you think of a silent student in a group discussion?

Wendy: I may feed him or her a question to let him/her take part in the discussion and ask for his/her opinions.

(Post-class group interview 5, January 9, 2017)

Among the four participants, Dennis tended to look at concrete situations in a reflective manner. He was always willing to participate in classroom activities. As he perceived himself as outgoing, humorous, and a talkative person who can endure pressure, he showed his enthusiasm and prudence in a research study about language learning. He shared:

I like discussing with people when I am given a question. And I believe it is a crucial ability to communicate with people. I participate in the class with rationality, meaning that I won’t ask unreasonable questions and answer questions without logic. I am active but also prudent in raising and answering questions. I will go over all the information in my mind before taking action. (Dennis, November 18, 2016)

Dennis happily accepted peers’ comments and he thought they were good for his progress. Sometimes, he would add his answers to the former student’s answer because he thought it was good for interactional learning. Wendy reported herself as a person who was active underneath the passiveness of her learning—she also described her personality as versatile. On most occasions, she was quiet in class but in private she was talkative. When asked about her voluntary response to the teacher’s question, she said:

I will think about the questions over and over again but not be the first to answer. I will listen to others’ answers and refer to those ideas. Then I will decide whether to answer or not. (Wendy, November 17, 2016)

Learning styles and personal preferences are different from learning strategies “used for task- or context-dependent situations” ( [39], p. 3) as they are fairly fixed. Lydia mentioned that she got used to the traditional classroom mode: teacher talks while students listen. She emphasized that Chinese thought patterns have impacted her comprehension of academic knowledge and that she usually only answered questions when teachers questioned her directly. She shared:

I am not introverted. If you want to talk to me, I would love to talk. I only don’t like talking too much. I am habituated to think over what teachers say. The teacher teaches us in front the class, and I sit silently. I can’t talk or communicate when I am deeply thinking. If I have questions, I seldom ask the teacher directly. I will ask him/her privately after class instead of raising questions voluntarily in class. I am used to listening to the teachers and taking notes. I have grown up not talking too much in class. I feel uncomfortable while talking too much. (Lydia, November 25, 2016)

Lydia pointed out that she passively accepted the classroom communicative activities and she did as the teacher asked. She believed that most of her classmates were also reluctant to communicate actively and only a few students were truly active and willing to communicate in English. Her low-key behavior and acceptance of a teacher-led classroom stemmed from past educational experiences and she had already gotten used to being silent in the classroom since middle school. It seemed that she embodied the stereotype of the “aloof academic” [40].

Contextual Factors

In this section, contextual factors include curricular content, topic, teacher’s role, teacher’s error correction and feedback, teaching style, teacher’s personality, interlocutor’s familiarity, and classroom setting and atmosphere, as well as interaction pattern. Chris thought the complexity of curricular content was most influential on her willingness to communicate. Dennis also mentioned that the only thing preventing his oral communication was the difficulty of the content knowledge but he was willing to continue asking questions if he could not understand a difficult point. Wendy thought the curricular content and academic concepts were so specialized that she considered herself as a stressed newcomer; however, she was still willing to engage in communication when she considered a topic interesting. For example, during the Media Discourse course she showed no interest in an oath speech related to politics because current political affairs did not interest her. Alternatively, she became an active communicator when a language topic concerning Chinese and Cantonese. She expressed that informal content interspersed with formal content could mobilize her initiative to communicate in English. She felt that some classroom digression could ease pressure from the heavy class topics. Wendy shared the following:

I will raise questions when I am interested in the topic and I have comprehended the content. Last class, although I didn’t watch the film that the presenter showed yet, I felt it was quite interesting after the presenter’s introduction and I questioned them based on my understanding. (Wendy, November 17, 2016)

Lydia thought the biggest challenge to her communicating in English was content knowledge. When taking a theory-based literature course she found it dull as it required reading of many reference materials and books:

The contents and knowledge points give me a big burden in class because I can’t comprehend them in English. There are many issues and problems that we need to contemplate in our courses. During the progress of deep thinking it is hard to articulate. (Lydia, November 25, 2016)

Chris and Dennis pointed out that “discussible problems” were important in classroom communication:

I think that to boost learners’ communication engagement, the curricular content and topics should be interesting. The questions should be discussable. (Dennis, November 18, 2016)

