Students’ Writer Identities and Writing Practice in Tertiary English-Medium Instruction in China

Abstract

This study adopts a case study approach to examine how students write in English-medium instruction contexts. It also explores why they write in this way from the perspective of writer identity. Four Chinese university students’ EMI course essays, as well as their interview and stimulated recall responses were collected. The analysis results presented three patterns of writer identity: (1) a member, as an EMI writer, of the academic community as the dominant self; (2) a student writer meeting the course requirements as the dominant self; (3) struggling between the two selves. Having different types of writer identities, the students wrote their EMI course essays in different ways. Their writings presented different features in terms of discoursal choice, language form and format. Suggestions for EMI teaching, EMI teacher training and curricula at the university level are provided.

1. Introduction

English-medium instruction (EMI), where English is used to teach academic subjects in non-native English-speaking countries, has become increasingly popular worldwide during the last few decades [1]. Despite the dual goals of EMI programs: the learning of content and subject matter and the development of English language proficiency [2], evidence has revealed that neither of these goals is met well [3]. It is reported that students face a variety of challenges when learning academic subjects with a second language as the medium of instruction. They find it difficult to understand the subject content written in English and the writing of course assignments, essays and reports is found to be found problematic [4,5].

Researchers have attempted to understand the problems of EMI writing by examining the types of EMI writing assessment and language demands in EMI texts [6,7]. However, writing not only involves language or textual knowledge and writing skills but it is also an act of identity. As Burgess and Ivanič note, “asking a person to write a particular type of text… will be requiring that person to identify with other people who write in this way” [8] (p. 228). Therefore, the exploration of how and why EMI students identify with people who write academic texts in English are essential in order to understand their writing practice in EMI contexts.

Despite the emergence of discussion on EMI learner identity [9,10,11], attention has seldom been paid to the students’ writer identity in EMI contexts. To better understand how EMI students write and why they write in this way, this study explores Chinese university EMI students’ writer identity, taking an in-depth look at their perceptions of themselves as EMI writers, and how those self-images influence their writing. The study also seeks the pedagogical implications for enhancing EMI learning and teaching.

2. Literature Review

2.1. EMI Writing and EMI Learner Identity

The effectiveness of EMI learning has been a major concern of EMI research for the past few decades. Although evidence shows that the English test scores of EMI students may be higher than those of first language (L1)-medium instruction students, very few students believe that EMI learning experiences help to improve their speaking and writing skills, and writing was perceived to be the most challenging in EMI learning [12,13]. A survey reported that a high percentage of EMI students advocated allowing the use of their L1 in the exams or assignments [13]. Despite a general impression from the aforementioned studies on EMI students’ perceptions of writing academic subject matter in English, we have little information about how the students actually write and what makes them write in this way.

A range of factors have been identified to explain academic achievement in EMI settings, which include English language proficiency, prior knowledge of the academic subject, gender, sociocultural resources, etc., [5,13,14,15]. Identity as an important factor of second language acquisition and academic learning has also begun to attract EMI researchers’ attention. It describes “how we define ourselves, how others define us, and how we represent ourselves to others” [16] (p. 259). Liu’s study of Thai EMI learners’ identity found that the students desired to be a member of the international community with good academic English skills and L1 English accents via EMI academic studies [11]. Hu and Gao’s investigation of Hong Kong secondary EMI students’ strategic learning concluded that the students viewed themselves as both subject content learners and language learners [14]. Hu and Wu’s study of the EMI learning of university students from the Chinese mainland found that the students played the roles of subject content learners, English users, passive English learners, and Chinese learners [5]. Those studies revealed the effect of students’ identities on their EMI learning practice. Informed by the aforementioned studies, this study attempts to further explore how EMI students write and why they write in this way from the perspective of writer identity.

2.2. Writer Identity from the Poststructural Perspective

Poststructuralists highlight the multiple, dynamic, and complex nature of identity and the agency of individuals in identity construction. From a poststructural perspective, writers construct different identities in different language communities with the influence of the sociocultural contexts of the acquired languages [17,18], and every writer identity is multidimensional, including the socially available possibilities for selfhood (roles a writer plays in a social context), the autobiographical self of the writer (the sense of who they are that a writer brings to the act of writing), the discoursal self (the discoursal representation of a writer about his/her interests, views of the world, values, beliefs, choices of wording, and other semiotic means of communication in particular written texts), the authorial self (a writer’s sense of his/her authoritativeness and authorship), and the perceived writer (the impression of the writer created by the reader) [8,19].

