Growing pains: graduate students grappling with English medium instruction in Kazakhstan

ABSTRACT

This study investigates graduate students’ experiences with English medium instruction (EMI) in Kazakhstan. The data reported here were collected through an online survey conducted in 10 public and private universities in Kazakhstan in March–July 2021. This survey received a total of 320 responses from graduate students with diverse age, gender, disciplinary, educational, and linguistic profiles. Through a combination of closed and open-ended questions, we aimed to determine how graduate students coped with EMI in their programs. We found that most respondents are struggling with various aspects of academic reading and writing. Low English proficiency is often cited as a cause of these struggles, but so are specific elements of academic writing. Inadequate socialization in English together with gaps in existing language and writing support are also seen as factors. Overall, the data confirm the existence of deep ecological tensions between policy aspirations and enactment conditions on the ground.

KEYWORDS:

• Academic writing
• English medium instruction
• multilingual graduate students
• Kazakhstan
• trilingual policy
• medium of instruction

Introduction

Universities around the world are seeking to internationalize to benefit from the numerous opportunities for academic cooperation and knowledge transfer that internationalization offers (Kirkpatrick, 2011). In the process, they aim to increase their competitiveness and impact beyond their home country and to move up in world university rankings (Kirkpatrick, 2012; Soler-Carbonell, Saarinen, & Kibbermann, 2017). International initiatives, such as the Bologna Process (1999) and the World University Rankings (Dumang & Symaco, 2020) have further accelerated the internationalization of higher education around the world (Dumanig & Symaco, 2020). As a result, there is greater emphasis on publishing in internationally recognized journals, enrolling international students, hiring international faculty, and building partnerships with top research universities. This drive for internationalization has given a boost to English language and English medium instruction (EMI) in universities in many countries around the world (Kirkpatrick & Liddicoat, 2017; Macaro, Curle, Pun, An, & Dearden, 2018).

Kazakhstan, arguably, was the first among post-Soviet countries to adopt a trilingual education policy and encourage English medium education in its universities. This reform policy agenda is reflected in First President Nursultan Nazarbayev’s 2007 address to the people of Kazakhstan, where his vision for the ‘New Kazakhstan in the new world’ is laid out. This vision includes trilingual education for Kazakh youth: Kazak as the state language, Russian as the regional language of post-Soviet countries, and English as the language of scientific and technological advancement and international cooperation. According to the State Program of Education Development (SPED) for 2011–2020 and ‘Roadmap 2010,’ by 2020, 95% of the population were expected to have full proficiency in Kazakh, 90% to be fluent in Russian, and 20% in English (SPED, 2011–2020). Kazakhstan was also the first post-Soviet country to join the Bologna Declaration in March 2010 (Oralova, 2012). Both the trilingual education policy and the adoption of the European education system through the Bologna Process have led to a sharp rise in the number of programs using English as a medium of instruction at Kazakhstani universities (Oralova, 2012; Seitzhanova, Plokhikh, Baiburiev, & Tsaregorodtseva, 2015). Research shows that the number of master’s programs taught entirely in English increased from 560 in 2002 to 3701 in 2011; in 2015, there were 42 university departments that had undergraduate and graduate programs in English (Seitzhanova et al., 2015). Many new master's and PhD programs in English have come into existence since then.

Situating challenges in EMI programs in Kazakhstan

Aspirations and ambitions aside, Kazakhstan – like many other countries (Aizawa & Rose, 2019; Coleman, 2011; Hamid & Nguyen, 2016; Hamid, Nguyen, & Baldauf, 2013; Kirkpatrick, 2011; Kirkpatrick & Liddicoat, 2019; Kirkpatrick & Bui, 2016; Manan, 2019) – faces critical issues and challenges associated with the growing number of English medium programs. For instance, a study conducted by the British Council and BISMA Central Asia Agency in 2010 found that only 4% of university faculty who taught courses in English had a high level of English language proficiency (Oralova, 2012). While there is general ‘agreement about the necessity of English for students and faculty, the current situation shows that there is a lack of readiness and motivation to learn English and use it for professional purposes’ (Oralova, 2012, p. 132). At the same time, students in EMI programs tend to focus more on improving their English language proficiency than learning the conceptual knowledge of the subjects that are being taught to them in English (Seitzhanova et al., 2015). This may be because faculty members spend a significant amount of time correcting students’ English in the classroom and on examinations. In some cases, the role of faculty in English medium programs has changed from that of discipline specialists to specialists who can teach their subject in English. They routinely face a choice between paying attention to students’ linguistic proficiency and focusing on their subject knowledge. It is therefore to be expected that, even in the absence of any explicit or conscious resistance to the EMI agenda, the processes associated with its implementation will not be frictionless.

