English-medium instruction at universities: global challenges

10 A Critical Perspective on the Use of English as a Medium of Instruction at Universities Elana Shohamy Introduction The use of ‘other’ (foreign, second) languages as medium of instruction has become a widespread phenomenon in the past decade in various educational contexts and settings worldwide. In general the approach applies to situations in which academic content, in a number of subjects (e.g. humanities, science), is being taught via languages which are not the native languages of the students but rather via the languages they seek to acquire. The approach is implemented in all levels of schooling, pre-school, elementary (early and late), secondary and especially tertiary. Content based instruction is a general category that embeds a highly complex phenomenon as it covers a variety of language educational practices (Hu, 2008). Some examples include CLIL (Content and Language Integrated Learning), partial immersion and full immersion (where the whole curriculum is being taught in other languages). The approach refers also to programs where students learn via different languages than those used at home or at their community and to situations where immigrants engage in programs where the language of instruction is different than their home languages and the languages they used in schools in the countries where they came from. The main ideology and assumption behind the above approaches are that the learning of academic subjects via another language is instrumental in enhancing language proficiency given that the learning employs meaningful content which in turn creates motivation to learn the language more efficiently (Cushing et al., 1997; Swain & Johnson, 1997; Swain & Lapkin, 2005). In all of these programs, academic content is learned simultaneously with language with the goal of enhancing both, i.e. not only language proficiency but also the academic knowledge. In other words, when ‘the other’ language serves as the medium of instruction students are not only Brought to you by | University of Groningen 196 Authenticated Download Date | 12/1/19 2:30 PM A Critical Perspective on the use of English as a Medium of Instruction at Universities 197 motivated to learn and use the language in reading, writing, speaking and listening, but also manage to improve their academic content (Swain & Lapkin, 2005). Indeed, higher levels of language proficiency were found in various types of immersion programs in the 70s and 80s, mostly in Canada (Johnson & Swain, 1997). These programs emerged mostly from the need for growing input of the language, especially in societies which defined themselves as bilingual. In a number of comparative studies where the language proficiency of students in regular foreign language programs was compared with those going through immersion programs, it was demonstrated that the latter group performed better and also succeeded in gaining academic knowledge. Yet, during the past decade these new approaches to learning languages spread widely to the variety of contexts mentioned above, and mostly for the learning of English as a second/foreign language in a new global world where it serves as the main lingua franca. The purpose is to increase the exposure to English, a language which is learned as an additional language in almost all countries worldwide along with the national languages which serve almost universally as medium of instruction in schools. Thus, the switch to the use of EMI (English Medium Instruction) instead of the national languages in a number of subjects and/or the whole curriculum reflects a big shift in the approach to language teaching. It also reflects an urgent need to drastically expand the use of English, not only as broader input and exposure but also, and more importantly, as it gives a superior status and prestige in relation to national languages. This trend follows a long tradition of associating high levels of academic knowledge with status of language, i.e. serious academic knowledge needs to be learned in high prestige languages, thus continuing a tradition whereby German and Latin were used as languages of instruction in higher education. Such is often the case in most colonial and post-colonial contexts where local languages are being taught in early grades and then there is a switch to languages which are viewed as more prestigious, in higher grades, and especially at universities. This tradition is especially dominant in postcolonial countries in Africa and Asia (Phillipson, 1992). This direct association of the prestigious academic knowledge and prestigious languages has been kept for a long time with little resentment and criticism. Yet the value of the variety of programs where the language of instruction is the language one seeks to acquire occurs in multiple contexts; it is to these different settings that I will now turn and demonstrate how the teaching of EMI at universities cannot be detached from broader settings where medium of instruction approaches are implemented. As we will see, while the principle of using another language as a medium of instruction is the same and the settings vary, there are major lessons that can be learned from Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM 198 Part 4: Institutional Policies at Higher Education Institutions each of the settings that may have an impact on the others. The analyses of four types of approaches to medium of instruction will be described, leading to critical arguments as to the use of EMI at universities which are driven mostly by ideological and economic motivations and less by educational benefits of maximizing academic knowledge. I will then argue that the teaching of English at HEIs (Higher Education Institutions) actually