Writers may struggle and show agency in the construction and development of their identities. Novice student writers often strive to fill the gap between the identity they try to construct in discourse and the actual identity perceived by readers [20]. For students who write in foreign languages, the new culture and writing conventions might result in great confusion. Apart from cultural differences, the power relationship is also one factor that influences the construction of writer identity. Primary phase teachers find it hard to balance their identities as teacher-writers and writer-teachers, attempting to negotiate the interpersonal and institutional identity [21]; established professors in universities need to make efforts to maintain their professional status and authorial writer identity [22].

2.3. Academic Writing and Writer Identities

Many studies have gauged the discursive choices that represent writer identities in students’ academic writings [23]. Research of writer identity presented in students’ writing concerns the impression given by student writers about their interests, values, positioning, and authority constructed in discourse [24,25,26]. For example, Li and Deng analyzed the linguistic features in the personal statements of a Chinese English as a second language (ESL) student and found that this student constructed the writer identity of a qualified applicant in personal statements through discoursal choices such as narration of life experiences, reference to themselves, and highlighting keywords and phrases [26]. Kanwal, Qadir, and Shaukat explored the identity presented in the academic writings of a postgraduate student and found that the student’s writing (both EMI and EFL) demonstrated a high level of lexical density, which along with other lexico-grammatical features of the writing (e.g., the use of personal pronouns, citations of published authors) indicated that he identified himself as a member of the academic community and he attempted to present a self-image as a professional academic writer [25]. Hyland examined ESL student writers’ authorial identity by comparing the use of first-person pronouns in writings by ESL student writers with that of expert writers [24]. The results showed that ESL students lack the commitment to their own claims and their use of first-person pronouns was problematic. Comparing the usage of self-mentioning markers between Chinese ESL students and native speakers, some researchers found that Chinese ESL students present a less visible and authoritative writer identity and tend to create a more collective writer identity in their argumentative essays and thesis writing [27,28].

Researchers have also noticed that the writer identities presented in the writings were sometimes different from the ones that they wanted to construct [19,26,29]. Ivanič’s research [19] about the writer identities of eight native adult students studying at the university revealed that while the student writers would like to show their own interests and subjective views in the academic community, they presented the self-image of an impersonal writer with an objective view of knowledge, due to the social pressure and limited resources provided to them. Fernsten found that an ESL student would like to present herself as a competent writer rather than as someone who has low language proficiency [30]. However, her writing identified her as a writer who lacked language fluency. Liu and Deng investigated the citation practices in the thesis writing of Chinese EFL students [31]. They found that while the students intended to construct a knowledgeable and credible discoursal self in the academic community, their problematic references and citations presented them as novice student writers.

3. Methods

This study attempts to understand how university students write in EMI contexts and why they write in this way. To have a detailed and holistic understanding of complex writer identity in natural settings, this study employed a case study approach, collecting multiple sources of data from four student cases, to address the following research questions:

• How do the four Chinese university students view themselves as EMI writers?
• How do their writer identities influence the students’ EMI writing?

3.1. The Context

This research was conducted on the Chinese mainland. Since 2001, the Ministry of Education (MOE) has highlighted the teaching of content using English as the medium of instruction in Chinese universities. To conform to this requirement, Chinese universities have offered a great number of EMI courses. University students who are enrolled in EMI courses take classes that are delivered in English, read English textbooks and other learning materials, and use English to do exercises and presentations, write papers, and take course examinations. The four participant students were from two universities. The university where Penny, Ariza, and Arthur were from is a prestigious one located in the southern part of China. Ariza and Arthur took an International Business undergraduate program which offered a few EMI courses, while Penny took a graduate program of Applied Linguistics, where the number of EMI courses surpassed that of Chinese-medium-instruction courses. Jasmine was from a Sino-US cooperative graduate program of Journalism and Communication offered by another prestigious university located in the east part of China. The program offered English-medium content courses only and most courses were taught by foreign teachers.

3.2. Participants

The four participants were selected based on their responses to a questionnaire survey. We considered the following criteria: (1) they were taking EMI courses at the point of data collection; (2) they had significant heterogeneity in their personal profiles; (3) they were willing to communicate and to participate in the study. By the point of data collection, the four students all had gained abundant writing experience in both Chinese and English. When we started the data collection, Ariza and Arthur had just begun their EMI writing one month before; Penny had a four-year experience of EMI writing and Jasmine had four months. An overview of their profiles is presented in

Table 1. Profile of participants.