The fact that graduate students struggle with academic reading and writing in English language is not unique to Kazakhstan. Research shows that despite meeting the English language admission requirements of universities, graduate students for whom English is a second, third, or additional language still struggle with academic reading and writing in English (Arkoudis & Tran, 2010). Although these students speak excellent English, they struggle when it comes to analyzing and synthesizing literature and writing an academic paper in a coherent and succinct manner (Kirkpatrick, 2011). They often report difficulties with grammar, lexis, and syntax (Al Fadda, 2012). Similarly, challenges with academic reading can cause conceptual confusion and create difficulties in analyzing essay prompts, for example (Singh, 2014). Many of these difficulties are due not only to language barriers but also to an inadequate understanding of academic writing standards and lecturer expectations (Kirkpatrick, 2012; Singh, 2015). Furthermore, there is a discrepancy between how students with English as a second or third language learn to write in their home cultures and the writing expectations at the university level (Al Fadda, 2012; Kirkpatrick & Liddicoat, 2017). Often the nature and style of argumentation in English may be at odds with the cultural and social norms in which students were previously socialized. This suggests that, for these students, learning academic reading and writing is not only a technical matter of understanding and redeploying notions of genre and grammar, but also an induction into a different set of worldviews, values, and beliefs. In some value systems and traditions, critical engagement with one’s own and others’ beliefs, practices, and traditions is not encouraged. As a result, the critical abilities and intellectual self-sufficiency necessary for educational success, especially at the higher education level, remain underdeveloped (Al Badi, 2015; Kirkpatrick, 2012).

In this complex emerging picture of how EMI implementation may clash with cultural norms and institutional practices on the ground, what is still lacking is an understanding of the nature, form, and level of the challenges graduate students face in Kazakhstan, the extent to which these challenges are rooted in their previous education, social background, and psychological make-up (perceptions of English, motivation, language learning ability, interest, and anxieties), and how these challenges affect their academic achievement, social life, and self-esteem. In other words, there is little empirical evidence about how these challenges with academic reading and writing affect the self-efficacy and well-being of graduate students in English medium universities in Kazakhstan. Our study was designed to contribute to our understanding in these areas.

Methodology

The quantitative data presented here were collected through an anonymous online survey in 10 Kazakhstani universities between March and June 2021. This survey is part of a larger collaborative project, which uses a sequential mixed-methods research design (Creswell, 2014; Floris, 2014; Onwuegbuzie & Collins, 2007) to investigate the challenges faced by graduate students in English medium programs in Kazakhstan. The 10 participating universities in this research project offer a variety of graduate programs in English and represent both the public and private education sectors in Kazakhstan. Their participation is voluntary and informed (detailed information about the study, its purpose, ethical considerations, and expected results were shared with the universities beforehand). The study was approved by the Nazarbayev University Institutional Research Ethics Committee.

The participants in the online survey are master’s and PhD students in EMI programs at the selected universities. As many as 320 students responded to the survey, representing different disciplines, degree programs, age groups, regions, and linguistic backgrounds (see Table 1).

Table 1. Demographic information about participants.

Demographics	Number	Percentage	Demographics	Number	Percentage
Gender			Age group (years)
Male	61	19.1	18–25	104	32.5
Female	203	63.4	26–35	106	33.1
Did not answer	56	17.5	36–45	49	15.3
Region			46–55	5	1.6
Rural	47	14.7	Did not answer	56	17.5
Urban	191	59.7	Years of study at current university
Semi-rural	18	5.6	0–2	173	54.06
Semi-urban	08	2.5	2–5	49	15.31
Did not answer	56	17.5	5–7	28	8.75
Degree program			7 and above	13	4.06
Master's	211	66	Did not answer	57	17.82
PhD	43	13.4	Discipline/field of study
Other	08	2.5	Education	122	38.1
Did not answer	58	18.1	Medical sciences	13	4.1
Medium of instruction of undergraduate degree			Engineering	16	5
Kazakh	57	17.8	Natural sciences	3	0.9
Russian	77	24.1	Social sciences	81	25.3
English	91	28.4	Other	28	8.8
Other (Kazakh + Russian)	36	11.3	Did not answer	57	17.8
Did not answer	59	18.4	Country of undergraduate degree
No of languages spoken fluently			Kazakhstan	239	74.7
1	6	1.8	Russia	05	1.6
2	73	22.8	Overseas	19	5.9
3	140	43.8	Did not answer	57	17.8
4	41	12.8	Language test taken for current degree program
Did not answer	60	18.8	IELTS	118	36.9
Mother tongue			TOEFL	08	2.5
Kazakh	192	60	GRE	04	1.3
Russian	57	17.8	GMAT	05	1.6
English	02	0.63	Other	119	37.2
Other	11	3.44	Did not answer	66	20.5
Did not answer	58	18.13
Language most frequently used outside the classroom
Kazakh	83	25.9	Russian	162	50.6
English	13	4.1	Other	05	1.6
Did not answer	57	17.8

As Table 1 shows, the pool of survey respondents was highly diversified in terms of gender, age, geographic location, education, and linguistic background. About 56% of participants speak three or more languages, with Kazakh and Russian being the first language for 60% and 17.8% of respondents, respectively. As many as 28.4% of participants studied in English medium undergraduate programs, and only a small number of them (n = 19, or 5.9%) studied overseas.