leads to discrimination against students whose English proficiency is not high and who are prevented from maximizing their academic knowledge (see Wilkinson, this volume). In the last part of the chapter I will argue for an approach that promotes multilingualism and translanguaging as part of effective learning, i.e. an approach to multi/bilingualism where speakers use a number of languages in harmonious ways and not limited to one language. Four Settings of Medium of Instruction Four settings of medium of instruction, which differ drastically in their contexts, goals and outcomes will now be described: learning content via L2 for immigrant students; using a school language which is different than home language; learning content via L2 for majority students; and learning through EMI at universities. Learning content via L2 for immigrant students The first setting is where immigrant students who arrive in a new country are immersed (or rather submersed) in schools where all academic knowledge is being taught via the national language, which is not fully familiar to the students. Indeed, the ideology of using the national languages as the only medium of instruction is implemented almost universally with immigrant students, whether in the US or in other countries, is clearly still associated with views of ‘one nation one language’. There are only rare cases where the medium of instruction will be different than the national language, or when immigrants are allowed to continue learning academic knowledge in their home language. However, in most educational contexts immigrant students learn the school content via a new language. While in some systems there is realization that there is a need to acquire some basic knowledge in the new language prior to acquiring knowledge, in most systems students are immersed into the new language in schools right after immigration. Yet, acquiring content when the language is not familiar is difficult if not impossible. Indeed, a number of studies (e.g. Levin et al., 2003; Levin & Shohamy, 2008; Thomas & Collier, 2002) have shown that students Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM A Critical Perspective on the use of English as a Medium of Instruction at Universities 199 in these situations face great difficulties in acquiring the academic knowledge while having low skills in the new language. These students pay a great price in terms of acquiring academic knowledge as ‘language’ is the main vehicle through which new academic knowledge is acquired. In our own studies (Levin et al., 2003; Levin & Shohamy, 2008), we showed that it takes immigrant students a long period of time to learn academic language and, consequently, they have problems to acquire academic content in mathematics, chemistry, history, etc. Specifically, we showed that it takes students about nine to 11 years to acquire academic language; this finding applied to students of high socioeconomic status (SES) such as those immigrating from the former USSR, whereas for students of lower SES –students immigrating from Ethiopia – academic knowledge was minimal and they were unable to reach similar levels of academic achievement as those born into the language. This difference in learning and testing between students born into national languages which are used as medium of instruction in schools, and those for whom the very language has not been fully acquired leads in many educational contexts to a policy of accommodations, in both teaching and testing (Abedi, 2009). Such accommodations include provisions of dictionaries, extra time on tests, and tests in the L1, among many others (see an example of accommodation in Van der Walt & Kidd, this volume). One dominant policy for helping these students cope with such difficulties which is applied mostly in the US is bilingual education. Under this policy immigrant students continue to learn in their first languages in separate classes and the language of instruction for learning academic subjects continues to be their L1 until they reach acceptable levels of English which will enable them to participate in main stream classes. This policy was implemented in the US mostly in the 80s and 90s and aimed to minimize the length of time it took students to acquire academic knowledge. Parallel to that, students also acquired the dominant language with the realization that this can take a long time and it is urgent to prevent the gaps that can occur as a result. Many of these bilingual education programs have been terminated as a result of political and assimilative ideologies of ‘one nation one language’ and strong beliefs that upon immigration students should be fully immersed in the new language in the shortest time regardless of how long it takes them to acquire the language. Using school language which is different than home language A second education policy takes place when a national language or post-colonial one (e.g. French or English in Africa) is used as the medium of Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM 200 Part 4: Institutional Policies at Higher Education Institutions instruction in schools for students who use different languages at home or in the community. This occurs with groups which are termed indigenous and/or those who make up second and third generation immigrants who continue to use different languages than those used in schools; consequently, in schools all content is taught via another language they are not familiar with. This is typical of many countries in Africa, South America and Asia, among many others. Take the example of Zanzibar, where many students are born into various tribal languages; upon attending school they learn via Kiswahili, the national language; then in 9th grade the language of instruction changes