Name	Ariza	Arthur	Penny	Jasmine
Age	20	20	23	23
Gender	Male	Male	Female	Female
Education level	Undergraduate	Undergraduate	Graduate	Graduate
Field of study	International Business	International Business	Applied Linguistics	Journalism and Communication
English level (Test/scores)	B2 (CET-6/590)	C1 (IELTS/7.0)	B2 (TEM-8/78)	C1 (IELTS/7.0)
EMI writing experience	1 month	1 month	4 years	4 months

. The scores of standardized English tests that participants took before, including the College English Test (CET), International English Language Testing System (IELTS), and Test for English Majors (TEM), were converted to the Common European Framework of Reference (CEFR) for a better comparison of their overall English language proficiency.

Ethical approvals of this research were obtained when participants consented to participate in the study. The participants understood that their privacy and the right to withdraw from the study were respected, and this study would not place them at undue risk.

3.3. Data Collection

Multiple sources of data, including written EMI assignments as well as semi-structured interviews and stimulated recall interview responses, were collected throughout one semester. The data triangulation through using multiple sources of data established the reliability of this study [32].

One or two semi-structured interviews (about one hour each) were conducted with each student for general information about their English and EMI learning experience, how they wrote for EMI courses, how they viewed their writing behaviors and their writing, what self-images they wanted to present in their writing and how they think of themselves as EMI writers, their relationship with their teachers, the readers, and others related to their writing.

All the essays the participant students submitted for the fulfillment of EMI courses they took during the semester were also collected for examination of how they write in EMI contexts.

Table 2. Information of the writings.

Name	Number	Subject	Task	Grade
Ariza	2	Organizational Behavior	Choose two cases from the supplemental materials, and analyze their organizational behavior issues and write case study reports.	91/100
Arthur	5	Consumer Behavior	Write a report about the consumer behavior of a company.	85/100
		Global Strategy Management	Choose a company to analyze and make strategic plans, including strategic analysis and strategic formulation.	87/100
		Organizational Behavior	Choose two cases from the supplemental materials, analyze their organizational behavior issues and write case study reports.	87/100
		International Trade	Use the economics learned in the class to perform an analysis of a real-world phenomenon.	70/100
Penny	1	Foreign Language Teaching Methodology and Practice	Write an in-depth analysis and evaluation of a teaching method covered in this course.	95/100
Jasmine	2	Journalism and Communication	Identify a key thinker in the field of communication and journalism, read through this thinker’s publications, and write a review essay of her or his works.	B+ (82–84/100)
		Global Hollywood	Write an argumentative essay answering this thesis question: How does Chollywood compete with Hollywood?	A+ (95–100/100)

presents more information about the writing tasks.

The stimulated recall interviews were conducted within 48 h of the completion of each essay [33], to further explore the way the participants wrote and how their perceived selves had influenced their writing. The EMI essays and the analysis of the discoursal choices were used as the stimuli of the interviews. During the stimulated recall interviews, the participants were asked to comment on the images of themselves they attempted to present in their writing, talk about why they chose certain words, and explain how their perceived selves had influenced their writing practice. Each stimulated recall interview lasted 50–80 min.

All the interviews were conducted in Chinese (the participants’ L1) and audiotaped with approval. They were transcribed verbatim for the convenience of analysis. Transcripts were all checked by participants to confirm the validity.

3.4. Data Analysis

Data analysis was conducted throughout the data collection process. To examine how EMI students write, the second author analyzed the discoursal choices in the collected writings, including linguistic features (e.g., sentence structures and lexical choices such as modal markers of certainty and uncertainty, self-mentioning markers, mood, reader-mention markers) and paralinguistic elements (e.g., formatting and pictures). The grammatical errors in the writings were also identified. The analysis was checked by the first author and discrepancies between the two authors were fully discussed and solved. To explore how the students viewed themselves as EMI writers, the first author identified the recurring words or ideas about self-image and interpersonal relationships in the interview transcripts, and coded the segments with phrases such as “confident in the writing”, “professional academic writer”, “obedient student writer”, “highlighting subjective views”, “concerning readers’ understanding”, “showing authoritativeness” and “competent English user”, etc. To reduce the biases in coding, the analysis results were constantly compared to different data sources for triangulation. Similar codes were grouped into categories that were further classified into themes.

4. Findings

Two identities that influenced the EMI students’ writing were found salient in our analysis: members of the academic community and student writers meeting the course requirements. According to the interview responses, conflict was often found between these two identities. Which self the students identified with more had greatly influenced their EMI writing. With the comparison across the four participant students in terms of their writer identities and their writing practice, three patterns of writer identity emerged: (1) a member of the academic community as the dominant self; (2) a student writer meeting the course requirements as the dominant self; (3) struggling between the two selves. The following sections report the three patterns of EMI students’ writer identities and how they influenced the writing practice.

