HIGHER EDUCATION STUDENTS’ PERSPECTIVES ABOUT EMI IN A SOUTH BRAZILIAN PUBLIC UNIVERSITY 1

: This work aims at investigating Brazilian students’ views on English as a Medium of Instruction (EMI) from a pre-implementation standpoint so that their interest, linguistic and pedagogical needs, concerns and suggestions are identified. Through online questionnaire, we identified a significant interest in EMI, associated to concerns with language support and the preference for a flexible regulation, considering the interest and needs of each area and department. Most envisage EMI as an opportunity to prepare for mobility and to compensate part of the experience to those who would not be able to afford it. The results contribute to an assessment of the feasibility of EMI and a local language policy aware of the necessary language support, professional development and offer regulation.


Introduction
Brazilian Higher Education scenario has been changing due to internationalization, and the pedagogical practices have been reassessed and redesigned in some specific institutions. The launching of Science without Borders 2 in 2011, a student mobility program that introduced a new set of academic literacies needs in additional languages, particularly in English, marked the starring of a countrywide internationalization process in Brazil's tertiary education. Between 2011 and 2015, Science without Borders (hereafter, SwB) offered 64,000 a language policy because it addresses additional language needs and interests of students, administrative staff and professors of UFSM (Author 3, 2018).
The academic community that participated in the elaboration of this document frequently mentioned the availability of curricular courses delivered in English as a strategy to overcome the internationalization challenge, a practice that can be understood under the scope of English as a Medium of Instruction (EMI). EMI consists of teaching curricular courses in English at university settings where the first language is not English (Dearden, 2015) and one of the reasons for its adoption in universities worldwide has been to attract incoming mobility.
Considering SwB, the program was distinguished by: • favoring outgoing mobility; 3 To stay 18 months in the foreign country/university, the exchange had to 'include a language course' (Ciência sem Fronteiras, 2019b). 4 The first SwB public calls that offered scholarships to universities in Portugal and Spanish (in 2012) received substantially more applications -12,126 and 9,918 respectively -than the first SwB public call to universities in the United States of America (in 2011) -7,997. Portugal was excluded from the SwB in 2013 after the second public call to Portugal received 28.191 applications, the highest number of applications in a SwB public call.

Fólio -Revista de Letras
Vitória da Conquista v. 14, n. 2 jul./dez. 2022 • revealing low proficiency in English of Brazilian university students; and • being limited to a very small percentage -0.9% -of the 6,739,689 undergraduate students enrolled in universities in 2011 (considering only public universities, it represents 3.6% of the 1,773,315 students) 5 We consider that the research on EMI as a potential practice at UFSM could work to fill gaps in all three mentioned aspects as it can • encourage incoming mobility; • contribute to improve English proficiency of UFSM students; and • be an opportunity of an internationalization experience 'at home' (Beelen and Jones, 2015) to those students who will not be able to participate in international student mobility.
Author 3 (2018) stated that the salience given to EMI by the community might steer the university's concern towards the use of English in class and challenge English language teaching experts at UFSM to offer pedagogical support. The absence of any specific policy at UFSM in relation to EMI to regulate the implementation of this practice also suggests the relevance of research about the adoption of EMI at UFSM that could inform its institutionalization. It is in this niche that our research group has been investigating (Author 3, 2018).
To the best of our knowledge, most available works assess students' perspective at a post EMI implementation period, when they have already enrolled and/or after concluding an EMI course. There seems to be little investigation regarding this public's opinion before institutionalizing the practice, consulting students' interest and needs in order to turn them collaborators in the construction of an institutional language policy. One exception would be Morell et al. (2014). Further to that, Macaro et al. (2018)