The online survey questionnaire contained 45 closed-ended and five open-ended questions to capture the participants’ perceptions and attitudes towards EMI and the challenges they face with academic reading and writing in their English medium programs. The survey used Likert scales from 1 to 3, where 1 stands for ‘strongly agree’ or ‘extremely satisfied,’ 2 for ‘somewhat agree/satisfied,’ and 3 for ‘disagree’ or ‘dissatisfied.’ The survey questions were aligned with the following main research questions:

1. What are the exact nature, forms, and levels of the challenges faced by graduate students in English medium universities in Kazakhstan?

2. How do graduate students in English medium programs express and articulate these challenges?

3. What opportunities and support are provided at the institutional level to students to help them overcome these challenges?

Although 320 students participated in the online survey, only 269 submissions were kept after data cleaning. Some of these submissions are incomplete, as some participants chose not to respond to some items. The 45 closed-ended questions were organized into seven sub-sections and each sub-section had four to seven questions. The sub-sections included levels of satisfaction with: English language proficiency; quality of academic reading and writing; degree of confidence during oral presentations; and English versus Russian or Kazakh for academic reading and writing. Examples of the survey questions included: how satisfied are you with your level of proficiency in the English language; how satisfied are you with the quality of your academic reading and writing in the English language; how confident do you feel when you make oral presentations in the English language; and do you find academic reading and writing in the English language more difficult than in the Kazakh or Russian language?

Similarly, the five open-ended questions were organized into three sub-sections: reading-specific challenges (what are the two or three most critical challenges that you face in academic reading in English?); writing-specific challenges (what are the two or three most critical challenges that you face in academic writing in English?); and students’ expectations and suggestions for improved writing support in English (to improve your academic reading and writing skills, what kind of help do you expect from your university, English language instructors, and faculty teaching content courses and supervising your research thesis?). To process the closed-ended questions, we used descriptive statistics to examine participants’ percentages and mean scores. The open-ended questions were analyzed via the grounded theory method. In particular, we used open, axial, and selective coding to identify specific categories, commonalities, and subcategories among participants’ responses (Creswell, 2014; Merriam & Tisdell, 2016).

Data analysis and findings

This section presents our analysis of the data, combining relevant information from the questionnaire and open-ended questions. First, we present the main results from the closed-ended section of the survey questionnaire, and then we analyze respondents’ descriptive answers to open-ended questions. The latter are grouped into three major categories: reading-specific challenges, writing-specific challenges, and expected or suggested level and type of academic writing support in English.

Satisfaction with English language proficiency (degree)

One key survey item in the closed-ended section of the questionnaire was eliciting information about the students’ level of satisfaction with their current English language proficiency. We were particularly interested in identifying any differences between students enrolled in PhD and master’s programs with respect to this key item. Table 2 presents a snapshot of the responses we collected, while Figure 1 tracks those responses across degree categories. Data show that, on the whole, most respondents are ‘somewhat satisfied’ with their level of English proficiency, with slightly more master’s students (68%) reporting this level of satisfaction than PhD students (64%). Overall, only 23.50% (n = 59) of respondents reported being ‘extremely satisfied’ with their English proficiency, this time with more PhD students (28%) selecting this option than master’s students (22%). Finally, only 9.57% of respondents (n = 24) reported being ‘dissatisfied’ with their level of English proficiency. The difference between the number of master’s students choosing this option and that of PhD students is negligible (16% and 18%, respectively). The overall results show that a significant portion of students, both master’s and PhD, are not fully satisfied with their level of English proficiency, suggesting that they struggle with English for academic and research purposes.

Figure 1. Graduate students’ reported satisfaction with their English proficiency across degree categories.

Table 2. Graduate students’ reported satisfaction with their English proficiency.

Answer	Percentage	Count
Extremely satisfied	23.50	59
Somewhat satisfied	66.93	168
Dissatisfied	9.57	24
Total	100	251

Confidence during presentations in English

This item prompted students to rate their level of confidence regarding their presentation skills in English. They could choose one of three options: ‘very confident,’ ‘somewhat confident,’ and ‘not confident.’ Most students surveyed (n = 130, or 51.80%) felt ‘somewhat confident,’ with minor differences between programs (51.7% of master’s students and 48.8% of PhD students). This difference is slightly higher among those who reported being ‘very confident’ (n = 73, or 29.08% of total respondents) about their presentation skills in English: 28.8% of master’s students and 33.3% of PhD students selected this option. As with the previous survey item, we see that more PhD students tend to choose this emphatic option than master’s students. Finally, a total of 19.12% (n = 48) of respondents felt ‘not confident,’ with again minimal differences across master’s and PhD programs (19.5% and 17.9%, respectively).

