From EMI to CLIL

EduLingua 4/1 (2018) 51 From EMI to CLIL: Methodological strategies for bilingual instruction at university1 Jesús Ángel González and Javier Barbero Andrés University of Cantabria DOI:10.14232/edulingua.2018.1.5 We report on a small-scale study carried out in the University of Cantabria, Spain, which investigated teaching strategies and methodology for CLIL at the tertiary level. The study was carried out in two stages: in the first one we designed a questionnaire for primary and secondary school teachers, and interviewed 10 CLIL teachers. The focus was on teaching strategies and methodology, with the explicit aim of collecting advice applicable to the tertiary level. In the second stage we have tested these tips in a university context through a second questionnaire and round of interviews. We have interviewed 10 EMI university lecturers and asked them about methodological strategies, and specifically about the advice suggested by their colleagues. Our findings suggest that CLIL teachers from lower levels are well acquainted with CLIL methodology and that EMI practitioners at university find their advice very useful, although they show reluctance towards deeper methodological changes. Key words: CLIL, EMI, teaching strategies, methodology, bilingual education 1. Introduction: CLIL and EMI Content and Language Integrated Learning (CLIL) has been defined as “a dual-focused educational approach in which an additional language is used for the learning and teaching of both content and language” (Coyle, Hood, & Marsh, 2010, p. 1). Since it was born in a primary and secondary school context, it is only natural that most of the research carried out about it has taken place in those two levels and not at a tertiary level, but this has been changing slowly in the last few years. Wilkinson (2004) and Fernández (2009), and, more recently, Fortanet (2013), Lasagabaster and Doiz (2016), O’Dowd (2018), and Schmidt-Unterberger (2018) have considered the possibility and challenges of implementing a CLIL approach at the tertiary level, or ICLHE (Integrating Content and Language in Higher Education), as CLIL at university is frequently called. This ICLHE-focused research is, in fact, becoming more and more necessary due to the increase of English-only programs and bilingual programs in public and private universities in Europe. Wachter and Maiworm (2014) documented this increase with a longitudinal study that shows an exponential increase in programmes taught in European Higher Education Institutions: from 700 programmes in 2002 to 8,089 in 2014. Similarly, a survey of 70 European universities by O’Dowd (2018) found that only 7% stated that they were offering no courses at all in English. According to Wilkinson, the fact that English is “increasingly gaining ground as the language of 1 This research was supported as part of the project Bilingual teaching and learning in Cantabria: From primary to tertiary education, funded by the University of Cantabria, UC 2016-GRE-10. 52 González and arbero From E to L L Methodological strategies instruction in universities” has started to provoke resistance even in English-friendly countries like the Netherlands and has fostered “a massive increase in academic studies on the phenomenon of English-taught degree programmes and courses” (2018, pp. 1-2). As Coleman phrased it, English is “the language of higher education in Europe” (2006, p. 1), and Dearden (2015) has shown that we can find a similar trend in the rest of the world. This increase means that bilingual teaching affects now a wider spectrum of students, including those who do not have the necessary proficiency to follow cognitively demanding University-level classes in English. As a result, in places like Southern Europe where previously EMI (English Medium Instruction) practitioners did not have to consider a different methodology, now CLIL (or ICLHE) has become necessary to make complex content available to students with a limited knowledge of English. According to Wachter and Maiworm (2014), the spread of EMI instruction in Southern Europe is not as extensive as in Central and Northern Europe, possibly as a result of a lower level of English proficiency. In the specific case of Spain, however, national and regional policies (including the generalization of bilingual teaching) have been implemented in the last few years to improve English proficiency in primary and secondary schools. Similarly, “the number of EMI programmes in [Spanish universities] has clearly increased in the last five years as have the international students attending Spanish HEIs [higher education institutions]” (Dafouz, 2018, p. 5). The combination of this increase and a comparatively low English proficiency among the students has meant that many EMI programmes in Spain need to keep the dual focus on language and content typical of CLIL (Dafouz, 2018). Before moving on to the description of our study, it is important to establish the terminological differences between EMI and CLIL, as well as the methodological implications of CLIL. In general, we can say that EMI is used in tertiary settings in countries where English is not the mother tongue, as shown in this definition: EMI (English Medium Instruction) entails the use of the English language to teach academic subjects in countries or jurisdictions where the first language (L1) of the majority of the population is not English. It may or may not include the implicit aim of increasing