The participants stated that by the middle and more so at the end of semester, familiarity with course content had been strengthened. As the classes progressed and the semester advanced, the master’s students gradually deepened their comprehension of the courses’ contents. They became gradually familiar with the curricular contents and academic concepts thereby facilitating communication in class. Understanding had been strengthened, motivating the students to share ideas with classmates and teachers during class. Dennis stated that his attitude changed as his understanding of the curricular contents increased:

Sociolinguistics didn’t appeal to me at first, but I got more and more interested and active in the course after understanding the challenging research and difficulties. (Dennis, November 18, 2016)

The teacher plays a crucial role in a classroom by helping learners to develop their WTC [27]. In the present study, the teachers guided discussions through sharing of their own experiences and explanation of course content. Positive feedback from teachers during communication with students increased their self-confidence and boosted their intention, willingness, and frequency of in-class communication. Teachers’ corrections are still necessary for learning but feedback should be delivered in a way that is accepted by students. Delayed error correction has been shown to increase WTC more so than immediate error correction [31]. Wendy responded well to the teachers’ reactions, correction, and responses to her classroom discussion:

Teachers’ positive feedback can encourage me to elaborate further on answers. If the feedback was negative, I would have been disappointed. But I would still try to think of a better answer anyway. (Wendy, Post-class interview 6, January 13, 2017)

Teachers also increased learners’ WTC through scaffolding learners to answer questions when the class fell silent. The teachers were found to liven up a quiet classroom atmosphere by scaffolding group discussions. As Dennis commented:

When the professor asked us if there were more questions to be discussed, all of us were silent. So, the professor raised a question for us and took part in the discussion together with us. The professor’s involvement in class can not only solve students’ problems systematically, but students can also become more willing to be involved in communicative activities. (Dennis, Post-class group interview 2, November 16, 2016)

Such a scaffolding and supportive teaching style has been recognized as a key factor influencing students’ situational WTC [7]. Students expected that a teacher’s instructional methods could be innovative and flexible—the teacher should become more frequently involved in the group interactions between and among students. Wendy specifically mentioned that she preferred a more relaxed teaching style as the reduced classroom pressure would increase her WTC. She also thought that teachers should be wary of students with lower language proficiencies and should adjust their teaching style to fit students’ abilities. This of course depends a lot on a teacher’s personality and whether the teacher feels it important to build rapport with students. The participants stated that they enjoyed being taught by teachers who were friendly, humorous, and vivacious; teachers should have a relaxed method of delivery of course content. The participants also mentioned that many of their teachers had accents and spoke quickly—this might cause communication breakdowns as the struggle with grasping new content or understanding new concepts would be exacerbated by the listening difficulties. However, all the participants agreed that as exposure to any teacher’s teaching increased, they more or less grew accustomed to the teacher’s speaking style and speed. Instead, they thought that personality factors of their teachers mattered more than their spoken English delivery. For example, a teacher with a humorous personality could reduce the distance between himself and the students which they felt allowed them to relax and thereby comprehend the course content. Dennis commented on these issues:

I think the personality of the interlocutors is very important. If one is bored, no matter what kind of interesting topics are being discussed, I will not raise questions or ask your opinions. Even though I may know a lot, I won’t want to share with others. Surely, personality will influence my WTC. (Dennis, November 18, 2016)

Dennis also discussed how not only the personality of their teachers but also those sitting around him in class affected his WTC:

I think peers have the least influence on my willingness to speak in class. I concentrate on my own performances and I don’t care about others. I hardly intend to talk with a specific person though I admit I feel free with close friends like Chris and Wendy. Honestly, the intimacy affects my willingness to communicate but I won’t refuse to talk with strangers in class. It depends on his or her personality. I don’t want to show my passion to someone inaccessible. (Dennis, November 18, 2016)

Wendy also said that she relied more on friends sitting around her and was willing to speak with them in English to confirm whether her proposed answers were correct before speaking in class. She feared making mistakes in class. She explained:

We close friends usually sit together. Chris and I have a tacit understanding about the problems we meet in class. I can talk to her without scruples. I feel unsure when sitting with a stranger. With acquaintances beside me, I have a sense of security. (Wendy, November 17, 2016)

Wendy and Lydia also pointed out that they were more willing to communicate with familiar people. They explained that their shy and quiet personalities prevented them from striking up a conversation with people they were not familiar with whether it was in individual or group settings. Kang [25] reported that a sense of security could stimulate WTC—for example, when learners converse with interlocutors that share the same L1 and cultural background.