These results show that, although most respondents report feeling ‘somewhat confident’ about their presentation skills in English (51.80%), this percentage is significantly smaller than the number of respondents who felt ‘somewhat satisfied’ with their general level of English proficiency (66.93%; see Table 2). The 15-point difference can be explained by an increase in the number of emphatic responses (‘very confident’), but also – and more meaningfully – a dramatic increase in the number of negative responses (‘not confident’). These observations support our hypothesis that graduate students in EMI programs are struggling with academic English.

Satisfaction with quality of academic reading and writing

This item prompted students to reflect on their level of satisfaction with the quality of their academic reading and writing in English. The results show similar patterns to the ones observed before. A majority of respondents (n = 164, or 65.34%) indicated that they are ‘somewhat satisfied’ with their academic reading and writing in English. Again, more master’s students chose this option than PhD students (67.3% compared to 56.4%). A lower percentage of respondents (n = 46, or 18.33%) indicated that they are ‘extremely satisfied,’ in comparison with those who chose this emphatic option for the first survey item (see Table 2); but, as before, the percentages were inverted, with more PhD students reporting extreme satisfaction than master’s students (see Figure 2). Similarly, relatively more respondents (n = 41, or 16.33%) admitted being ‘dissatisfied’ with their reading and writing skills in English, compared to those who were dissatisfied with their general level of English proficiency (see Table 2), the differences among degree groups being, again, negligible. Cumulatively, these results allow us to get a more nuanced picture of the particular challenges that graduate students are facing.

Figure 2. Students’ satisfaction with academic reading and writing in English (breakdown by program).

English versus Russian or Kazakh for academic reading and writing

Students were asked which language(s) they enjoyed more and found easier to read and write in academically. As Table 3 illustrates, Russian remains the language of choice for most respondents (n = 128, or 50.79%), regardless of program or progression. Only 27.38% (n = 69) of them indicated that they enjoyed reading and writing in English, while 19.44% (n = 49) picked Kazakh. In other words, the vast majority of surveyed students (72.62%) do not enjoy reading and writing in English, the official medium of instruction in their programs. These numbers show that Russian still holds sway in Kazakhstan’s education system, more than 30 years after independence.

Table 3. Languages in which students enjoy reading and writing academically.

Answer	Percentage	Count
Kazakh	19.44	49
Russian	50.79	128
English	27.38	69
Other	2.38	6
Total	100	252

Survey participants were also prompted to react to the following statement: ‘Academic reading and writing in the English language are more difficult than in Kazakh or Russian.’ In response, 45.56% of respondents (n = 113) disagreed, 39.92% (n = 99) agreed, and 14.52% (n = 36) were neutral. Had the comparison been between English and Russian only, the percentage of students who agreed with the statement would have likely been higher, as most students find academic reading and writing in Kazakh far more difficult than in English. That is because, in the past, Russian overshadowed Kazakh as the main language of instruction in schools and universities. It is only recently, since the adoption of the trilingual education policy, that English, Kazakh, and Russian are used as mediums of instruction in Kazakhstan. What we can see from these results is that a significant portion of the student population surveyed find academic reading and writing in English challenging. Furthermore, the fact that 45.56% of students disagreed with the statement does not mean that they have no difficulty in understanding and using conceptual language in English. They may have interpreted the statement differently, or assessed their English proficiency based on their IELTS scores or their grades on course assignments. As we know, both scores can be misleading.

In the same vein, respondents were asked to indicate how happy they were with their progress in academic reading and writing at this point in their career/degree. The data show that 41.89% of respondents (n = 93) were ‘extremely satisfied,’ and 45.05% (n = 100) ‘somewhat satisfied.’ Only 13.07% (n = 29) showed dissatisfaction. Nevertheless, the fact that around 58% of respondents are either somewhat satisfied or dissatisfied with their academic reading and writing at this point in their studies indicates that the majority of these students are lagging in their progress in academic reading and writing in English language.

Insights from open-ended questions: reading-specific challenges

In addition to the Likert-scale quantitative items, the survey also included five open-ended questions that prompted students for descriptive answers based on their specific situations, inviting them to elaborate on their views and provide concrete details about their personal struggles with academic reading and writing in English. The response rate for these questions ranged from 133 to 178 students.

The first question sought to identify the most common challenges students faced with academic reading in English. Two key concerns emerged from the analysis of their responses (n = 149, or slightly over 55% of total survey respondents): ‘linguistic barriers’ and ‘academic barriers’ to content understanding. The most frequently mentioned concerns under both are unfamiliar vocabulary (including specialized terminology) and the complexity of academic style (including grammar and phonetics), which reportedly result in students’ low confidence or even anxiety (‘fear’) about their ability to identify the central idea in a reading. Other – ‘technical’ – concerns mentioned by the participants were low reading speeds, high volume and length of academic texts, lack of time for reading or lack of practice, and limited access to credible sources.