students’ English language abilities, (Brown & Bradford, 2017, p. 330, based on Dearden, 2015, our italics) In contrast, CLIL tends to be used in primary and secondary schools where the final goal is to develop language proficiency and content knowledge: CLIL is an approach to education that integrates language and content learning; planning for, fostering, and assessing both, though the focus may shift from one to the other. CLIL is also a method of teaching which draws heavily on constructivist and socio-cultural notions of learning to provide students with opportunities for meaningful input and output in L2 and meaningful engagement with content (Brown & Bradford, 2017, p. 331, our italics) EduLingua 4/1 (2018) 53 As summarized in Table 1, EMI refers only to University programmes taught in English (no other language), it has a more global dimension than CLIL (which is mainly European), and it may have ideological implications. Besides, and this is important for our purposes here, it may ignore language learning completely, since its main aim is the internationalization of the University (a top-down process), that is to say bringing in students from abroad who are supposed to be already proficient in English. In contrast, CLIL is used for a variety of languages (not only English), it was developed in Europe (where we can find most of the examples of CLIL programmes), and it was at first a bottom-up local or regional process, even though now we can find more examples of top-down programmes. Since it was created in order to deal with students who were not proficient in their L2, one of the objectives, together with content, has always been language learning. As a result, it has become “more than an approach; a widely accepted method of teaching CLIL courses has emerged” (Brown & Bradford, 2017, p. 330). Table 1 EMI vs. CLIL EMI • English only (not L1) • Global (ideological implications) • Top-down (internationalization) • Focus on content, not language • No particular interest in methodology CLIL • Several languages • Mainly European. • Bottom-up at first, later more topdown. • Double focus on content and language. • Specific methodology So, what does this specific CLIL method or methodology imply? Brown and Bradford have summarized some of its main features: CLIL lessons are expected to engage students in the four communicative skills and to balance the students’ receptive and productive experience with the target language; teachers approach language lexically rather than grammatically; content is based on a curriculum rather than on topics of general interest or current affairs; classes need to pay attention to four key elements of CLIL teaching, known as the four Cs of CLIL (content, communication, cognition, and culture) and, finally, CLIL lessons rest on constructivist principles of active and cooperative learning and the co-construction of knowledge (Brown & Bradford, 2017, pp. 330-331, based on Coyle, Hood, & Marsh, 2010). To these elements, we should add the following (related or inter-related to the previous ones): 1) Teachers should provide context to context-reduced language tasks (Halbach, 2012) 2) Since the priority of CLIL is the content rather than the language (Ball, Kelly, & Clegg, 2015, p. 36; the term they use for this is “conceptual fronting”), it is necessary to make key language salient by embedding and scaffolding students’ 54 González and arbero From E to L L Methodological strategies performance through modelling, posing of questions or breaking up the tasks to make them more manageable (Halbach, 2012) 3) Use visual support (images, diagrams, slides) 4) Support student output (Ball, Kelly, & Clegg, 2015, p. 44) 5) Following constructivist principles, focus on the students’ experience, and therefore activate prior knowledge (Dale, Van der Es, & Tanner, p. 2010) The authors of this paper considered, after revising the EMI and CLIL literature reviewed previously, that rather than simply informing EMI teachers at our institution about the teaching strategies inherent in CLIL methodology, it would be very interesting to build bridges between different levels of education, take advantage of the experience of our colleagues in primary and secondary education and ask them which of these methodological principles they thought would be important for their university colleagues. The aim of the study, therefore, was not to ask them about the principles of CLIL methodology (which we have summarized above), but which teaching strategies related to “the CLIL method” could be applied at university in classes were content is more demanding from a cognitive point of view and where students’ English proficiency is higher than in secondary school but often lower than required for those tasks. 