4.1. A Member of the Academic Community as the Dominant Self

Penny had a strong sense of self as a member of the EMI writers’ academic community. Although she acknowledged her student identity and she was clear about the rare possibility of her writings being published in academic journals, she still thought that writing academic essays in English for the EMI courses had labeled her as an EMI writer in the academic community. She said,

… We are still academic English writers… [I am still] at the peripheral area [of the community], knowing very little about it, but one needs to do as the Romans do when in Rome. This is its rule, so you need to get used to it.

(Penny_Interview)

The above interview excerpt showed that Penny felt certain about her academic EMI writer identity although she was not confident in her academic English writing skills. She believed that the writing community she was in had certain rules that all the members should follow. She explained why she thought it important to follow the rules of the community:

With these rules… because while I am a writer, I am also a reader… I can precisely locate the things I search for when I am reading the papers. These rules have somewhat made a standard for academic writing, which a novice writer like me can refer to.

(Penny_Interview)

According to the above interview excerpt, in Penny’s opinion, the writing community is where some members (i.e., writers) share ideas and information with some others (i.e., readers) via their written words, which are organized and presented in styles and forms specifically aimed at the community. The rules and regulations of the organization enable an effective communication between members of the community.

Due to her identification with the community member, she attempted to write like and build a self-image as a professional academic writer, who features both logical thinking and the use of accurate language in the writing. Therefore, she was careful about the preciseness and reliability of her writing, hoping that all the ideas and information she conveyed to the readers were supported by reliable sources. In consequence, she constantly referred to others’ work in her essay, using attributive tags such as “according to”, “argue” and in-text citations to show that the information or ideas were from others’ work as illustrated in the following examples.

Example 1.

According to Butler (2017), in Asia-Pacific regions, TBLT is introduced as a top-down educational policy. (Penny_Essay_Foreign Language Teaching Methodology and Practice).

Example 2.

Long (1985; 2016) has constantly argued that the starting point of TBLT should be a needs analysis. In his words, task should always reflect real-world communicative use, or “things they will do in and through L2” (p. 6). (Penny_ Essay_ Foreign Language Teaching Methodology and Practice).

She tended to quote the original words directly when she felt paraphrases would ruin the preciseness of the writing:

They [the in-text citations] are used to show that the ideas are from others’ work. For example, the term “situational authenticity” was created by this person, and I feel it necessary to indicate that. Then, here, I use his original words, because I think…my paraphrase cannot express the exact meaning… It is to maintain the preciseness… The use of original words and sentences makes it more reliable.

(Penny_Stimulated recall)

The above data excerpt indicates that in Penny’s writing, preciseness and reliability were the major concerns for discoursal choices. In addition to that, Penny also highlighted the correctness of the grammar in her writing to build her self-image of a professional academic English writer and she had spent much time checking the grammar. Her writing that was collected for this study showed pretty good quality in terms of grammar use, with only one grammatical mistake identified. She explained that the error was because of the time limit and she did not think that one or two grammatical mistakes would hurt her self-image as a professional academic EMI writer in general. She explained:

I think it (the essay) is an academic writing in nature. Although our teacher does not emphasize it (the accuracy of language), I still hope it is presented in this way so I feel your (my) language needs to be formal… I would be very careful when I write each sentence to avoid mistakes. I check the spelling of every word that I am not sure about immediately. I would not leave the uncertain form for later revision.

(Penny_Interview)

As shown above, the form of language was not focused upon in the EMI course as is the case in many other EMI programs [34]. However, Penny was still strict with the language form used in her EMI writing. The strong sense of self as an EMI writer in the academic community also made Penny think from the readers’ perspective. Evidence from the interview with her showed that she would use a variety of strategies to help the readers understand more easily (e.g., using parallel sentences starting with “first”, “secondly”, and “thirdly”, etc.) and to prepare for the forthcoming argument (e.g., asking rhetorical questions). In her writing, the inclusive language (e.g., “We need to go back to find out how task is selected in TBLT”) was also identified to invite readers to be more engaged in the reading:

The reason why I used “we”… You see what I put here is “We need to [go back to] find out how task is selected”… It is like we all come to conduct a review.

