University students’ attitudes towards English Medium Instruction and academic language support: A view from Indonesia

Despite the rapid growth of English Medium Instruction (EMI) in Indonesian universities, studies have revealed students’ challenges in meeting the required English competence while tackling academic content. This study used both quantitative and qualitative approaches to investigate the attitudes of 136 students toward EMI through an online questionnaire in five Indonesian universities. It also examined the academic language support available through interviews with six students and three teaching staff of the language centers administered by the universities. The findings of this study revealed that the students mainly perceived EMI as a tool to access better employment and improve their English proficiency. They also had various expectations towards EMI teachers, which might be influenced by their views of EMI as a means of learning English. Despite the imposition of English-only instruction, the students had a positive attitude towards the use of language(s) other than English in EMI courses. This study also reported several models of language support offered by the universities, which mainly focused on English for Academic Purposes (EAP) and were deemed insufficient to address the students’ disciplinary needs. These findings suggest that the pedagogical practices of EMI should be critically adapted to cater to the needs of the local context. This study calls for the re-conceptualization of EMI and the systematic language support of English for Specific Academic Purposes (ESAP) embedded in the curriculum.

of EMI programs, which reached up to 110 universities in 2021 (Lamb et al., 2021), primarily due to an autonomous movement of the universities (Lamb et al., 2021;Santoso & Kinasih, 2022;Simbolon, 2021).While the Indonesian government has encouraged EMI provision at the university level (Simbolon, 2021), there is no official regulation that guides its implementation.Such absence has resulted in the misconception that implementing EMI means using 'native' English in content classrooms (Santoso & Kinasih, 2022) in order to improve English proficiency and master academic content (Galloway & Ruegg, 2020).
Generally, EMI refers to the utilization of the English language to teach content subjects in non-English-speaking countries (Macaro, 2018).This definition, however, may be problematic in Asian contexts where improving English language proficiency has become part of the country's agenda (Galloway & Ruegg, 2020).Scholars have also sought to extend the definition to include Anglophone countries that are increasingly becoming multilingual (Galloway & Rose, 2021).While content and English language instruction labeled as Content and Language Integrated Learning (CLIL) and Content-based Instruction (CBI) adopt a dual focus on both academic instruction and language learning, EMI does not explicitly aim to enhance students' English competence (Dearden & Macaro, 2016;Rose & Galloway, 2019).Thus, how English Language Teaching (ELT) is situated within EMI is unclear, further raising a crucial question: Is ELT part of EMI?Nevertheless, many universities regard improved English language proficiency as a result of studying content in English (McKinley & Rose, 2022).This perspective implies that English proficiency will simultaneously develop along with the construction of knowledge in subject disciplines (Rose & Galloway, 2019).Pecorari and Malmström (2018, p. 497) have called for the importance of understanding "a very natural symbiosis" that both ELT and EMI offer.As Lasagabaster (2018) notes, the implementation of EMI should be rooted in the principles of English for Specific Purposes (ESP) and English for Academic Purposes (EAP) courses, which are remarkably absent in many contexts.In other words, ELT should be put at the heart of EMI, which raises critical arguments regarding how universities should situate and integrate ELT as a core part of their EMI policy (McKinley & Rose, 2022).To link ELT with EMI provision, many universities in Asia, such as Japan (Galloway & Ruegg, 2020;McKinley, 2018), Korea (Chang et al., 2017), and China (Hu et al., 2014) have provided various types of language support for students.Nevertheless, more research is still needed to investigate feasible support models appropriate for specific teaching contexts.
Most studies conducted in Indonesia have examined university teachers' and/or institutional stakeholders' views of EMI (Dewi, 2017;Floris, 2014;Hamied & Lengkanawati, 2018;Santoso & Kinasih, 2022;Simbolon, 2018Simbolon, , 2021)), but there is a lack of research on investigating this issue from students' points of views.In this study, we address this with an investigation of student attitudes towards EMI, including their motives for enrolling in EMI, as well as their attitudes towards EMI teachers and language use in EMI.This study also aimed to examine language support available for EMI students in Indonesia, a research focus that still remains unexplored.Understanding students' attitudes towards EMI and available language support is deemed essential to inform the EMI policy planning and implementation.