to English and this policy is applied to all academic subjects. In research conducted by a team from Bristol University led by Professor Rea-Dickens and others (Rea-Dickens & Afitska, 2009; Rea-Dickens et al., 2011), it was shown that students who learned all their academic subjects in Kiswahili performed poorly on academic tests which were carried out in English in subjects such as chemistry and mathematics. In a series of experiments they demonstrated that the low achievement on these tests can be attributed to the language of the test. Thus, while the learning through English is viewed by the educational systems and parents as a route to success and prestige, students’ achievement seems to suffer. Situations like this also occur in a variety of contexts where the language of instruction is a standard variety while students are mostly exposed to a spoken variety, i.e., a dialect. In Cyprus, for example, students who use a vernacular language at home are forced into learning academic knowledge in standard Greek and they report on difficulties in academic achievements (Papapavlou, 2001). At times, students use their home languages in kindergarten or first or second grades, but switch quickly to the more prestigious language. In Israel, Arab students whose home language is Arabic continue to use it as medium of instruction in schools. Once they attend secondary schools, they learn sciences in the dominant language, Hebrew; upon enrolling at university, Hebrew becomes the only medium of instruction. Yet, even within Arabic, while students use a spoken variety of Arabic at home, upon entering first grade they are forced to switch to Modern Standard Arabic, which is different than their spoken dialect. Learning content via L2 for majority students Up to now we referred to educational policies which can be viewed as subtractive, since a new language is used as a medium of instruction while there are no investments in the home language at schools. The third setting is one where students choose to learn one or several subjects in another Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM A Critical Perspective on the use of English as a Medium of Instruction at Universities 201 language in additive ways, aiming at achieving bilingualism. In other words, the learning of the new language via meaningful academic content is believed to enhance the level of the language while learning the content, i.e. two goals by one policy. The application of the policy takes place mostly via EMI where students learn some or most of academic subjects in schools in English, given its global status and intense need to study it. As noted above, the approach is based upon claims that the combination of new content in a new language will enhance both the language and the content providing authentic learning environments in a language that is needed for functioning in the global world which is used in all domains of life, business, academy, the internet, etc., hence most of these programs use English, or other power languages such as French, German, Spanish and Chinese as languages of instruction. Very limited empirical research is available as to the success and impact of these programs in reaching the goals; questions still remain as to whether the learning of academic content in a language that is being acquired does result in acceptable gains in academic content and in language. Further questions can relate to whether students in fact gain enough of the language given the extensive input. As a result of the extensive research on immersion in the past, this immersion approach is being viewed as successful for language learning, although many questions arise as to the implementation of the program in other contexts (see Lasagabaster & Sierra, 2010). Yet, little is known about how the learning of academic language in a prestigious language affects the acquisition of the academic content; nevertheless, it is a widespread phenomenon around the world in programs such as CLIL and the like. Learning through EMI at universities The fourth setting refers to the widespread approach at many universities worldwide to teach academic subjects via English. The phenomenon is a reflection of the combination of two power entities: on the one hand, the power of the English language at this day and age, a language that a large number of learners seek to acquire given the belief that it will provide access to central bastions of society and, on the other hand, the power and status associated with the prestige of universities that grant degrees and provide access to the workplace. Given the status of English in the world today many HEIs adopt English as the main medium of instruction and thus defy the policies of most universities to teach in the national languages. These programs can still be viewed as additive as they assume that students who attend universities have acquired high levels of literacy in their first language. One major argument of universities to use English is not only that Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM 202 Part 4: Institutional Policies at Higher Education Institutions it will be effective for learning language and content, but mostly, since English is viewed as an international language, universities will be able to attract international students who are proficient in English and provide the universities with economic benefits (Doiz et al., 2011; see also Wilkinson, this volume). English provides a very special case as it is through English that scientific and academic knowledge is being transmitted especially in academic journals. This is even more so at higher education which, as noted above, tends to be associated with high prestige languages for the teaching of academic subjects. Consequently, universities