Out of 149 students who responded to the open-ended question about academic reading, 56 (37.59%) respondents indicated that poor vocabulary in English was the main cause for their difficulties with reading, understanding, and interpreting the academic texts they encountered. They reported struggling with jargons and unfamiliar words. For one respondent, ‘the use of technical language’ was a critical problem, which made it difficult ‘to concentrate on meaning.’ This resulted in a lack of ‘motivation to continue reading.’ Another respondent noted, ‘I do not understand meaning of many words, when I translate them, it takes a lot of my time. That is why I respond to questions with low confidence level.’ Frequently stumbling on unfamiliar words, another respondent agreed, makes it hard to ‘capture the main idea in the text in first reading,’ and slows down the ‘reading speed.’ According to another student, ‘Lack of vocabulary (synonyms) makes me misunderstand the meaning of concepts when I read academic articles in English.’ Some respondents referenced the cognitive overload they experienced due to inadequate academic vocabulary: ‘the need to concentrate hard to understand text which led to quick fatigue.’ Another respondent attributed his reading challenges to his lack of exposure to the English language in early schooling:

… from elementary classes I have studied math and chemistry in Kazakh, and now in Masters’, to grasp advanced mathematics and advanced chemistry in English is very difficult. Terms and rules are unclear. It took [me] 8 hours per week to understand them.

As many as 79 (53%) respondents found the style in their English academic texts to be too complex and complicated. As a result, they were unable to fully and easily grasp course contents: for example, ‘During reading articles in English some authors’ writing style is unclear because they write in very difficult English.’ Others found academic English more complex and difficult stylistically than the day-to-day spoken language: for example, ‘Long and complicated English sentences required more time and effort to understand’ and ‘There are difficulties in understanding some complicated philosophical texts’. Another respondent observed:

Too many special terminology, academic texts are long, sometimes difficult to read. I am happy when I read a scientific piece of writing with easy language because I finally can fully understand what I am reading about.

While responses like these may indicate some level of resistance to EMI, the data from the closed-ended section of the questionnaire show that most respondents nevertheless consider the English language as important to their academic and professional career now and in the future. When prompted to evaluate this importance, 85% of respondents chose ‘very important,’ while 14% chose ‘somewhat important’ (total n = 261). These results do not completely negate the possibility of resistance to EMI, but our survey was not designed to test for it. Data from the qualitative phase of our larger project may provide more insights into this topic.

Insights from open-ended questions: writing-specific challenges

The responses to the second open-ended survey question confirmed that academic writing in English is one of the major stumbling blocks for most students in EMI programs. This finding is corroborated by the responses to closed-ended questions, which show that 70.26% (n = 163) of respondents endorse the view that academic writing in English is one of the biggest challenges for most students in their cohorts. Similarly, students’ answers to the open-ended questions revealed a wide range of challenges surrounding academic writing in English. The factors mentioned before in connection with the difficulties thrown up by reading academic literature in English resurface: inadequate vocabulary, the complexity of academic writing styles – from general organization to sentence structure and mechanics (spelling and punctuation) – and a lack of proper training and practice emerged as the most frequent issues cited by respondents. These factors partially explain why graduate students feel they are limited in their ability to express their ideas clearly and concisely. For example, to the question which prompted students to list two or three of the most challenging aspects of academic writing for them (n = 143), one student notes: ‘Vocabulary and sometimes I do not know how to properly address/write what I have in my mind.’ Explicit formulations like this appear with some frequency (10%). However, other concerns are also contributing to this limitation in the students’ expressive range. For example, their previous writing experiences in Russian and Kazakh remain an impediment and are complicated by the strict requirements of academic writing in English, such as documentation practices (citing and referencing), proper paraphrasing of sources cited from literature, and formatting. For instance, one respondent notes, ‘I am not confident with integrating academic literature into literature review,’ and finds that their writing suffers from other limitations as well, such as ‘little critical thinking style.’ Organization and logical coherence are stumbling blocks for some: ‘most challenging part of academic writing for me is … the coherence between paragraphs of sentences, linking of different ideas’; another respondent cites ‘organizing ideas, coherence, and lack of academic vocabulary’ as the main problems they encountered. Yet another student expressed a ‘fear of wordiness and repetition of phrases.’ One student summarizes their recent experience with academic writing and writing instruction as follows:

Rules of academic writing are too complicated. I am now studying [taking a course] ‘Academic Writing for PhD students’. This is one of the difficult courses I ever studied. I have to read a lot of materials to have a long list of references and every time refer to previous studies in my writing. I understand that there is formalized special structure of an academic work, but rules are complicated and some of them sound archaic. I just want to express my thoughts and do not put numerous references in my text to support my words. I would enjoy academic writing more if I could write it in a simple language in a free format.