(Penny_Stimulated recall)

According to the above data excerpt, Penny used the first-person pronoun “we” in her writing to invite the readers to think together with her by recalling what had been mentioned before. However, the first-person pronouns were seldom used in her writing. That is because despite the effort to write like a professional writer, Penny considered herself as a novice academic writer and she therefore avoided showing her subjectiveness in her writing. She said,

I have only read a few papers of some experts. I think I need to read all these classic works, including both books and articles, and I also need to have practical experience if I want to contribute to this field… At the current phase, I do not think my own views are worth mentioning. How I think about it is not worth mentioning.

(Penny_Stimulated recall)

Obviously, Penny did not feel much self-confidence in expressing her own ideas in her writing, because she thought she had not acquired enough knowledge and experience. To weaken her subjectiveness in the writing, she would use the passive voice or terms such as “this paper” rather than the first-person pronoun “I” to hide herself as the subject of the act.

Penny’s lack of confidence in her identity as a professional academic EMI writer is understandable. It is because she did not develop a sense of what a professional academic EMI writer is like until her final year of undergraduate study, during the process of writing her thesis in English:

My thesis supervisor was used to precise academic writing, so he would think the meaning of part of my writing was unclear. I did not realize this problem at the beginning… Since I finished my thesis writing, I had realized that good academic EMI writing should follow a set of rules and regulations.

(Penny_Interview)

Identity reconstruction may occur when one negotiates his/her identity with another’s [35]. Penny’s thesis supervisor presented an image of a professional academic EMI writer. Under his supervision, Penny saw the expectations that her supervisor had of her and rethought herself as an academic writer from her supervisor’s perspective. In this process, Penny gradually gained a sense of what a professional academic EMI writer was like and what kind of writer she wanted to be during her thesis writing. The identity formed in this period had obviously influenced Penny’s EMI writing later in her postgraduate study.

4.2. A Student Writer as the Dominant Self

Unlike Penny, Ariza and Jasmine did not think they were in an academic writing community. Instead, EMI student seems to be the only identity they held during the EMI writing process. Ariza had rich experience of academic writing in Chinese. He had published two case studies and one article. He also had one report accepted by the government. In Ariza’s opinion, while formal academic writing is all about the innovation of knowledge in a field, course essays involve the review of course content only, and therefore do not count as formal academic writing:

Innovation means to increase the brand-new knowledge in human history. It is not that our coursework does not involve creative thinking, but it has no innovation… Formal academic writing requires you to have your own things… You are exploring some new knowledge. You are delivering it to others, contributing to the world… [The purpose of course essays] is reviewing.

(Ariza_Interview)

The above data excerpt shows that Ariza thought the course essays did not contain enough brand-new ideas to be labeled as academic writing. He further explained his understanding of the differences between formal academic writings and course essays:

There may be an initial idea [in my course essay], but I will not further explore it with data or experiment or interviews… It will take a long time to make the idea a real contribution, to make it a formal academic writing.

(Ariza_Interview)

In his opinion, formal academic writing is something that contains ideas supported by empirical evidence. The course essays, therefore, do not count as formal academic writing to him even if they might have some initial contribution to the field. Ariza acknowledged that the quality of his EMI writing was worse than that of his Chinese writing:

If what I write in Chinese is a paper, it will be published in a book. My writing of cases will also be accepted… However, what we write in English are just assignments.

(Ariza_Interview)

The above interview excerpt indicates that Ariza made a clear distinction between his Chinese writings and EMI essays, defining the former as academic writing and the latter as an assignment. This distinction may have put Ariza in a pure position of a student during the EMI writing process. He wrote the EMI essay the way the course lecturer required. In his opinion, teachers have absolute authority. He said,

… I am a student, so my teacher’s writing must be better than mine… When I know he is a teacher, I would be influenced by his authority, thinking that he must write better than I do.

(Ariza_Interview)

Ariza’s perceived relationship with his teacher reflects his strong sense of self as a student, which has greatly influenced his writing practice. He explained that his teacher was very strict with the language he used in the Chinese papers in terms of the conciseness and accuracy. However, the EMI teacher did not care about it. “Even if my writing isn’t very clear, they can still understand it”, said him. For this reason, Ariza was not strict with the use of language in his EMI writing. Many language mistakes were identified in his essay, including informal language, incorrect grammar, problematic wording, etc. Here are three examples:

Example 3.

On the one hand, it was evidently unfair and ineffective that only one person focus on data and model; on the other, working alone might lead to a feeling of helpless, which make it hard to fulfill the love & belonging needs of the person. (Ariza_Essay_Organizational Behavior 1).

Example 4.