comply with a prestigious and high status language, which becomes more powerful than the local and national languages and stipulate that academic language must be taught in English. Academic spaces are more and more English spaces as there is close association between high level content with high and prestigious language in all academic domains. The effectiveness of these programs and the extent to which university students reach expected levels of academic knowledge when they are taught via a language that many of the students are not fully proficient in is still an open and under-researched question. Most of the studies predominantly examine attitudes towards such policy (see Cots and Doiz et al. in this volume) and not actual academic or language outcomes. The next section of the chapter will address this very context of EMI at university. A Critical View of EMI at University Raising questions and challenges as to the costs and benefits of this policy within the settings of multilingual societies will be the focus of this section. These questions are presented without overlooking the need to learn English as a major lingua franca of the world today in which most academic materials are produced. After all, almost all academic articles and books are published exclusively in English (Curry & Lillis, 2004; Lillis & Curry, 2010). Issue 1: Content versus language The first challenge of EMI relates to the issue of how successful achievement of academic content is when it is acquired via a language students are not fully familiar with. Similar questions can be asked as to how successful the improvement of English is when the goal is to use the content for arriving at higher levels of English proficiency. In ideal situations the familiar language serves as an effective vehicle for acquiring academic content, what happens then when this tool is not fully functional? Does it mean that Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM A Critical Perspective on the use of English as a Medium of Instruction at Universities 203 the academic content will not be achieved to its fullest? As noted above, in settings where immigrants are not highly familiar with the medium of instruction, they face major difficulties in the acquisition of content in academic subjects. It is a long process (9–11 years in schools) and affects the level of academic knowledge acquired negatively. This is so dramatic that most educational systems introduce sets of accommodations in tests and in teaching (Abedi, 2004, 2009; Levi-Keren, 2008, among others; and Van der Walt & Kidd, this volume) which are expected to put a remedy to the limited academic knowledge that is achieved without these accommodations. Thus, the main challenge that is raised here relates to the effectiveness of the measures implemented in order to acquire academic knowledge which is, after all, the most important goal of university education. What if research reveals that the processing of academic texts at universities in a language that is not fully familiar to students results in lower achievement of academic knowledge in comparison to what it would have been had the students read the same material in a language in which they are highly proficient? One cannot devalue the importance of reading any piece of text in the original. Still, even if we appreciate the value of reading in the original, we need to challenge the ability of students in academic settings to use English in all the language skills, to read, write, comprehend and participate orally in discussions, speeches in academic contexts of classes, conferences, etc. In this vein, current critiques of the domination of Standard English in journal articles should be considered. There is a strong call by many researchers worldwide to include more languages in academic articles so to defy the full domination of standard English as the only language of academy which compromises equality and opportunities to learn (Curry & Lillis, 2004; Lillis & Curry, 2010). My call here is mostly for extensive research to examine empirically the cost and benefits of the use of EMI at HEIs; the main goal being the need to explore how much language is being gained by such programs as well as how much academic content is being achieved. Based on research reported above in contexts of immigration, there is evidence that this type of learning is not very effective. There are still a number of related issues. One is the availability of professors who have high levels of knowledge in both areas, academic content and English, not just in reading but also in speaking. It is often the case that academic professors will have high knowledge in one of the areas and not in the other. Another issue relates to the ease and comfort in which students interact with one another in a language that is a second language for both. In EMI contexts students are required to communicate in a language which Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM 204 Part 4: Institutional Policies at Higher Education Institutions does not represent the natural type of discourse between students and teachers, and among students, possibly harming the natural flow and rapport among students and teachers, especially when fluency is not reached and ideas are not communicated fully. Issue 2: Inequality in the global status of English for different groups The second issue addresses inequalities resulting from the ‘relative’ importance of English for students of different backgrounds, especially those termed ‘minority’ students, indigenous and immigrants, for whom their home languages are different than the national languages, so that English is considered a third language and often the national language has not yet been fully acquired. The situation results in major inequalities and lower achievements in tertiary education. In an earlier article (Shohamy, 2007: 128), I challenged the term ‘global languages’ to refer only to English while arguing that a global language is a term that has a different meaning for immigrants and indigenous people. Specifically: Globalization, in my view, is a relative and complex term referring to hierarchies and prestige of a number of languages within given contexts. Thus, for some groups, the languages of the global world may refer to those used in the ‘next town over,’ whereas for others it refers to the languages ‘of the world.’ In fact, any language can be considered global when it is considered more prestigious and powerful than one’s own, and when it is used as a gatekeeping device for entering institutions that enable upward mobility. Thus, in multilingual societies, which are the dominant language reality nowadays, consisting of indigenous, immigrant, and transnational groups, a number of languages can be perceived as global and not only English. These refer to all other languages that individuals and groups seek to acquire to gain participation and inclusion and economic, educational, and social access. This is so because it is rarely the case that those who are outsiders to the language are capable of reaching full levels of proficiency in relation to those who acquired the languages from birth and continue to use them. These language acquirers are marked as ‘the others’. The main issue here is that for all the students who use a different language in their communities, the study of academic content via English means studying in a third language, whereas for majority students English Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM A Critical Perspective on the use of English as a Medium of Instruction at Universities 205 is a second language and offers better learning conditions; it is usually the case that proficiency in the third language is lower than in the second language, especially since second languages are learned for a longer period of time while students begin studying the ‘third’ language much later and for shorter duration. It is also the case that there is more exposure to the second language than to the third. For example, at the university where I teach, Arab students find it difficult to function in English. For them Hebrew is a second language; it is a language that they are surrounded by. Even facing the situation of having to learn in Hebrew is already a major challenge. Moreover, there may be fewer opportunities to acquire academic knowledge if the language of instruction is more remote and less familiar. It is more likely that students for whom English is a second language and not the third, have a major advantage over their counterparts for whom it is a third language. This does not mean that students who are bilinguals cannot acquire a third language with more ease than those who are monolinguals, but the proficiency in the language that for some is second and for others is third is not the same; clearly this can vary depending on the conditions, such as status of the language, quality of teaching and programs, etc. These unequal conditions can be viewed as cases of discrimination and lack of academic and language rights for the prime goal of acquiring content at university. Here too, there is a need for serious research as to the extent to which these students lose academic knowledge. Issue 3: Biases due to assessment in second languages A final issue brought up here relates to biases due to methods of assessment. In most programs where English is the medium of instruction, English is also the language of the tests which assess the achievements of the students in these academic subjects. Yet, using tests in English with students for whom English is a second or a third language can result in unjust and invalid scores on tests. Low achievements due to lack of language skills have been demonstrated in a number of studies, repeatedly showing how immigrant students reach lower scores on the tests in comparison to what their scores would have been had they been tested in tests which are bilingual, or in tests which are conducted in their stronger language. In fact, this is the major motivation behind the policy of tests accommodations mentioned above aimed at helping students to overcome the lack of proficiency in the language of these tests. Those tests often have high stake consequences for their future. There is therefore a strong policy to introduce test accommodations in all levels of education in order to have students manifest their true and ideal academic performance. In fact, their performance is often masked when the tests are being conducted via the less familiar Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM 206 Part 4: Institutional Policies at Higher Education Institutions language, in this case English, even in situations when that knowledge is being learned through EMI. Yet, there is very limited use of any accommodations for students in testing at university level, even though empirical evidence shows the effectiveness of such accommodations. In fact, a number of studies on accommodations results indicate that low proficiency in the language can harm the scores of the students on the tests. Robinson (2010) demonstrates that when students are given mathematic tests in their first language, they tend to get higher scores than they do when they are tested in the second language. In our own research (Levin & Shohamy, 2008), we found that 9th grade immigrant students who came from the former USSR and were tested in mathematics in a bilingual – Hebrew/Russian – version (each question was presented in Hebrew and in Russian) performed significantly better than an identical group of students