As highlighted in this excerpt, the genre-specific conventions and documentation practices in English academic writing are major obstacles for many students, who perceive these formal constraints (‘rules of academic writing’) as opaque barriers and a time sink. This may point to specific weaknesses in some pedagogical approaches to academic writing, which may be placing too much of an emphasis on the imitation of rules and norms at the expense of invention, in the rhetorical sense of these words. Further information about how academic writing is being taught in English medium universities in Kazakhstan is needed to identify such weaknesses with any clarity, but – as the next section shows – the responses to our open-ended questions already afford glimpses of how the development of students’ academic literacy may be better supported.

Student’s expectations and suggestions for improved writing support in English

At the end of the closed-ended section of the survey, respondents were polled on their level of satisfaction with the language and writing support available to them at their universities. The responses reveal mixed reactions: while 47.51% (n = 105) of respondents are ‘extremely satisfied,’ 35.29% (n = 78) are only ‘somewhat satisfied,’ and 17.19% (n = 38) feel ‘dissatisfied’ with the type, frequency, and quality of support they currently receive at their universities. These numbers suggest that the academic writing support available to students is insufficient.

As a follow-up (open-ended) question, respondents were prompted to indicate the kind of support they would like to receive from their English language instructors, faculty teaching content-based courses, their thesis supervisors, and their school or university. Many students (41.08%, n = 76) expect the university to put in place a comprehensive support program to assist students in developing their academic reading and writing skills in English. Others (34.59%, n = 64) think English language instructors should take responsibility in this regard. Fewer respondents believe that thesis supervisors (15.14%, n = 28) and faculty teaching content-based courses (9.19%, n = 17) should provide ongoing language support to students (see Table 4).

Table 4. Students’ expectations and suggestions for English language and writing support.

Item	Answer	Percentage	Count
1	University/school	41.08	76
2	English language instructors	34.59	64
3	Faculty teaching content-based courses	9.19	17
4	Thesis supervisors	15.14	28
	Total	100	185

The remaining two open-ended questions in the survey were used to elicit more concrete proposals from students about the type, level, and timeliness of academic writing support they need. Some respondents proposed that training in academic English should begin much earlier. As one respondent suggested, ‘… academic English should be taught from 9–10 grade at schools or at least in undergraduate studies, regardless of what major a student pursues.’ Another respondent added that schools should not only teach English, but also ‘educate students about academic referencing, plagiarism and other technical aspects that we struggle with the most.’

Other respondents emphasized teacher training. They believed that most English teachers were not properly trained to teach academic English. For instance, one respondent proposed to ‘add more English lessons in secondary schools taught by qualified English language teachers.’ Another respondent noted that ‘local teachers are not trained well in academic English especially in the local universities. They should be provided with proper training in academic literacy.’ Some respondents also expressed dissatisfaction with the input they receive from their English language instructors. For instance, ‘The academic English instructors at my department just run useless Zoom sessions without paying attention to individual students’ problems.’ Another respondent suggested that the Common European Framework of Reference is ‘vague and not helping school children to master proper reading and writing skills necessary to be successful at an English-speaking university.’ A number of respondents expressed their dissatisfaction over the quality of feedback they receive from instructors on their written assignments. For instance, one respondent demanded that academic instructors ‘explain the use of academic language in a practical way and discuss with us the errors we make and how we could write better.’ Another student asked for more opportunities to workshop their writing:

Our instructors should provide short live special trainings to us. I mean right now we receive explanations during lessons, then independently we have to write something, then submit our work for check, receive comments, re-write the work, submit again and this cycle goes on and on. It would be good to have sample writing with live comments during lessons.

Interestingly, one respondent expressed dissatisfaction with foreign instructors and emphasized the need for local instructors with training in academic English:

There are many local teachers with American and English diplomas, and I think that their teaching will be much more useful and understandable than the teaching done by the foreign teachers. I studied at a preparatory program in an English medium university and there were teachers without strong background in teaching English. So I cannot say that I improved my academic skills in this program. Academic English teachers from the … [ABC country] are very cold people and very strict, I got the impression that they want to work with students who already have excellent English …

These responses suggest that the best approach to improving the support system available to graduate students for the development of their academic literacies and writing skills is a multilayered one, ranging from school-wide or university-wide programs and services supporting students’ development of academic literacy skills in English to instructor training programs on pedagogical approaches to teaching writing. Strikingly, graduate students in Kazakhstan seem to have more than a dim awareness that linguistic proficiency, while necessary, will not be sufficient to ensure their educational success.

Discussion

For a fuller understanding of the results of this study, it is crucial to contextualize the EMI processes, concerns, and challenges reported by our participants. Doing so will help us better understand the complexities surrounding the implementation of the trilingual education policy in Kazakhstan.