2. The study 2.1 Context to the study The institution where this research has been carried out is the University of Cantabria (UC), a medium-sized institution in Northern Spain that, like many others in Europe, has been increasing the number of courses in English noticeably in the last few years. The UC is the only public HEI in the small autonomous region of Cantabria. With a student body of approximately 10,000, it is a research-oriented university that receives around 400 incoming students every year and sends out a similar number. It is part of several international networks (Santander, Compostela, AUIP, CINDA) and is very active internationally not only in Erasmus+, but also in bilateral agreements with universities from the USA and Latin America. The increase of courses in English at the UC is part of an internationalization strategy aimed both at the attraction of Englishspeaking students and at the “internationalization at home” of Spanish students, who are required to hold a B2 certificate in English in order to graduate. As a result, the University offers a total of 152 courses in English in all its degrees, taught by lecturers who are required to have a C1 level in English in order to be part of the EMI programme. There are two types of courses taught in English: those specifically EduLingua 4/1 (2018) 55 designed for exchange students (normally students with a B2-C1 level of English or even native English speakers) and courses which are part of regular Bachelor or aster’s Degrees (and therefore mostly taken by Spanish students with a proficiency in English ranging approximately from A2+ to C1, but with a high percentage of A2-B1 students). Although Schmidt-Unterberger argues against the use of CLIL or ICLHE in higher education settings, due to the fact that “language development is not amongst the set learning objectives” (2018, pp. 2-3), this is not the case in our university. The A2-B1 students are strongly motivated to reach B2 level, but their English is hardly enough to follow cognitively demanding contents in English, which is why a CLIL methodology seems to be the most appropriate in this context. These EMI courses are offered in the third or fourth year, so most of the students’ ages range roughly from 20 to 25. 2.2 Materials and methods The study was carried out in two stages, as part of two different projects of teaching innovation at the University of Cantabria. In the first stage (2011-2012), whose description and results have been published in González and Barbero (2013), we designed a set of questions for primary and secondary school teachers, with 16 questions going from the more general aspects of CLIL to specific questions about methodology and, more specifically, methodological advice for university teachers. We then interviewed 10 teachers working in primary and secondary schools, both public and state-subsidized schools. In the sample we included both CLIL teachers (teaching 7 different subjects) and English-language teachers taking part in CLIL programs, but all of them with a reasonable experience with CLIL (from 2 to 12 years). The set of questions was the basis for the semi-structured interviews, which were carried out by the authors in English or Spanish (depending on the participants’ mother tongue), and following c racken’s suggestions about the “long interview” (McCracken, 1998). They were then transcribed by an assistant, and coded into categories and analysed in detail by the authors. Taking into account all the teachers’ answers, we developed a summary of their advice, which we called a methodological decalogue for university lecturers in bilingual programmes (González & arbero, 2013, p. 18). In the second stage (2012-2017), we tested the decalogue in a university context in two different ways: first, in 8-hour training courses for teachers in six different instances from 2012 to 2017 (a course called “CLIL: Methodological Orientations for a Better Practice” supported by a blog called “From Bilingual Teaching to CLIL” that can be found at http://bilingualteachingunican.blogspot.com), where the attendants provided informal feedback about the decalogue; and then, in a second round of interviews with 10 university lecturers with experience in EMI from very different backgrounds and colleges (see Table 2 below). 56 González and arbero From E to L L Methodological strategies Table 2 Colleges and courses Colleges • Education • Electrical Engineering • Civil Engineering • Chemical Engineering • Medicine • Mathematics • Economy Courses • hildren´s Literature and Literary Education • English Phonetics • English Acquisition as a Foreign Language • Foreign Language Teaching and Learning • Research Methodology in Language Acquisition • Electrical Drives • Energy Systems • Molecular Biology of the Cell • Fundamental Pharmacology • Drug Dependence and Addiction • Statistical Inference • Macroeconomics • International Business: a European Perspective • International Business Economics: Modelling and Simulation • European Economic Policy • Open Economy Macroeconomics Excluding two courses created by the university for exchange students with a better level of English, all the other courses taught by our participants are part of a achelor’s or aster’s degree and therefore mainly attended by Spanish students with various levels of English, as mentioned before. The 10 University lecturers teaching these courses include 7 men and 3 women with a variety of linguistic and professional backgrounds. Their ages range from 30 to 65, they all had had some previous experience teaching in English (both in Spain and abroad), and, with the exception of a native English speaker and an English philologist with a C2 in English, they all have a C1 in English, with more self-declared proficiency in written than oral skills. Class size is highly variable (from 10 to 60 students), as well as the teaching conditions: some of the classes are lectures, whereas others include laboratory practice or the possibility of splitting groups for more practical sessions. This kind of variety highlights one of the basic difficulties when implementing CLIL at a tertiary level, namely the heterogeneity of degrees and teaching