who were tested in a monolingual Hebrew-only version. This difference lasted up to eight years and in some cases far beyond, in spite of the fact that the teaching was done in Hebrew only. Thus, the use of the students’ L1 provided them with a significant advantage on academic tests. This led to a firm recommendation to use bi and multilingual tests for immigrants. Monolingual types of assessment can be interpreted as discriminatory towards students whose proficiency in the second language is not high as it Figure 10.1 Comparison of scores between groups receiving a math test in a bilingual versus monolingual version Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM A Critical Perspective on the use of English as a Medium of Instruction at Universities 207 results in lower grades than those they would have achieved had they been tested in their L1. Furthermore, there is ample evidence that students, even when they learn via a second language, continue to rely on their first language (Levi-Keren, 2008; Logan-Terry & Wright, 2010). All this leads to recommendations as to the need to minimize bias due to the language used in the tests and to use multi/bilingual types of tests in EMI programs at university to provide students with the best opportunity to show their academic knowledge (see Shohamy, 2011 for discussion of multilingual tests). Yet, monolingual tests in English are the only way assessment is generally done in EMI academic courses, thus defying the bi/multilingual discourse that takes place in the classroom and how students actually process language via two linguistic codes. Conclusions – Toward Multilingual Languages Policies at HEIs The issues raised in this chapter relate to the extent to which the approach of EMI at HEIs can be considered an effective educational policy in academic learning. This issue was contextualized within broader language educational policies of teaching languages via academic contents. Four different settings of such policies were identified: some were considered subtractive like the teaching of a new language to immigrants and to students who use a different language in their homes and/or communities, and others additive such as CLIL and the very approach which is the focus of this book – EMI at HEIs around the world. While these four settings differ in many ways, there are great similarities among them and results of one setting can be applicable to another. Thus, research on immigrants can be considered when it comes to research on university learning; although the settings are different, these other findings can serve at least as working hypothesis for further research. This chapter raised a number of challenges and criticism regarding EMI. The first related to the extent to which the learning of academic content via a language students are not highly familiar with provides students with conditions that are not ideal for learning and acquiring academic knowledge, which is the main goal of tertiary education. There is a need to examine whether the loss of academic learning detected among immigrant students also occurs at university level. Second, questions are raised regarding EMI when English is a third language, as is the case with immigrants students and those who use different home and community languages Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM 208 Part 4: Institutional Policies at Higher Education Institutions which are different than school languages, because they are being penalized by EMI. It was argued that this approach poses unequal conditions to these students as it is difficult for them to perform as equally well as those for whom English is a second language. Finally, the third issue that was raised related to the language of the tests. Even if the teaching is delivered in English, the fact that high stake assessment is carried out in a second (or third) language means that students cannot fully manifest their academic knowledge in the tests. Thus, conducting tests in the second language can lead to invalid results in terms of the actual content knowledge that the students have in the academic subjects. This too should lead to a different policy where students are being tested in two or three languages, as all these languages together provide important resources for academic functioning (Shohamy, 2011). Nevertheless, EMI programmes continue to exist and are expanding at an accelerated rate in more countries than ever before, due to the fact that universities’ policies are driven by economic considerations. The teaching of English is deemed an attraction, not only for learning English but also for attracting students from other countries, and offering programs in English means that students from a number of other countries can enrol. This policy does not demand foreign students to acquire local and/or national languages; it is the status of English as a world language that creates this situation. Yet, this policy may harm local students, since, whereas the use of English may have advantages in terms of input, it may have detrimental effects on the academic knowledge. The implementation of EMI calls then for serious research before it is adopted at a wider scale in order to guarantee an egalatarian education. Policies of multilingualism where a number of languages are used at universities provide more smooth and fluid ways to teach content in line with current views of language learning. Currently, a number of trends in terms of language learning emerge. First, the trend towards translanguaging and the notion of ELF (English as a Lingua Franca) which refer to the recognition that English is no longer the possession of those who were born to it, but rather a language that belongs to all its speakers who develop different and unique local