Participants did, however, have different ideas about classroom interaction. Chris preferred teacher-centered speaking, where the teacher dominated teacher-student interaction. She preferred not to get involved with group tasks requiring spoken communication. If she had no option and had to participate in a group task that required speaking, she preferred talking with group members than presenting with group members. She shared the following feelings:

I dislike group presentations because we need to prepare slides and scripts, and resolve disagreements with group members. But for discussion, I just say what I have to say and I am done. (Chris, November 14, 2016)

Dennis preferred a one-to-many communication style where he could individually share his opinions and comments in front of the class. However, he disliked being interrupted. He found group discussions time-consuming and inefficient. He shared the following reflection:

In the class of Media Discourse, we often have group discussion. But the teacher can just ask us to answer those questions one by one. The result of the discussion is that the presenter says the answer for group members one after another. It is unnecessary. (Dennis, Individual interview, November 18, 2016)

Wendy preferred pair and small group discussion as she believed collaborative learning and idea “collision” were beneficial because they would further spark discussion. However, she thought smaller group sizes were necessary as they provided more opportunities for in-depth communication. She shared:

I am more willing to communicate in small sized groups, like in pairs. If it is a dyadic situation, we both have the same discourse allocation and we both can express our ideas. I prefer few people because the opportunity and extent of communication can be increased. If it is free talk of the whole class, I like to have private discussions with peers. (Wendy, November 17, 2016)

Lydia preferred one-to-many presentations. She said that the size of her classes was usually around 8 to 9 students and they were given the opportunity to express their ideas one by one. There was no need for pair or small group discussion in most of her classes, so she had become accustomed to this type of interaction.

The participants did agree that lively or heated group and classroom discussions would promote their WTC. Dennis, for example, said that if everyone in the group kept silent and no one wanted to lead the discussion, he would keep silent. Chris remarked that when senior postgraduates visited their courses, they were able to take on a leader role in the classes that often helped the first-year students speak up more. For example, in the Media Discourse course, the professor invited some of his doctoral students and second-year master students to play a leading role in the class discussions. Chris felt that these senior students had already acquired critical and academic thinking skills that she and her classmates sometimes lacked and Wendy shared that she could think more clearly and was more motivated to share her opinions and communicate when the senior postgraduates took charge of the lessons.

Social-Cultural and Educational Factors

In this section, social-cultural and educational factors including knowledge gap, reliance on L1, challenge of EMI class, social pressure, and cultural orientation are discussed. The four participants shared their experiences of education in Macau and they thought there was a gap of knowledge that hindered their WTC at the beginning of their master’s study. As the semester progressed, they became accustomed to the academic atmosphere and their WTC increased.

Wendy and Lydia stated that English as the medium of instruction was a big challenge for them because even though they were undergraduate English majors, most teachers in their undergraduate years were Chinese speakers and were equipped with the ability to provide Chinese explanations when necessary. Lydia said she needed to adapt herself to the full English classroom environment since the teachers were from different countries or regions, requiring her to adjust to the varying accents and speaking speeds. She recalled:

In my undergraduate study, most English teachers were Chinese. They could give explanations [in Chinese] when students were confused. Now we cannot expect those foreign teachers to give explanations [in Chinese]. (Lydia, November 25, 2016)

The master’s courses stressed students’ English communication abilities and assumed that students would become engaged in class discussion. Wendy mentioned that the fast pace of master’s study and the requirement to become an independent researcher caused a difficulty for her to get involved in class discussions. She explained:

I keep silent because sometimes I can’t follow the teacher’s pace. But during my undergraduate study, teachers would give students some room as a buffer. They taught class half in English and half in Chinese.