contexts. In the first questionnaire (for primary and secondary school teachers) we inquired about different topics related to their teaching practice, such as linguistic competence, teacher training, experience as a CLIL teacher, theoretical knowledge about CLIL, materials, cooperation between language and content teachers, skill balance, general difficulties, advantages of CLIL, or use of ICT, but the main focus was on methodology and teaching strategies, the final question being: “What kind of methodological advice would you give to university lecturers who would like to implement CLIL at the University of Cantabria?” (the complete set of questions was published in González & Barbero, 2013). EduLingua 4/1 (2018) 57 The main focus of our set of questions for university lecturers (see Appendix) was the decalogue, but we offered the participants the chance to speak about their experience more freely. Therefore, the participants also spoke about the EMI program at UC and about general questions of organization and educational policy, like the professional or economic benefits of being involved in such a program, or the main difficulties that they encountered teaching in English. All the semi-structured interviews were carried out in English or Spanish (depending on the participants’ mother tongue), and they were recorded and subsequently transcribed using a procedure similar to the one used in the first stage. They are confidential and anonymous and each took approximately 30 minutes. 3. Results 3.1 First stage: primary and secondary school teachers’ recommendations concerning CLIL The most important results we obtained in the first stage were the methodological recommendations, which we summarized in the aforementioned methodological decalogue, already published in our 2013 study: 1) Communication is a must. English should be used as much as possible, but the mother tongue can also be used in case of communication blockage. “Translanguaging” (Kim, 2018) is perfectly acceptable. 2) Scaffolding is essential. Identify language demands and provide support strategies. Use visual aids and written language whenever necessary. All students, but particularly all those whose listening skills are not the best, will appreciate the use of slides summarizing the main ideas stated in class. Model and break up the tasks if appropriate. 3) A reference lexical corpus is required for every task. Advance work (with warm-up activities like video comprehension, Webquests or the like) on specific vocabulary should be done prior to the explanation of cognitively challenging content. 4) Use ICT, in particular software and on-line material, in English. 5) Use a student-centred approach. Put yourself in the students’ position. Provide the opportunity for as much hands-on learning as possible. Use pair work and group work. 6) In assessment, content should be a priority over language: linguistic competence in the foreign language is an added value which should be rewarded, but the lack of fluency in the foreign language should not be a major obstacle for a positive evaluation 58 González and arbero From E to L L Methodological strategies 7) Use diverse assessment instruments: self-assessment, peer assessment, rubrics, and language and content portfolios. 8) Repeat and consolidate. Present information in different formats. 9) Plan carefully in order to be flexible. 10) Turn problems into opportunities. Be bold as far as methodology is concerned and take advantage of this new educational context to work on a different paradigm. Teachers are facilitators and mediators between language and content, not mere transmitters of knowledge. Assess your own teaching practice (with instruments like the EPOSTL, or “The CLIL Teachers’ ompetences Grid”). (González and arbero, 2013, p. 18) We can see that most of these tips (like the emphasis on lexis, the scaffolding techniques, or the constructivist student-centred approach) also appear in the summary of CLIL methodology that we have provided in section 1, which means that in general the teachers we interviewed are well acquainted with CLIL teaching techniques and that they believe that most of these principles and techniques can also be applied in tertiary Education. As an example, we can see some of the comments provided by the teachers: Participant #1: A CLIL teacher must be really open to new methodological approaches and to new ways of assessment. The thing is: how do you assess students who are studying your content through the foreign language? And how do you facilitate that assessment process? Many times you have to put yourself in the place of the student, which is something we teachers are not very used to doing. Participant #6: They should forget about traditional teaching. I feel sorry for those professors who try to preserve that approach!! Student groups should be smaller, classes should be based on debate-based continuous interaction. Participant #7: Professors should not take for granted that their students have the level of English they are supposed to have. Participant #8: Visual aids are really important, if students have that kind of material they understand everything much better and in that sense ICTs are essential. Participant #9: They should look for more practical ways to show content to students. Traditional teaching simply does not work and ICT should have a very relevant role together with collaborative work. 3.2 Second stage: university lecturers’ responses about CLIL implementation In the second stage, we recorded the reaction of our university participants to the decalogue provided by their colleagues. In general, we can say that the university lecturers reacted favourably to the suggestions. We have organized their reactions, EduLingua 4/1 (2018) 59 depending on their agreement with their recommendations, in three different groups: complete agreement, partial agreement, and disagreement. 