Internationalization and EMI on Language Teaching
Various approaches to the teaching of curricular subjects and/through additional languages have arisen, many of which have contributed to the strengthening of Internationalization at Home (hereafter, IaH). IaH is an alternative to traditional Internationalization, because it is conceived as a process of internationalization that can be conducted at the local university environment (Beelen and Jones, 2015) as opposed to an understanding of internationalization as mobility only. In fact, Internationalization is a process which refers to an 'integrating an international, intercultural, or global dimension into the purpose, functions or delivery of post-secondary education' (Knight, 2004 quoted in Beelen andJones, 2015: 60) in such a way that although it includes mobility, it does not limit itself to this aspect.
According to Corrales, Paba Rey and Santiago-Escamilla (2016: 322), 'while IaH does not require that courses be taught in a particular language in order to become internationalized (Beelen & Jones, 2015), more and more universities have shifted to using a common language' in this process, and EMI 'parece estar recebendo mais atenção nos contextos de educação superior 6 ' (Baumvol and Sarmento, 2016: 73).
EMI, as understood by our group, is a comprehensive kind of use of English focused as a teaching language for academic courses/subjects others than English itself without a specific interest on language learning aims in regions where it is not the first language of the majority of people (Dearden, 2015;Macaro et al., 2018;Schmidt-Unterberger, 2018;Dafouz, 2021). Different from a defense for an English-only policy, we recognize its importance on a globalized world and more specifically on an increasingly globalized/internationalized education connected to diversity of research and pedagogy agendas in which a framework of English-Medium Education in Multilingual University Settings (EMEMUS) (Dafouz and Smit, 2016;2020) seems advisable. For the purpose of this research, we will continue to refer to EMI -considering it as a more widely familiar term -, but we want to make it explicit our position in favor of the employment of EMI as one tool in academic environments of multilingual nature. Since the start of the Bologna Process (The European Higher Education Area, 1999) the aim of internationalization has been to facilitate mobility by mainly standardizing documentation and easing academic credits transfer (Baumvol, 2016). In this scenario, Englishalready considered the 'Higher Education language' (Coleman 2006 quoted in Baumvol and Sarmento, 2016: 73) -was chosen as a medium of instruction, i.e., a language of instruction, due to its growing presence in academic environments. EMI differs from other approaches such as Content and Language Integrated Learning (CLIL) because it does not have the explicit purpose of teaching language in class. EMI is focused on disciplinary content taught through English. According to Morgado and Coelho (2013: 11), while in EMI we may find content teachers devising strategies (simplifying, classifying, translating, etc.) to help students understand content, and in ESP we find language teachers helping students to learn content-specific language, in CLIL language and content are integrated.
Based on Morgado and Coelho (2013), we understand that a successful adoption of the practice involves the articulation of both preparatory and supplementary courses on (i) EMI policies, practices and epistemology, and (ii) additional language education. Whilst the former focuses on the specific subjects and disciplinary conventions, the latter is interested in the language education of students and professors, specifically in their processes and products (genres) that develop Academic Literacies in English. This view suggests that EMI can benefit from support of the local English language specialist community, such as ESP, EAP (English for Academic purposes) and Academic Literacies programs and projects 7 .
Moreover, participating in classes in which English is used in authentic communicative situations to learn disciplinary content may also improve students' linguistic proficiency (Muñoz, 2012 quoted in Baumvol and. Besides this, EMI is a tendency built by multiple factors, some of which Martinez (2016) lists as factors leading most universities to decide to implement EMI: • attract students from other countries; • prepare students for mobility and a globalized labor market; • raise the profile and ranking position of the university. (Martinez 2016: 192  Further to this, Martinez states that EMI may enhance the participants' future opportunities, as it might bring an 'upward for students who, for example, might obtain better employment through being educated in English, upward for the university for which EMI can increase its international draw and place in international rankings' (Martinez, 2016: 193).
Although there seems to be a number of benefits from EMI adoption, it is yet an issue to be discussed with the local community.
Dearden's report on EMI worldwide (2015) analyzes the example of the Politecnico di Milano, in Italy, in which the adoption of EMI caused reaction from the teaching staff. The university decided that most of its courses 'would be taught and assessed entirely in English rather than Italian' (Dearden, 2015: 18). Otherwise, they 'risked isolation and would be unable to compete as an international institution' (Dearden, 2015: 18). Given this scenario, professors claimed this decision would diminish the quality of teaching due to translation to English. As a result, a 'compromise seems to have been reached and the university website now shows undergraduate and graduate courses taught in English and in Italian' (Dearden, 2015: 18).
The Italian case casts some light on an aspect to be careful with in order to implement EMI: the fashion. Martinez (2016: 192) indicates that EMI practices can be initiated by either a 'mostly top-down (e.g., from the university administration) or bottom-up (e.g., from the faculty and/or students) fashion'. The Politecnico di Milano case seems to fall on the first format. Further to fashion, Martinez (2016based on Cummins, 2000 indicates that this case conjugates an aspect of 'subtractive' EMI police -or EMI-only policy -, in which the first language loses nearly all its space in the curriculum. Especially if the EMI practice is of a topdown origin and combines this subtractive aspect, it can suffer from a greater resistance to be accepted.