varieties. Thus, the English of those whose home language is Dutch is not identical to those whose home language is Kiswahili, as each ELF consists of strong elements of local languages. Thus, language mixtures, hybrids, code-mixing and different trans-languaged varieties are considered legitimate languages that enable effective communication even with those who do not share the same ELF as well as with those for whom English is a first language (Wei & Martin, 2009). The second trend is that of Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM A Critical Perspective on the use of English as a Medium of Instruction at Universities 209 multilingualism, in which English is a global language along with local/ national languages. It is becoming clear that more languages are needed for functioning in global societies. Thus, regional lingua francas as Chinese, Arabic or Swedish are being encouraged along with local languages of various communities. The main belief nowadays is in multilingualism and the use of EMI at HEIs in fact defies this goal. The use of a number of languages may be a more effective policy at HEIs. References Abedi, J. (2004) The No Child Left Behind Act and English language learners: Assessment and accountability issues. Educational Researcher 33 (1), 4–1. Abedi, J. (2009) Utilizing accommodations in assessment. In E. Shohamy and N.H. Hornberger (eds) Encyclopedia of Language and Education, 2nd Edition, Vol 7: Language Testing and Assessment (pp. 331–347). Berlin: Springer. Cushing Weigle, S. and Jensen, L. (1997) Issues in assessment for content-based instruction. In M.A. Snow and D. Brinton (eds) The Content-based Classroom: Perspectives on Integrating Language and Content (pp. 201–212). White Plains, NY: Longman. Curry, M.J. and Lillis, T. (2004) Multilingual scholars and the imperative to publish in English: Negotiating interests, demands and rewards. TESOL Quarterly 38 (4), 663–688. Doiz, A., Lasagabaster, D. and Sierra, J.M. (2011) Internationalisation, multilingualism, and English-medium instruction. World Englishes 30 (3), 345–359. Hu, G. (2008) The misleading academic discourse on Chinese-English bilingual education in China. Review of Educational Research 78 (2), 195–231. Johnson T.K. and Swain, M. (1997) Immersion Education: International Perspective. Cambridge, UK: Cambridge University Press. Lasagabaster, D. and Sierra, J. M. (2010) Immersion and CLIL in English: More differences than similarities. ELT Journal 64, 376–395. Levi-Keren, M. (2008) Factors explaining biases in Mathematic tests among immigrant students in Israel. PhD dissertation, Tel Aviv University [in Hebrew]. Levin, T., Shohamy, E. and Spolsky, B. (2003) Academic Achievements of Immigrants in Schools. Report submitted to the Ministry of Education. Tel Aviv University [in Hebrew]. Levin, T. and Shohamy, E. (2008) Achievement of immigrant students in Mathematics and academic Hebrew in Israeli school: A large scale evaluation study. Studies in Educational Evaluation 34, 1–14. Lillis, T. and Curry, M.J. (2010) Academic Writing in a Global Context: The Politics and Practices of Publishing in English. New York: Routledge. Logan-Terry, A. and Wright, L. (2010) Making thinking visible: An analysis of English language learners interactions with access-based science assessment items. AccELLerate 2 (4), 11–14. Phillipson, R. (1992) Linguistic Imperialism. Oxford, UK: Oxford University Press. Papapavlou, A.N. (2001) ‘Mind your speech’: Language attitudes in Cyprus. Journal of Multilingual and Multicultural Development 22 (6), 491–501. Rea-Dickens, P., Yu, G. and Afitska, A. (2009) The consequences of examining through an unfamiliar language of instruction and its impact for school-age learners in Sub-Saharan African school systems. In L. Taylor and C.J. Weir (eds) Language Testing Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM 210 Part 4: Institutional Policies at Higher Education Institutions Matters: Investigating the Wider Social and Educational Impact of Assessment. Studies in Language Testing 31 (pp. 190–214). Cambridge: Cambridge University Press. Rea-Dickens, P., Yu, G., Afitska, O., Olivero, F., Ingram, N, Erduran, S., Khamis, Z., Mohamed, A., Mbaraok, M., Mwevura, H. and Said, S. (2011) Investigating the Language Factor in School Examinations: Exploratory Studies. SPINE Working Papers (2). Bristol: University of Bristol. Accessed 16 May 2011. http://www.bristol.ac.uk/ spine/publication%20and%20reports Robinson, J. (2010) The effects of test translation on young English learners’ mathematics performance. Educational Researcher 39 (8), 582–590. Shohamy, E. (2007) Reinterpreting globalization in multilingual contexts. International Multilingual Research Journal 1 (2), 127–133. Shohamy, E. (2011) Assessing multilingual competencies: Adopting construct valid assessment policies. Modern Language Journal 95 (3), 418–429. Swain, M. and Johnson, R.K. (1997) Immersion education: A category within bilingual education. In R.K. Johnson and M. Swain (eds) Immersion Education: International Perspective (pp. 1–16). Cambridge, UK: Cambridge University Press. Swain, M. and Lapkin, S. (2005) The evolving sociopolitical context of immersion education in Canada: Some implications for program development. International Journal of Applied Linguistics 15, 169–186. Thomas, W. and Collier, V. (2002) A national study of school effectiveness for language minority students’ long-term academic achievement: Final report. Project 1.1. Santa Cruz, CA: Center for Research on Education, Diversity and Excellence (CREDE). Wei, L. and Martin, P. (2009) Special issue: Conflicts and tensions in classroom codeswitching. Bilingual Education and Bilingualism 13. Brought to you by | University of Groningen Authenticated Download Date | 12/1/19 2:30 PM