And the culture emphasizes fairness, reminding the human resource manager to see the potentials in every job seeker, especially those who are not well-educated. (Ariza_Essay_Organizational Behavior 2).

Example 5.

What’s more, Haidilao’s frontline employees have more decision-making power than employees of other dining enterprises. (Ariza_Essay_Organizational Behavior 2).

The citations in his course essays were also problematic. They were found not to match any citation styles in Academic English writing. He confirmed his carelessness in the stimulated recall. He explained:

I found the citation style was quite annoying. When I started to get restless in the later stage of writing, I was even less in the mood to adjust. I probably would not have written the references if the teacher did not ask us to list them.

(Ariza_Stimulated recall 2)

Ariza’s discoursal choices and his attitude towards referencing in his EMI course essays demonstrated how his writing was influenced by his self-image as a student who did not work hard on the aspects that would not hurt his course scores. It was also the case with Jasmine. Although Jasmine did not have the experience of having papers published as Ariza did, she did not view the course essays she wrote as academic writings either, but as something expressing personal ideas instead. Unlike Ariza, Jasmine’s student identity seemed to be mainly due to her lack of confidence in her academic writing ability. She felt that she was expected by the course lecturer to write academically, but she did not make much effort to do so. While she understood that to write an academic paper, the author needed to refer to others’ work, she showed great subjectivity in her writing. She said,

My writing may have shown strong subjectivity. For example, it is an essay about social media, actually I am telling about my personal views about social media without mentioning the references of previous research on this topic.

(Jasmine_Interview)

She further explained why she wrote in this way:

I think our teacher hopes that we write academically, but because I never had systematic academic writing training, my writing is not so professional or academic.

(Jasmine_Interview 1)

Thinking that she was not able to write like an academic writer, Jasmine only considered how to write the course essays to meet the course requirements. Although she planned to express personal views in her course essay, she added some references as the lecturer asked. However, the in-text citation and the reference list in her essays did not match with each other either. She acknowledged that the use of references was problematic:

Actually, I did not refer much to references. I just wrote based on the teaching content and the information I got from my classmates’ presentations. It is quite subjective. However, our teacher required us to refer to references. So I had no choice but to add some [in the reference list].

(Jasmine_Stimulated recall 2)

She explained that the careless use of references was because the course lecturer did not highlight the accuracy and the format. She said,

I think the teacher may not care about this aspect… He may mainly focus on the structure and the content, not the format.

(Jasmine_Interview 2)

Furthermore, because the lecturer did not put emphasis on the language, Jasmine often chose to ignore the language mistakes in her EMI writing. As with Ariza’s essay, quite a number of inappropriate words and incorrect points of grammar were identified in her course essays. Two examples are presented below:

Example 6.

The first is media governance. the second is media distribution. And media content. (Jasmine_Essay_Global Hollywood).

Example 7.

They focus more and more on service quality & product experience, but less on product and brand loyalty. And the key point is that they are ready to change suppliers. (Jasmine_Essay_Theoretical Basis of Journalism and Communication).

4.3. Struggling between the Two Selves

Arthur considered himself as a member of the academic community and attempted to present a self-image as a professional academic writer in his EMI course essays as Penny did. Similar to Penny, Arthur also thought that as a kind of academic writing, the writing of EMI course essays should follow a set of rules and regulations. He perceived the correctness of grammar and the richness of vocabulary to be the basic requirement of good English writing. He therefore was very careful about the grammar and words used in his essays. He reported in the interview that he always checked for grammatical mistakes with the help of the online tool and polished the essays by replacing the repeated words with their synonyms after writing:

I will use “Grammarly” to check and see if there are any grammatical mistakes. This is a necessary step. After that, I will also read through it and see if all the sentences are clear… Maybe sometimes I will find that I have repeatedly used some words… Then you (I) will replace them with their synonyms while checking to increase the richness of vocabulary.

(Arthur_Interview 1)

The above interview excerpt demonstrates how Arthur made efforts to make his writing look professional. He also strived to figure out what professional academic English writing is like by analyzing the writings of experts. He said,

I feel that their writings are different. To be specific, the argumentation is cunning, using proper examples, without any redundant words… If you read the sentences, you will find that there are not many clauses, and many words used are cunning.

(Arthur_Interview 1)

According to the interview excerpt, one rule that Arthur summarized from reading academic articles was that academic writing should be concise. He therefore avoided writing lengthy sentences in the EMI course essays as illustrated in the following essay excerpts.

Example 8.

We put forward several general suggestions in term of online and offline. Online strategy mainly focuses on developing the game itself. First of all, it’s of vital importance to enhance the overall technology level… In addition, heroine’s characteristic (i.e., character setting) matters. (Arthur _Essay_ Consumer Behavior).