EMI in the Brazilian Context
In Brazil, EMI practices are usually isolated due to its origin on a few professors' interest to bring the additional language closer to the teaching reality (Baumvol and Sarmento, 2016). In addition, Dearden (2015: 23) indicates that the public opinion about EMI refers to the practice as 'something for the elite' and not among the most useful investments in education. Martinez (2016) also shows that although there has been a growth in the number of universities adopting EMI since 2010, the number of courses taught in this modality is discreet. To put in the author's words: 'if EMI is a swimming pool into which many countries have dived head first, most universities in Brazil are still just barely 'dipping their toes' in it' (Martinez 2016: 208). (2019), mentioning a Folha de São Paulo report, showed university courses taught in English act as a strategy to call attention of foreign students and in most cases comprise short-term courses and undergraduate (regular or elective) courses. Martinez (2016) shows that universities seemed to require no specific level from students, being the expression 'no proficiency restrictions' found in different Brazilian universities (Martinez 2016: 205). The author cites the annual publication of English First English Proficiency

Guimarães, Finardi and Casotti
Index, which set Brazil as a low proficiency country in 2015. This may be one main reason why institutions in general demonstrated this tendency of not demanding high proficiency levels, once the chances of little enrollment would probably arise. Another aspect that this scenario raises is the necessity of language courses, pointed by Morgado and Coelho (2013) as a fundamental aspect of EMI, which will probably have a significant share on the success or failure of the whole project.
It is important to highlight that several studies cited by Macaro et al. (2018) 8 showed students' preoccupation with their proficiency to cope with classes in English. Some of them reported believing to be learning less than if they had opted for instruction in their mother language.

Morell et al. (2014) reflect on professors' and students' perspectives and needs regard-
ing EMI at a Spanish university. In relation to students, the authors stated that the group feared that their proficiency level would be too low to register in an EMI course and that should be part of the university's programs and that the course they took was a contribution to the improvement of their international and intercultural skills (Corrales, Paba Rey and Santiago-Escamilla, 2016).

Method
The present study was designed to be a cartographic study aiming at unveiling students' perceptions, interests and needs regarding institutionalization of EMI as a potential pedagogical approach at UFSM's teaching culture. Considering the necessity of collecting vast amounts of data from a large population, the adopted research instrument was the crosssectional, web-based questionnaire survey. Following Dörnyei (2003: 09), we believe that questionnaires are advantageous because they are quite efficient in such aspects as '(a) researcher time, (b) researcher effort, and (c) financial resources'. Further, it can be said the 'online' characteristic may be helpful in mitigating eventual embarrassment of researcher and/or participants and may offer participants autonomy to manage their time to answer the questions. post-graduation were excluded, due to the possibility for them to take classes in which English is both the content and the medium of instruction.

The Research Instrument
The self-administered online questionnaire was designed in Portuguese, expecting to enhance respondents' understanding, and consisted of 33 items. These items included four types of questions: multiple-choice, checkbox, semantic differential scales and short-answer.
It was created on Google Forms platform once it is more user-friendly and appealing to students, who widely use it for academic purposes. By the end of the questionnaire, respondents The questionnaire was developed considering some disadvantages of this data collection instrument indicated by Dörney (2003). Unmotivation to answer was approached by pointing to the relevance of the study for internationalization and to the institutional support of the International Office (SAI -Secretaria de Apoio Internacional); pedagogical support was also indicated as a future implication of the study. Lack of specialized knowledge in the field of academic literacy and EMI was predicted and the questions were formulated to be straightforward and to contain ordinary terminology or to include definitions of terms; instructions also clarified when more than one option could be chosen in a question.

Data Collecting and Analysis Procedures
The questionnaire hyperlink was distributed thought institutional Data Processing Center -which manages a database with all e-mail addresses of UFSM's students -following approval of the Ethics Committee. The sample was determined by self-selection, i.e., volunteering, by the participants. Reminder notices were published during the survey period 9 in an

Results
From 26,042 potential respondents, 689 (2.65%) accessed the form and 679 (2.61%) participated in the study by answering it in full 10 . We took the institution's PDI as a parameter to validate the representativeness of our sample. Under this parameter, our survey reached a larger percentage of participation than PDI (Table 1).  (2016) Data revealed 86.3% of students are interested on EMI while only 1.5% reported no interest in the practice. We also considered participants who might not be interested at the moment, regardless the reason. This third group represented 12.2% of participants. Given this scenario, there were more than 500 students who reportedly stated they would engage in EMI courses. In the next subsections, we present these groups perceptions and needs.

Students' perception: limiting factors and relevance of EMI
When asked about the reasons why EMI is not widely adopted at UFSM, students mainly pointed to unfamiliarity with the concept of EMI (75.1%) and to their own low proficiency in the language (58.5%) ( Table 2). Professors' proficiency does not appear in the first group of factors, it is the eighth determining factor according to students. Focusing on the five most mentioned reasons limiting EMI, only one is about language proficiency. To some Facebook pages such as 'Spotted UFSM' https://www.facebook.com/pages/category/Community/Spotted-UFSM-930267327004173/. 10 Seven answers we identified as duplicates, i.e., containing the same answers of another response and; therefore, were disregarded. More information on this problem is available on the platform developer's forum: <https://support.google.com/docs/thread/40048414?hl=en>. Other three did not agree with informed consent and did not actually respond. Accessed 05 July 2020.