The concerns expressed by the participants in our survey cannot be seen in isolation from the macro-level goals of policymaking and the complexity of often competing institutional practices on the ground. Hamid et al. (2013) aptly observe that a rather simplistic understanding of the medium of instruction tends to exist at the macro-policy level, which results in the pursuit of cheap solutions to highly complex language problems. The understanding of the medium of instruction at the policy level often does not reflect the struggles teachers and students experience at a micro-level. The successful enactment of medium of instruction in general, and EMI in particular, entails more than a declaration of policy setting up ambitious goals of internationalization, global competitiveness, and human capital development. Without careful planning and systematic implementation, any policy becomes mere rhetoric. Many countries have set the development of English language proficiency as an apparent goal; however, the ‘politico-economic goals of internationalising education, developing human capital and participating in a globalized economy appear to surpass mere educational/linguistic expectations’ (Hamid et al., 2013, p. 8; Doiz, Lasagabaster, & Sierra, 2012). The fact that English continues to be perceived as a vehicle for human capital development, internationalization, and globalization ambitiously pushes many countries to the agenda of EMI; however, evidence from many contexts demonstrates that little attention is being paid to the development of language competence through adequate allocations of financial, human, and material resources to teaching institutions (Hamid et al., 2013; Kaplan & Baldauf, 2003). This appears to be the case in Kazakhstan as well.

The results from our online survey raise concerns about the way the trilingual education policy is being planned and executed on the ground. As the data show, academic English poses notable challenges to a significant number of students from participating universities. The policy as envisioned at the macro-level seems to run into difficulties on the ground, a typical example of the gap between policy aspirations and enactment that many other countries are familiar with (Hamid et al., 2013; Kirkpatrick & Bui, 2016; Macaro, 2018; Macaro et al., 2018; Macaro, Hultgren, Kirkpatrick, & Lasagabaster, 2019). In our study, this gap translates into a significant number of students expressing dissatisfaction with their level of academic English and reporting low levels of confidence and high anxiety during English presentations. It also translates into potentially longer times to completion. During site selection and entry negotiations at one of the participating universities, we learned that some PhD students at this university have been unable to complete their degree within the allotted time, despite being enrolled full-time. Some are in Year 9 and have yet to complete their thesis. One of the major challenges these students have often cited in their progress reports is associated with their low academic reading and writing skills in English. They lack the necessary linguistic proficiency and academic literacy to understand, analyze, and synthesize academic literature and write up their research. These micro-level challenges define the complexities of the trilingual policy in the country.

In short, our survey data suggest that there is a high level of ecological tensions (Hornberger & Korne, 2016) between policy aspirations and EMI enactment conditions, which hold back the development of academic English. Students’ low level of academic literacy in English is one of the main stumbling blocks to the smooth functioning of EMI in surveyed universities. A lack of adequate knowledge of vocabulary, grammar, and mechanics (spelling, phonetics, punctuation) together with a limited understanding of the procedural complexities of writing for academic and research purposes (citing and referencing, paraphrasing, analyzing and synthesizing information) make both reading and writing in English arduous. Given the nature of these challenges indicated in the data, a crucial question arises, namely, what their root causes might be. Data from the open-ended part of the survey questionnaire are very suggestive in this regard as a large segment of the participants suggested that the mainstream schooling system fails to provide a strong foundation for their academic English. Therefore, they expressed the need for a more robust system of teaching academic English, especially basic writing skills, beginning at an earlier stage of their education. This suggestion is reasonable, as adequately proficient and well-trained professional English teachers and academic writing instructors are integral to the development of students’ academic literacy in this language. But mainstream state-owned schools suffer from a paucity of such teachers, which in turn holds the governments’ reform policy back. Due to these constraints, the education reform policy is currently stalled (Karabassova, 2020). The much-touted education reforms and the teaching of English as a medium of instruction for science, technology, engineering, and mathematics subjects is fraught with numerous strategic and implementation challenges (Goodman & Karabassova, 2018; Karabassova, 2018, 2021; Zhetpisbayeva & Shelestova, 2015; Zhetpisbayeva, Shelestova, & Abildina, 2016), such as teachers’ low English proficiency and poor understanding of content and language integrated learning approaches. For instance, according to Karabassova (2020), a significant number of teachers admit that they do not have good command of English although they completed English language courses with a Common European Framework of Reference level of B2 or higher. Feeling the pressure of teaching in a foreign language in which they have neither proficiency nor proper training (Karabassova, 2021, p. 568), instructors may resist the ambitious rolling out of the educational reform on a large scale, which requires the adoption of new pedagogical models in the mainstream schools.

