As Macaro et al. (2018) note, the use of English as a medium of instruction (EMI), i.e., teaching academic subjects in English, has increased exponentially in higher education worldwide. Kazakhstan, a Central Asian republic, is no exception. The number of universities reportedly offering whole degree programs in English, or courses in English as part of a national trilingual education policy of teaching in Kazakh, Russian, and English, has risen from 2 in 2008 to 42 out of 125 institutions in 2016 (Goodman & Karabassova, 2018).
The purpose of this study is to address the following questions that arise from the implementation of EMI in a context where English is a foreign language, where students move among one, two, or three languages simultaneously, and where higher education nationwide is in the process of reform to reach “international standards of excellence” (Hartley & Ruby, 2017, p. 7): 1) to what extent do students develop academic skills in three languages? 2) to what extent do students transfer these skills among languages? 3) which supports in Kazakhstani universities facilitate students’ development and knowledge production in three languages in a postgraduate context? 4) what is the relationship between those supports and students’ self-reported academic skills?
This study is framed by one main concept and three nested frameworks. We start broadly with the concept of multilingual competence (Cook, 2007; Jessner, Allgauer, Hackl, & Hofer, 2016; Todeva & Cenoz, 2009), defined in our study as the strategic knowledge and skills of three (or more) languages in receptive and/or productive communication for different academic purposes and audiences. We define academic knowledge and skills through the lens of Tardy’s (2004) framework of genre knowledge, which consists of four components: 1) formal knowledge–knowledge of structural elements of a genre; 2) process knowledge–knowledge of the steps to produce a text; 3) rhetorical knowledge–understanding the purposes and audiences of a text; and 4) subject-matter knowledge. Transfer is defined as “the influence of a person’s knowledge of one language on that person’s knowledge or use of another language” (Jarvis & Pavlenko, 2008 as cited in Haim, 2015, p. 698). Language transfer can potentially affect all linguistic subsystems of the language user (Odlin, 2010), as well as the processes of language acquisition and language use (Jarvis & Pavlenko, 2008 as cited in Haim, 2015, p. 698). These cross-linguistic processes can be multidirectional (Cenoz & Gorter, 2011), where the L3 can influence the L1 and be influenced by the L1, or take place between the L2 and the L3 (Cenoz et al., 2001, p. 2). Finally, we have constructed from empirical literature, and test in this study, a framework of institutional support organized around four themes: 1) structural (institutional policies and resources), 2) pedagogical (teaching approaches and adjustments in language use for EMI), 3) linguistic (school linguistic environment), and 4) cultural (values, beliefs, and motivations of stakeholders).
This is a multi-site, mixed-methods study consisting of surveys and interviews or focus groups. Six universities were selected which offer Master’s and/or PhD programs, either fully in English or as part of “polylingual” programs, in three regions of the country: Center (the capital region and predominantly Russian-speaking), South (predominantly Kazakh speaking), and North-East (predominantly Russian-speaking).
During the 2018-2019 school year, students (n=283) completed surveys in which they self-reported their development of genre knowledge in three languages, their transfer of knowledge among languages, the availability of various institutional supports, and the perceived utility of institutional supports for genre knowledge development. Focus groups with teachers (n=34) and students (n=43) of different disciplines (e.g. engineering, journalism) were conducted to gather situated understandings of genre knowledge and institutional supports for interpreting the survey data. Interviews with administrators (n=30) were conducted to shed light on changes in postgraduate education in their institutions with the focus on innovations in EMI/polylingual research programs, and institutional support to faculty and postgraduate students in developing capacities for working in English.
Preliminary Analysis and Findings
Descriptive statistics indicate that students have higher genre knowledge in Russian, followed by Kazakh, then English. A one-way between-groups analysis of variance (ANOVA) shows these differences are significant at the p <.01 level. However, principal component analysis suggests the survey’s four-part framework of genre knowledge holds more explanatory value for English than for Kazakh and Russian. Students’ transfer of knowledge is multidirectional as expected, but students are more likely to transfer knowledge from their mother tongue (Russian or Kazakh) to English. Students who identify as balanced Russian-Kazakh bilinguals report higher ability to transfer genre knowledge among all three languages, suggesting a relationship between transfer and multilingual competence. Principal component analysis suggests there are 4 main categories with a total of 12 subcategories of institutional supports. Of these subcategories, multiple regression analysis suggests discrete skills development and general linguistic environment are the greatest contributors to students’ genre knowledge development in all three languages.
Qualitative data from students and teachers suggest that a major challenge is admissions criteria for entry into English-medium classes which allow or force low or zero-beginner English speakers to be in English-medium courses. Teachers are not trained in methods of teaching their subject in English as a foreign language to groups with varied levels of English knowledge. As a result, pedagogical practices are not conducive to students’ genre knowledge development at the post-graduate level. Moreover, Kazakh for academic purposes is not as well supported, possibly due to the ideology that Kazakh is a language of cultural identity, whereas English is the ‘language of knowledge’.
Preliminary Implications for Theory and Practice
At a theoretical level, the institutional support framework has been upheld by the quantitative analysis. More research is needed, however, on which discrete skills are necessary for academic skills development in Russian and Kazakh at the postgraduate level, and what measures are necessary to support them. Findings need to be shared with university stakeholders to reflect on structural, cultural, pedagogical, and linguistic steps or measures universities can take to improve academic skills development in these programs.