Fólio -Revista de Letras
Vitória da Conquista v. 14, n. 2 jul./dez. 2022 extent, students signal one major aspect to be discussed with them is the practice, along with the educational system and how they can relate. Moreover, around 21% of participants believe EMI is not used at UFSM because it could exclude students from low socioeconomic backgrounds. Not the lack of interest (near 19%), infrastructure (around 10%) or relevance of the language in their area of study were pointed as fundamental aspects limiting EMI at UFSM. In specific about a possible exclusion of low-socioeconomic background students, indepth study is needed to provided any further interpretation on the matter. What we can so far report is that the questionnaire allowed us to compare the interest rate between students who held or not Benefício Socioeconômico, a subsidy provided by UFSM to students from low socioeconomic background as a means to assist on their maintenance at UFSM during the period of the major or graduate course which include access to three free-of-charge meals a day and possibility to live at on-campus, university-maintained, student housing. Although composing a minor group in relation to fivefold non-BSE students in the sample, we perceived a slightly higher interest rate on the former group (87.9%) than in the latter (86.1%).
All the Not interested respondents are from the non-BSE group.
About students' opinion with reference to an official policy for the implementation of EMI, only 6.8% think it should not be institutionalized while the majority considers it an what data showed about limiting factors, from the three most selected options, two are focused on the characteristics of an EMI project (educators' instruction and flexible norms) rather than the offer per se. Some participants added other viewpoints (Table 4) in relation to the institutionalization of EMI. In most cases, participants feared a top-down fashion EMI, especially considering an institutionalization/regulation as a means to impose the practice. Participants #51, #60, #341, and #394 mention language courses offer as a step towards institutionalization. #66 and #504 also highlight the relevance of a progressive introduction, possibly via elective courses instead of mandatory ones. Participant #114 is an example of discourse against topdown policies as no actual contrariety on interested students to engage on EMI is reported, but on institutionalization of the practice. The word institutionalization seems to have been taken as a proposal to fully replace Portuguese as a Medium of Instruction rather than a means to ensure the legal status of EMI and the integrity of those involved. Acho importante a oferta para alunos que se interessam, mas sou contra a institucionalização
*sic, emphasis added Source: the authors.
In relation to the specific role of EMI in internationalization, participants frequently reported that EMI may contribute to the preparation for mobility (84.5%) or for the global professional market (70.1%) ( Table 5). Only 1.6% stated they could not identify any positive result stemming from EMI adoption. Students also highlighted EMI's potential to create a welcoming environment to international staff and students and establish a plurilingual setting to study (English during instruction, Portuguese in ordinary interactions and other languages students and staff might include). However, these roles related to pluricultural setting and inclusion are yet secondary to a traditional view of internationalization, which is conceived as a synonym to mobility.
This is an important aspect to be discussed in EMI courses or forums. In the next section, we focus on students' language and institutional needs.

Students' needs: towards EMI policy
A significant majority of the participants (67.5%) indicates the need of broadening linguistic education in English at UFSM (Table 6). The necessity of specialists in English to monitor the EMI students follows it with 48.3% of mentions. The data collected showed that a substantial part of the students (nearly 44%) would be more willing to enroll in EMI courses if any official advantage results from it. In addition, they consider necessary to set a policy to administer the EMI offer. It is important to note that students are more often concerned with the need of language assistance to them than to their instructors, what is consonance with their assessment of their own low proficiency rather than their professors' as a limiting factor to EMI implementation.
In our mapping of EMI, we also considered important understanding the current role of the English language in the activities conducted at UFSM's classes. In general, 82.1% of the students declared they use English in their classes. Data showed nearly 73% read bibliography in English indicated by the professors and 48.7% understand technical terminology from their field in English (Table 7).

Table 7 -Presence of English in class-related activities
Activity % Outside classroom, I study/read bibliography in English, by indication of the professor 72.8 I understand disciplinary terminology in English used by the professor during explanations 48.7 During classes, I study/read bibliography in English, by indication of the professor 33.9 I read slides used by the professor during classes 20.6 I write assignments in English to hand in to the professor 11.5 I cope with classes orally delivered in English 6.0 Take course tests/exams in English 6.0 I interact orally in English with classmates from other countries during in class group activities 4.6 I present seminars/assignments orally in English with slides written in English 4.3 A surprising number of students (6%) has taken course examinations and other 6.7% have delivered oral presentations in English (with the slides written in Portuguese or English). In addition, 6% stated they had classes orally delivered in English. These results signal to a relevant presence of English in the academic culture and in the local academic literacy practices at UFSM. We consider it remarkable because nor EMI or any other similar approach is currently officially institutionalized. If we concentrate on the five most mentioned class-related activities, four are connected to written skills (reading and writing) and, pedagogically, only three are during the session/class.