Then, there is no getting away from the fact that English remains a foreign language in Kazakhstan. Since it is neither an official language nor a cultural language in the country, students’ contact hours with and exposure to English in their sociocultural ecology are limited. This makes it hard for them to perform in it academically, as the sociocultural theory of learning predicts. On this theory, learning and cognitive development are socially mediated through interaction and collaboration with more skilled peers or members of their community (Vygotsky, 1978). Therefore, learners should be provided with extensive interaction opportunities with peers, teachers, and fellows in their proximal environment. The disjunction between school language (in this case, English) and learners’ sociocultural ecology such as home or community can compound academic challenges, and cause numerous learning dysfunctions such as incomprehension of subject material, rote learning, reduced creativity/critical thinking, and parents’ disengagement from the teaching/learning processes (Manan, David, & Dumanig, 2015). A clear indicator of Kazakh students’ lack of exposure to and socialization in the English language in their early schooling years may be observed in Table 1, which shows that only 11.76% (n = 26) of total respondents attended English medium schools, as compared to 45.70% (n = 101) who attended Kazakh medium schools and 37.56% (n = 83) who went to Russian medium schools. (The remaining 4.98% [n = 11] attended mixed Kazakh–Russian schools.) These numbers indicate that most students participating in this survey had negligible exposure to the English language in schools. As a result, they do not have a solid foundation in academic English and are not adequately prepared for university-level work in this language.

Furthermore, as studies of bi/multilingualism show, the development of academic language generally takes longer than basic conversational fluency in a language, a difference which Cummins (2000) captured through his distinction between cognitive academic language proficiency and basic interpersonal communicative skills. These refer to major aspects of language proficiency that learners must acquire while learning a second or additional language. Basic interpersonal communicative skills include the basic language system used in face-to-face communication in informal contexts (intimate or colloquial registers), whereas cognitive academic language proficiency includes formal and academic registers of the language, and such proficiency is strongly correlated with academic success. Cummins (2000) estimated that it takes about five to seven years for learners to acquire academic proficiency in the target language. This means that learners ought to be provided with sufficient time and support to acquire the necessary academic reading and writing skills to succeed in their degrees. Specifically, school students in countries like Kazakhstan should be given sufficient time, exposure, and proficiency/literacy support in the English language before they can undertake studies in EMI at the university level.

Alongside ecological tensions, some procedural concerns also emerge from the data. Procedural concerns refer to how teachers manage classroom teaching and learning and facilitate the flow of information. Participants have indicated repeatedly that they do not receive proper feedback or enough opportunities to workshop their writing and improve. A more fine-grained investigation of students’ experiences (through follow-up semi-structured interviews and focus groups) is planned to understand such procedural concerns, including a perceived lack of sensitivity to multicultural and translingual literacies on the part of foreign instructors in Kazakh universities.

Conclusion and implications

This study provides insights into the ecological tensions between EMI policy aspirations and enactment by analyzing the nature, forms, and level of challenges faced by graduate students with academic reading and writing in English medium universities in Kazakhstan. It finds that most graduate students in participating universities struggle with EMI in their programs due to their inadequate academic literacy skills such as low English language proficiency, lack of understanding of academic writing genres and styles, and inadequate socialization in English during their school and college years. These issues point to limitations in their English comprehension and expression range, and ‘without the ability to read for comprehension and write for expression, students’ understanding may be limited and may lack the complex background knowledge needed to engage in critical thinking’ (Kim & Kim, 2021, p. 2), and hence in university-level academic and research work. These challenges are further intensified by the gaps in existing English language and academic literacy support systems provided by universities. Hence, graduate students in EMI universities are forced to take an entrepreneurial and piecemeal approach to acquiring the necessary linguistic proficiency and academic literacy skills, which can lead to extended completion times and high financial and emotional costs.

This study has implications for policy and practice in relation to EMI in Kazakhstani universities. Since English is neither an official nor a cultural language in Kazakhstan, a ‘one-size-fits-all’ EMI policy implemented haphazardly is unlikely to work (Kuteeva & Airey, 2014). The policy must take into account ground realities and the voices of both teachers and students who are affected by it. A forceful implementation of EMI without appropriate consideration of the realities and conditions at each university is likely to have many negative consequences (Tsui & Ngo, 2016). Therefore, EMI implementation should be accompanied by building EMI capacity in universities through professional development of faculty, academic writing courses for students, and provision of appropriate academic writing support and services such as writing centers. At the policy level, there must be an understanding that EMI cannot reduce to teaching certain subjects in English, but rather requires the creation of sufficient sociocultural opportunities – on and off campus – for professional interaction and collaboration in English.

Acknowledgments

The authors greatly appreciate the valuable contributions made by all research participants from the selected universities, the anonymous peer reviewers for their insightful comments, and their team members (Uli Schamiloglu, Duishon Shamatov, and Gulnissa Zhunussova) who are part of the larger project. They also thank the research assistants (Lyudmila Fillipova and Seth Antwi) who assisted in conducting the survey and organizing the survey data.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research study was funded by the Nazarbayev University Collaborative Research Grant 021220CRP1322, “Graduate students’ struggles with academic reading and writing in English medium universities in Kazakhstan.