3.2.1 Complete agreement with the recommendations All the participants showed complete agreement with the following recommendations: 2, 3, 4, 6, 8 and 9. They all use PowerPoint slides and other forms of visual support such as images or diagrams, although some of them were not familiar with the concept of scaffolding. Participant #2, for example, mentions several forms of methodological support: “The methodology should be different. For example, we should use more linguistic simplification and visual aids, more repetitions and paraphrasing, as well as more interactive strategies” (#2). All the interviewees place particular emphasis on lexis, and 6 of them provide the students with a glossary or another form of lexical support (#3). They all use ICT, particularly Moodle and the internet, where they find worthwhile teaching material such as videos or presentations (#4). Since they consider their job to teach content, they don’t worry too much about language accuracy. In three cases, they openly admitted that they do not care about language accuracy at all; much more so for assessment, where content is the only priority. Participant #4 mentions some of the strategies used in order to minimize language-related mistakes in assessment: “We normally deliver multiple choice tests and short answer tests. We don´t take into account the possible linguistic mistakes they may have. They don´t tend to have difficulties in these tests. Oral presentations are different, the higher their level is, the better for their presentations, we must admit that definitely influences their final marks…” (#6). They all tend to repeat things in different formats: Four of them mentioned combining theory and practice in order to consolidate content and skills (#8). While all of them spend a long time planning the classes, they are also ready to be flexible and use more time than previously planned if they feel that students have not understood a certain point (#9). Let us remember that in most of these university courses, there is much more freedom than in subjects of primary and secondary education, which are more constricted by a common curriculum, so in that sense lecturers can afford to be more flexible than their colleagues: Three actually mention that you need to skip topics when you teach in English, but that does not seem to be an unsurmountable problem. 3.2.2 Partial agreement with the recommendations The participants have shown partial agreement with tips number 1, 7 and 10. While they all agree that communication is essential, four of them said that they only use English (#1). Let us remember that there are many types of courses taught by these lecturers, and in some of them, they teach exchange students more proficient in English than in Spanish. While most teachers use a variety of assessment procedures (in our University, 60 González and arbero From E to L L Methodological strategies only 60% of the final mark can be based on a final exam, so they all need to use other forms of continuous assessment), half of them feel reluctant to use self-assessment or peer-assessment techniques. Participant #6, for example, says that, “I think that the idea of continuous assessment and alternative forms of assessment is not serious. In the real world, you risk your neck every day. What I need to make sure is that at the end of the day they master some basic statistical techniques. That’s it” (#7). Although they like the sentence “turn problems into opportunities”, they are not so convinced about the implications. More specifically, six participants say that they are not convinced that teaching in English means that they are working in a new educational context or a different paradigm. Participant #4, for example, points out, “When I teach in English, I don’t think I do anything differently. The content is the same, and apart from maybe checking more often that they are following you, I do the same things”; and participant #1 does not think that there are important methodological changes “I think that the methodology is similar, but the rhythm changes. The students need more time and space to assimilate things” (#10). 3.2.3 Disagreement with the recommendations Finally, the tip they have agreed with the least is Tip #5 (“use a student-centred approach”) which is the one most related to the constructivist-collaborative approach and the one that would imply the deepest changes in methodology. Nine of them think that they would like their students to work in pairs or groups, but they say they cannot do it in their classes, in some cases because the classes are overcrowded, in others because of the characteristics of the content that they have to teach. Eight lecturers mention that they are very reluctant about the student-centred approach, and the teacher’s role as facilitator. n fact, three of the participants blame the students themselves, because they do not accept it or do not want to change their traditional, more passive role. This reluctance also crops up in other parts of the questionnaire. For example, although nine