Previous experiences with EMI and further uses of English in class
Despite 6% answered they had been to classes delivered in English, only sixteen students (2.3%) reported they had already taken EMI courses. Six of these students (37.5%) classified their proficiency at the Beginner or Basic level, whereas five (31.2%) at the Advanced or Proficient level. The remaining participants presented discrepant self-assessments of their communicative skills, usually giving lower assessments to their oral skills and higher to written skills.
About the reasons to enroll in an EMI course, the students mainly pointed to the absence of another option (50%), followed by the desire of improving their proficiency in English (37.5%) ( Table 8). The role of the additional language to the disciplinary field was also frequently mentioned. However, it is important to highlight the first reason to have enrolled on EMI course was not a fully personal decision, and only one student did it for the experience (6.2%). Considering the setting of parameters for an EMI policy these traces of top-down approach (only option, mandatory course) should not be encouraged. In answering what difficulties they have faced during the EMI classes, 37.5% reported their low proficiency as the major obstacle and their shyness to interact orally was pointed as equally impactful on the experience. However, 31.2% declared having no problems to cope with the EMI course (Table 9), which is consistent with the number of self-assessed Advanced/Proficient EMI students. Maintaining a pattern demonstrated by the whole group, students who took EMI courses at UFSM demonstrate a better evaluation of their professors' proficiency than their own. Another relevant finding is that from the sixteen EMI students, fourteen (87.5%) reported they would continue taking EMI courses. The other two students declared not interested at the moment. None of the students who had the EMI experience answered 'not interested' in the practice.

Conclusion
In this paper, we aimed at developing a short cartography of students' interest, perspectives and needs in relation to EMI at UFSM. Through our survey, we noticed there is a stark interest on English-mediated courses. Based on students' perspectives on limiting factors and the relevance of EMI on internationalization interconnected with their language and institutional needs, it is clear the EMI project at UFSM must not limit itself on offering curricular courses on the additional language.
The unfamiliarity with EMI and the lack of discussion on the practice students reported as aspects that lead to the limited employment of EMI at the institution ( well as the crystalized association of internationalization as mobility they still seem to hold (Table 5) signal to a necessity of including a wide discussion about the practice at an initial phase of its institutionalization, prior to the offer per se. This discussion could be established though small talks on selected aspects (e.g., whether should be a minimum proficiency level, if the discipline would be simultaneously offered in Portuguese, assessment) and continuously instigating students to actively participate in these talks. The participants also considered their professors should receive a specific professional development on the practice and that an EMI policy should respect the particularities of each area and study level.
Regarding linguistic education, the community showed concern with the availability of language education courses. Most of them (64%) have never taken English language courses at UFSM and around 20% had not even known about the existence of projects UFSM have on this area before answering the questionnaire. The language support; however, is not circumscribed to a preparing phase, but it is also connected to the monitoring of EMI students (Table 6). Further, more than monitoring to assist the comprehension of the content, the participants who took EMI courses demonstrated greater need of language support to EMI courses. Especially in this aspect, the institution could help on the enhancement of the promotion of the language projects and their courses offer.
The majority of students who took EMI courses pointed the reason why they enrolled in these courses was that it was the only available option (Table 8). Although this group is modest, this might indicate top-down fashion (Martinez, 2016). Moreover, several students seemed concerned with the idea of institutionalization of this modality via institutional regulation, understanding it would be an imposed practice, rather than a regularized one. On the other hand, this unclearness signaled the adoption of top-down fashion would not be well accepted by the community.
Because of this, it is necessary to ratify that under no circumstance we advocate for an EMI-only policy. Conversely, we believe EMI should be configured as a means to empower Brazilian science and academy by not 'ignoring the imperialist [culture]'s language, but rather knowing how to make use of it in favor of the Third World [countries] 11 ' (Moita Lopes 1996: 59). Therefore, the preparing and monitoring projects within the umbrella-project of EMI at UFSM should also address that we can consider English as a Tyrannosaurus rex (Swales, 1997) due to its carnivorous trait in relation to other languages, especially in academic publications.
This EMI configuration would then be one that allows a widening in Brazilian academia