This was also mentioned by the other participants as well—there was a strong indication that they relied heavily on L1 Chinese. Based on their previous educational experiences that included either a balance between Chinese and English as a medium of instruction or teachers that could offer further elaboration in Chinese, the participants found it difficult to follow their master’s teachers’ hopes to always speak English in class. The participants also felt there was a mixed expectation inside and outside the classroom as there are limited opportunities to speak English outside the classroom in Macau. Their habitual reliance on their L1 and the limited opportunities could have impeded their WTC [41]. As most of the communication dealt with content knowledge or concepts that the participants felt were hard to discuss using English, if more opportunities for practical speaking conversation occurred in classes, it may have increased their WTC in English both inside and outside the classroom [42].

Participants also expressed their anxiety and worry of potential negative judgment from peer students; the social pressure decreased their WTC. Dennis admitted his eagerness for self-expression in class—he wanted to create a good impression on others. While Dennis was active during class discussions, he admitted a constant worry about the possibility of his classmates considering him as pretentious or as dominating during classroom interactions. Wendy, on the other hand, said she was worried that if she spoke too much in class that it would disturb peers’ thought processes.

The Chinese cultural heritage of students has been shown to influence their WTC [33]. Unlike previous studies that reported that a submissive learning style impeded students’ WTF [7], the present study found the participants were more concerned about how they would appear in the eyes of their peers if they were active contributors to class conversation. As Chris said:

I don’t think we are qualified to give peers comments and advice since I think their presentations seem better than mine. (Post-class group interview 1, November 15, 2016)

The professors were regarded as the class authority by the participants and while they were willing to communicate about particular topics, giving feedback to peers was seen as a role that should only be played by the teachers [43].

Conclusions on Master’s Students’ Situational WTC

This qualitative investigation aimed to uncover factors influencing master’s students situational WTC in the graduate classroom context where English is used as the medium of instruction. Three dimensions (i.e., individual factors, environmental factors, social-cultural and educational factors) were found to have influenced the master’s students’ situational WTC. The factors underlying the master’s students’ WTC did not independently affect their WTC but rather the effect of these factors was interrelated. Moreover, the weight of their combined influence was unpredictable for each master’s student. While it was clear that the influence these factors had on the master’s students’ WTC was individualized as well as related to the classroom context, it is difficult to make generalizations to other contexts. This is because at particular times and in particular circumstances, some factors had more of an influence on the master’s students’ WTC than others.

Students’ WTC often fluctuated in class: WTC first appeared to be rather low but suddenly under the joint effect of individual and contextual factors, WTC increased. The master’s students’ affective orientation changed during class depending on content, teaching methodology, topic, interlocutor support, and the interactional patterns among these factors. For example, in the Bilingualism course, to exemplify the phenomenon of code-switching, the teacher played a music video, which contained Japanese, English, and Cantonese lyrics. Wendy had stated that she did not want to interrupt and ask questions or make comments because she noticed her classmates taking notes and did not want to disturb them. She was concerned that her asking questions would interrupt peers’ concentration; this was an example of constraint from communicating rather than unwillingness to communicate. Then, in contrast to her previous silence in the class, Wendy suddenly spoke out, “I am familiar with the famous Hong Kong singer. I love his music.” She was excited and motivated to share her opinions with her teacher and classmates. Her self-confidence and learning motivation were stirred by the content, and she actively communicated with the teacher about the linguistic concept being discussed—code-switching. Wendy’s L2 self-confidence played an intermediating role by reducing anxiety while simultaneously stirring up a motive for her to communicate with the teacher. If the content used to discuss the academic topic had not been familiar to her, then it is possible that her silence would have not been broken.

Self-confidence and emotions are also affected by contextual factors. Students’ frequency of communication was affected by context, especially when they were trying to converse with uncooperative interlocutors in a group environment. Wendy, for example, remarked, “I will give silent group members some clues or questions to bring them into the discussion.” Students’ learning styles and prior educational experiences may also have a lot to do with how they communicate in class; if a student is accustomed to a particular mode of communication, there may be a period of adjustment where the speaker gets used to the new environmental expectations. Wendy reported that she “c[a]me here to promote my English proficiency and my ideal intention is to be an expert at English.” However, she was often reluctant to answer questions about new topics because she lacked self-confidence. She put a lot of effort into trying to understand academic concepts and theories in class but she usually took care of necessary communication outside of class: a lot of conversing with professors occurred after the formal class had ended.