participants are familiar with the term CLIL, to the question: “Do you think that teaching in English implies not just code-switching but also a methodological change?” four consider EMI as simply changing the language of teaching. And when asked about teaching techniques, the commonest one mentioned by the teachers is still the lecture, used frequently by eight of the participants and occasionally by two. The following two comments illustrate the teachers’ reluctance with the student-centered approach: Participant #10: I can see the teachers changing their role as facilitators, but not the students changing their more passive role, because they expect to have things done for them, and work as little as possible. And this has nothing to do with the language… I think that people ask too much from teachers, when it should actually be the opposite. EduLingua 4/1 (2018) 61 Participant #9: I agree with the student-centred approach, you should teach students how to learn on their own, but it doesn’t happen. The students don’t accept it, because they think that what they are doing is not being validated by their teacher. 3.2.4 General questions about the EMI program As shown before, we also asked the participants about other more general questions related to EMI, or the particular organization of EMI at their university, and their answers are in general consistent with the literature about EMI and CLIL. Most of them are quite happy with the program and they mention personal reasons like professional development (teaching specialization), personal fulfilment, or the satisfaction of having overcome a challenge, although they also mention the reduction in teaching load, which, at the time, meant that credits taught in English counted twice as much as credits taught in Spanish. The positive aspects of EMI they mention most often are the improvement in students’ English proficiency (8 lecturers), and the internationalization of the students’ body (6). As an example, we can see this comment from one of the participants: Participant#5: I´m very happy with this program. Personally, it gives me a lot and I really think we should keep on working on it all in the long term. As to the problems related to teaching in English, material production does not seem to be an issue, since most of the bibliography used in those courses is already in English. As far as assessment is concerned, most of the participants have found ways to minimize the effect of students’ low English proficiency in their assessment, as mentioned before: multiple-choice tests, lab tests, or other forms of assessment with limited L2 production. The participants complain about lack of training (both methodological and linguistic, even though all of them have at least a C1 level), about the students’ level of English, and about organizational issues related to budget restrictions, like the number of students in class or professional compensations. From a more holistic point of view, some of them wonder about the coherence of teaching in English in Spain, and whether having just a few courses in English in a degree is enough. The following comment illustrates this point: Participant #7: Maybe we’ve gone too far with this obsession with English. Why should everything be taught in English? I think we should rethink it all and filter a little bit what kind of content or subjects we want to teach through English. Sometimes students have the perception that passing subjects in English is easier as teachers tend to be more lenient, and I´m talking about Medicine studies…!! 62 González and arbero From E to L L Methodological strategies 4. Discussion Among the limitations of the study we first need to mention the small scale and range: it is based on a limited number of participants from only one university. However, we have found considerable agreement in the opinions expressed by our participants. In the first stage of our research, we were able to establish that CLIL practitioners at primary and secondary schools have a very clear understanding of the methodology and teaching techniques implied by CLIL, and that they think that most of these strategies are applicable at a tertiary level. In the second stage we have been testing these tips in a university context through a second set of questions and round of interviews. We have seen that EMI teachers at our university found the decalogue based on the experience of their colleagues from primary and secondary education very useful for their teaching practice. They showed complete agreement with 60% of the tips and partial agreement with 30%. In addition, they seem to be very conscious of the students’ linguistic needs and they try to help them to learn the content by using conceptual fronting, repeating, scaffolding, embedding, making language salient, and using ICT. However, although our findings suggest that EMI practitioners at our university seem to be willing to teach in English, and are ready to change many aspects of their teaching methods, they do not seem to be ready to make the deeper changes necessary to implement a CLIL methodology in tertiary education, L L’s “enriched pedagogy” which is “its major achievement” according to Ball, Kelly & Clegg (2015, p. 4). Therefore, we have to agree with O’Dowd (2018) (“lecturers themselves often do not believe in or see this need for a methodological shift … to more student-centred approaches”, p. 3) and Dearden and Macaro (2016) who detected a “distinct lack of awareness