Example 9.

This paper sheds light on the few key factors of international pricing and analyzes the relationship between pricing strategies and entry modes… Identifying customer’s demand is essential in deciding price… Cost is an inevitable topic when discussing international pricing… Prestige of certain firm and certain nation is revealed in the price. (Arthur _Essay_ Global Strategy Management).

A second rule of writing professionally in Arthur’s opinion was to use terminology. He said,

I think the use of terminology is a symbol of standard and professional writing… I may not be able to find appropriate terminologies when I am drafting the essays, but I will add them after I finish the writing.

(Arthur_Interview 1)

Like Penny, Arthur’s writing also represents his authorial identity which shows his perceived relationship with the readers. He used a number of strategies to engage the readers. For example, he attempted to help the readers understand more easily by using transitional language to show the logic and the structure; he used the first-person pronoun “we” to remind the readers to pay more attention; and he asked rhetorical questions to invite readers to think and to interact with him.

While Arthur viewed himself as a member of the academic community and attempted to act like a professional writer, he cared a lot about how the course lecturer would evaluate his essays. Although the course lecturers did not highlight it, Arthur believed that they would appreciate his efforts to make his writing look professional in terms of language, content, and format. This belief seems to have been further verified in his writing practice. He explained why he put a reference list in his essays while it was not required by the lecturer:

First of all, it is a habit of an academic practitioner. I will add a reference list whenever I prepare for a presentation or a report, no matter whether it is required. I will also do this in the final assessment of some courses. This is because of my reverence and respect for knowledge. I want to make my writings more rigorous and express my gratitude to the authors providing insightful ideas… I have spent so much time looking for the literature. [I think] our teachers will give me more scores when they see my efforts.

(Arthur_Stimulated recall 5)

Based on this data excerpt, Arthur thought that writing in a professional and academic way would help to increase his scores in the course study, which indicates that his writing practice was influenced by his student identity along with his professional writer identity. However, some of his ideas of what professional academic writing is like conflicted with the lecturers’ requirements. For example, while he wanted to indicate all the sources he had referred to with citations, some of the reference sources did not appear in his essays because they did not meet the lecturer’s criteria of literature selection:

Actually, I read a great amount of literature, but some of them did not meet our teacher’s criteria of literature selection… I remember that our teacher said we should not cite The Economist. To avoid trouble, I did not put it in the reference list.

(Arthur_Stimulated recall 5)

Similar conflicts were also found in many other cases. For instance, the lecturer suggested they had better not cite Chinese literature and so Arthur did even when he thought that the citation of two or three items from Chinese literature would make his writing look better. He said,

Our teacher said you should not write [the references] in Chinese or translate them into English. That means that we should not cite [Chinese literature]. I think he must mean this… So at the end, I read two or three articles of literature from Zhiwang (a Chinese database). It would look good if they were added to the reference list… But I did not because our teacher said so.

(Arthur_Stimulated recall 2)

The above data excerpt shows that Arthur was not sure about whether Chinese citations were accepted in the EMI essays. He would rather weaken the perceived professionality of his writings when he inferred from his teacher’s words that he should not do this. He made a similar choice when the word and time limits hurt the perceived professionality of his writings:

Because of the word limit, we have deleted something in our draft… and these requirements have somewhat made us delete something in our writing. This deletion may not make the writing more concise, but it makes your argumentation to be like defective goods.

(Arthur_Stimulated recall 1)

Arthur deleted part of his writing to meet the course requirement although he thought the deletion would make the writing defective. He also confessed that he did not feel certain about some viewpoints he put in the essay, because he did not have time to finish reading a case that he thought may confirm this idea. He said,

If time allows… if I can submit it this weekend, I may further revise it, including the uncertain part where I used “might”. I would find evidence to support it. Because I did not have enough time, I had no choice but to submit it the way it was.

(Arthur_Stimulated recall 2)

He further explained that the reason why he retained the uncertain ideas was that the lecturer required argumentation for both advantages and disadvantages in the essay and these ideas were the only disadvantages he could think of. The aforementioned cases indicate that Arthur often struggled between his sense of self as a member of the academic community and his student identity during the process of EMI writing. Obviously, when his beliefs about professional academic writing conflicted with the course requirements, Arthur’s student identity had surpassed his self as a member of the academic community.