The most frequently mentioned WTC demotivator was difficulty in comprehending the content of the courses. The maladjustment to full English instruction for graduate studies and poor knowledge in research paper writing, academic theories, and presentation skills set up a general educational background disparity for the students. Nonetheless, as time progressed and the courses moved forward, these master’s students became familiar with the communication environment and the educational expectations. They became accustomed to different teacher’s pace of instruction and accents. The students’ WTC, for various reasons, increased over time.

Situated within EMI contexts across a range of academic courses for postgraduate students in a Chinese university, this study is one of the few endeavors that extend WTC research from secondary/undergraduate students to master’s students. While the findings of this study echo previous research by reinforcing the argument that L2 WTC depends on linguistic, psychological, environmental, and socio-cultural factors [1, 7], this study does yield significant implications for the promotion of master’s students’ WTC in EMI-mediated academic classrooms. First, while one rationale for EMI in postgraduate education is the belief that it would create an immersion environment for more English use, the findings of this study refutes this optimism and reveals that EMI in itself had not been found to increase WTC in classrooms. This is because EMI had contributed an additional layer of difficulties for the master’s students to understand the subject content, let alone to participate in classroom discussions. This difficulty was particularly evident when teacher scaffolding was absent. Although being aware of the benefits of EMI courses, postgraduates sometimes were reluctant to communicate in English on account of dynamic and complex factors [11]. As Wendy stated, her goal of coming to Macau was to promote her English proficiency; however, both individual and contextual factors led to her unwillingness to communicate, which was inconsistent with her cognitive learning goal. Second, this study reveals that for master’s students, the “professional,” “technical,” and “academic” nature of the course content tended to discourage their WTC. This is an especially important finding considering the current study involved postgraduates majoring in English Studies that were enrolled in several different English-related academic courses ranging from English literature to linguistics. This finding echoes Syed et al.’s [12] observation that the “thickness” of the academic content at the postgraduate level could inhibit master’s students’ WTC. As narrated by Lydia and Wendy, the academic content in their EMI courses developed a sense of having a “knowledge gap” in their learning, and such a gap inhibited their L2 WTC. The current study’s findings were uncovered from postgraduates enrolled in many different courses including Media Discourse, Cinema Studies, Understanding & Interpreting Research, the Holocaust in Literature, Memoir & Film, and Bilingualism unlike previous studies while other studies involving postgraduates [3, 11]; Syed and colleagues [3] only investigated students enrolled in a Business Communication module. The current study has added to the WTC literature by showing the change in WTC for different curricular contents and topics in connection with different degrees of interest and motivation. To enhance WTC, there is thus an essential need for the academic content in EMI classrooms to be made, in Lydia’s words, not only “interesting,” but also “discussable” and accessible to master’s students since the academic contents are new to first-year postgraduate students. Last but not least, the findings of the study also reveal how master’s students’ WTC could be promoted by teacher-scaffolded activities such as inviting senior research students to join the classroom-based group discussions. Teachers could give more instructive and informative references and books to the master students to preview before class. Teachers may consider providing more informative background introductions towards new aspects of courses. As narrated by Chris and Wendy, their WTC in group discussion was ignited when it was led by those of more academic competence in the subject areas. While early studies usually attribute students’ unwillingness to communicate in Chinese-L1 and other similar environments to either psychological factors [3, 28] or social-cultural factors such as “fear of losing face by making mistakes” [7, 33], this study suggests that, for master’s students in EMI-mediated classrooms of academic learning, underneath their low level of WTC are genuine linguistic and content barriers to be crossed. Apart from the lecturer or the instructor, the upperclassmen or senior postgraduates as assistants or activators could help the new students to comprehend the difficulties and lead these students into communication or discussions. Meanwhile, teachers could play the role of monitor and motivator that provides feedback to motivate students to share ideas. For postgraduate students, there is thus an important need to distinguish WTC in using English for general communication from WTC in EMI-mediated academic classrooms. As demonstrated by Chris, her WTC in daily L2 communication had not been transferred to her WTC in EMI classrooms mainly due to her unfamiliarity with the academic content. Overall, this study reinforces the value of adopting a situational perspective to understand master students’ WTC. Further research is warranted to investigate how the factors arising from multiple sources identified in the present study could be better orchestrated for more effective WTC from postgraduate students in their L2-mediated academic learning.