of a need to change pedagogy in order to help students … to cope with content delivered through a second language”, although they found that instructors seem to be “relatively open to the need for a revised pedagogy” (p. 479). The implications of these findings seem to be far-reaching. First of all, it is important that both university administrators and lecturers become aware of the fact that teaching in a foreign language is a process that involves not simply changing the language of instruction, but also the manner in which the classes are managed, prepared and taught. This means that, in line with O’Dowd (2018), Dafouz (2018), and Wilkinson (2018), we think there is a need of specific training for the instructors taking part in these programs. This training should involve not only linguistic tools, but also, and more importantly, a redefinition of teaching and assessment methods. Another implication is that both language teachers and researchers should help to build more bridges between different levels of education. We tend to think of knowledge transfer as a top-down process (from higher education to lower levels), but our study suggests that there is much to be learned from bottom-up processes too, particularly if we focus on pedagogy and teaching techniques. EduLingua 4/1 (2018) 63 5. Conclusion This article has presented the findings of a two-stage study which has aimed at building a bridge between different levels of education. Having detected a need for a change in teaching strategies in our EMI context, we decided to ask the CLIL practitioners for advice and therefore designed a set of questions for primary and secondary school CLIL teachers. The results of this first stage (the recommendations summarized in a decalogue of methodological tips) were found very useful by the university lecturers, who are very much aware of the need to make cognitively demanding content available to students with a limited proficiency in English. Having said this, the lecturers do not seem to be ready to go beyond conventional lecture-based teaching and embrace a CLIL student-centred approach. This paper hopes to raise awareness of the need to incorporate the best contributions of the CLIL method in EMI contexts where the students have a limited proficiency in English. References Ball, P., Kelly, K., & Clegg J. (2015). Putting CLIL into practice. Oxford: OUP. Brown, H., & Bradford, A. (2017). EMI, CLIL & CBI: Differing approaches and goals. In Clements, P., Krause, A., & Brown, H. (Eds.), Transformation in Language Education (pp. 328-334). Tokyo: JALT. Coleman, J. A. (2006). English-medium teaching in European higher education. Language Teaching, 39(1), 1-14. Coyle, D., Hood P., & Marsh, D. (2010). CLIL Content and Language Integrated Learning. Cambridge: CUP. Dafouz, E. (2018). English-medium instruction and teacher education programmes in higher education: ideological forces and imagined identities at work. International Journal of Bilingual Education and Bilingualism, 21(5), 540-552. https://doi.org/10.1080/13670050.2018.1487926. Dale, L., Van der Es W., & Tanner, R. (2010). CLIL Skills. Leiden: ICLON. Dearden J. (2015). English as medium of instruction: A growing global phenomenon. London: British Council. Dearden, J., & Macaro, E. (2016). Higher education teachers’ attitudes towards English Medium Instruction: A three-country comparison. Studies in Second Language Learning and Teaching, 6, 455-486. Fernández, D. J. (2009). CLIL at the university level: Relating language teaching with and through content teaching. Latin American Journal of Content and Language Integrated Learning, 2(2), 10-26. Fortanet, I. (Ed.). (2013). CLIL in higher education: Towards a multilingual language policy. Bristol: Multilingual Matters. 64 González and arbero From E to L L Methodological strategies González, J. A. & arbero, J. (2013). Building bridges between different levels of education: Methodological proposals for CLIL at university. Language Value, 5(1), 1-23. Halbach, A. (2012). Adapting content subject for bilingual teaching. Encuentro, 21, 34-41. Kim, S. (2018). “It was kind of a given that we were all multilingual”: Transnational youth identity work in digital translanguaging. Linguistics and Education, 43, 39-52. McCracken, G. (1988). The long interview. Thousand Oaks: Sage. Lasagabaster, D., & Doiz, A. (eds.). (2016). CLIL Experiences in secondary and tertiary education: In search of good practices. Bern: Peter Lang. O’Dowd, R. (2018). The training and accreditation of teachers for English Medium Instruction: an overview of practice in European universities. International Journal of Bilingual Education and Bilingualism, 21(5), 553-563. https://doi.org/10.1080/13670050.2018.1491945. Schmidt-Unterberger, B. (2018). The English-medium paradigm: A conceptualization of English-medium teaching in higher education. International Journal of Bilingual Education and Bilingualism, 21(5), 527-539. https://doi.org/10.1080/13670050.2018.1491949. Wächter, ., & aiworm F. (eds.). (2014). English-taught programmes in European higher education. Bonn: Lemmens. Wilkinson, R. (ed.). (2004). Integrating content and language. Meeting the challenge of multilingual higher education. Maastricht: Universitaire Pers Maastricht. Wilkinson, R. (2018). Content and language integration at universities? Collaborative reflections. International Journal of Bilingual Education and Bilingualism, 23(5), 607615. https://doi.org/10.1080/13670050.2018.1491948.