5. Discussion

Studies of EMI learning have examined its effectiveness, students’ perceptions, their learning difficulties and the causes, as well as the strategies used to overcome the difficulties (e.g., [12,13,14]). To further contribute to the field, this study adopts a case study approach to have an in-depth examination of how EMI students learn, with a focus on their EMI writing practice. It also attempts to explore the rationale behind their writing practice from the perspective of writer identity. Unlike the studies that examine the discoursal choices of ESL students (e.g., [24,27,28]), this study does not find a unified pattern across the four participant students. This may suggest a variety of students’ writing in EMI contexts. While the analysis results of this study echo Kanwal et al.’s study [25], concluding that EMI students may write as professional academic writers do, this study reveals more possibilities of how EMI students write. Not all the participant students in this study attempted to follow the rules and regulations of academic writing. Two of them (i.e., Ariza and Jasmine) did not care about the language and the format used in their EMI course essays, and a great number of grammatical mistakes, problematic referencing, and strong subjectiveness were identified in their writings.

Previous studies have revealed that L2 writing would be affected by factors such as the learners’ English proficiency level and the writing task complexity [36,37]. According to the results, while some of the four students shared the same educational level (Penny and Jasmine), the same English proficiency level (Penny and Ariza), the same subject field and the same writing task (Arthur and Ariza), as well as similar writing scores (Penny and Jasmine), they showed different patterns in terms of their EMI writing practice and writer identities. Evidence shows that the different writing practices across the four students were led by the different ways they viewed themselves and their relationships with others. The aforementioned students’ carelessness in terms of the language and format was due to their perceived selves as student writers meeting the course requirements, which seldom highlighted aspects other than the subject matter [5,34]. The students who viewed themselves as members of the academic community instead tended to represent their self-image as professional academic writers in their EMI writings, paying attention to all the aspects, including the language, the content and the format.

Not only the two students who perceived student writer as their dominant self during the EMI writing process, but also the other two students perceived the high authority of their teachers and showed the great influence of the course lecturers on their writing practice. In the case of Penny, her undergraduate thesis supervisor played a model role of a professional academic English writer. Under his influence, Penny started to form a sense of self as a member of the academic community and tried to act like a professional writer. She maintained this identity and writing practice in her postgraduate EMI learning as they did not conflict with the EMI course requirements. However, as shown in Arthur’s case, when the course requirements did not match the students’ beliefs of what professional academic writing should be like, the students let their student identity be salient among their multiple identities and wrote to meet the course requirements [35]. This suggests an unequal power relationship between students and teachers [38] and a great influence of teachers on the construction of students’ writer identities. For this reason, teacher training is necessary to help EMI teachers be aware of their important roles in their students’ reconstruction of writer identities in EMI contexts. It is important for EMI teachers to define the course essay as academic writing and to require their students to pay attention not only to the content but also to the language, so that the dual goals of EMI learning can be both achieved.

As indicated in Ariza’s case, the EMI students may consider student writer as the only identity of their own during the EMI writing process because of their lack of skills in academic writing, even if they understand they are expected to write like a professional academic writer. Based on this, it is necessary for the university to provide systematic curricula to prepare students with academic English skills before taking EMI courses, helping them gain knowledge about the construction of academic voice and awareness of different linguistic resources [39].

6. Conclusions

To understand how students write in EMI contexts and why they write in this way, this study analyzed the EMI course essays of four Chinese university students and their interview responses. It concluded that the students’ EMI writing practice was greatly influenced by the writer identities they held during the EMI writing process. Three patterns of writer identities emerged in this study: (1) a member, as an EMI writer, of the academic community as the dominant self; (2) a student writer meeting the course requirements as the dominant self; (3) struggling between the two selves. Under the influence of these identities, the students held different attitudes towards the discoursal choices, the language form, the content, and the format (referencing in particular) of their EMI writings. Although the factor of the students’ writer identities was not the focus of our study, we found that during the process of their writer identity formation, their teachers played essential roles. Given the great influence of students’ writer identities on their EMI writing practice, we suggest future studies further explore how different writer identities are formed, examining whether students’ English proficiency levels, educational levels, writing grades, fields of study, etc., will affect their writer identity formation.

Despite the small sample size, this study is expected to contribute to a better understanding of students’ EMI writing practices and to inform pedagogy for enhancing EMI learning. We suggest that EMI teachers help students form a sense of self as an EMI-writer member of the academic community by asking them to follow the rules and regulations of academic writing when writing EMI essays. Teacher training is therefore needed to increase EMI teachers’ knowledge about those rules and regulations. We also call for systematic curricula where academic English and EMI courses go together to better achieve the dual goals of EMI.