English Medium Instruction in Indonesia
The significance of English as a global language in politics, economy, and education has contributed to the proliferation of educational institutions offering EMI.In 2007, the Indonesian government established international standard schools for primary and secondary education throughout the country (Departemen Pendidikan Nasional, 2007).Nevertheless, this policy was officially canceled in 2013 as it attracted many criticisms, such as creating unequal access to education for all Indonesian students, exclusively targeting students from highincome families, and not having enough qualified EMI teachers (Cahyani et al., 2016;Dewi, 2017).Despite the cancellation of the policy, many schools, especially private institutions, still use English as a Medium of Instruction (MoI) without being labeled as international standard schools.They usually adopt an international curriculum that offers courses through EMI.In this case, private schools sometimes hire foreign teachers, who might be appealing to parents who can afford the expensive fees charged by the schools.
In 2015, the Minister of Research, Technology, and Higher Education supported Indonesian universities in aiming for internationalization through EMI provision (Simbolon, 2018).The driving forces of EMI in Indonesia have been identified in prior studies, summarized as (1) preparing students to compete globally, (2) improving students' English competence, (3) achieving higher university rankings, (4) increasing the intake of international students, (5) improving graduates' employability, (6) meeting the institutional demand (Hamied & Lengkanawati, 2018;Lamb et al., 2021;Santoso & Kinasih, 2022).These motives have been constantly reported by much research in different contexts (Galloway et al., 2017;Kim, 2020;Macaro et al., 2018).In reality, however, the implementation of EMI has often been based on pragmatic reasons, such as gaining more profits from wealthy families (Coleman et al., 2023).
It is intriguing to note that EMI has been operationalized in various forms.Both state and private universities use the terms 'double degree,' 'international classes,' or 'bilingual programs' to describe the enactment of EMI in this context (Santoso & Kinasih, 2022).The various understandings of EMI among stakeholders are evident since no policy has been stipulated to specifically regulate EMI provision in Indonesia (Simbolon, 2018(Simbolon, , 2021)).In the formal regulation, EMI is supported by Law No. 12/2012 concerning Higher Education, stating that a foreign language can be used as an MoI without explicitly mentioning the utilization of the English language (Pemerintah Indonesia, 2012).Consequently, many universities tend to set official guidelines for EMI provision on their own.This bottom-up approach in policymaking should be able to meet institutional needs according to the specific context.However, the absence of explicit national policies and implementation strategies, along with monitoring and evaluation of the program, has resulted in some repercussions.Teachers, for example, often grapple with the lecturing mode, the role of ELT in content learning, and pedagogical strategies in teaching content through English (Hamid & Rifai, 2023;Yuan et al., 2020).Moreover, many universities rely on setting an entry requirement demanding a certain level of English competence.However, as Galloway and Ruegg (2020) argue, such a standardized English test could not guarantee EMI students' performance in the long run.In fact, even if students can fulfill this prerequisite language proficiency, they often face difficulties in tackling the academic content (Galloway et al., 2017).Another issue is related to how much English should be used in EMI, whether English should be exclusively used, or if there is room for other language(s) to occur.Some studies have reported that despite the enactment of an English-only policy in many EMI courses, the stakeholders acknowledged the role of student's own language and used it to facilitate the teaching and learning process (Santoso & Kinasih, 2022;Simbolon, 2018Simbolon, , 2021)).

Student Support in EMI
The growth in EMI provision has unfortunately not been supported by empirical research focusing on English language-related difficulties in EMI.Although English language competence has been reported to significantly influence students' academic achievement in EMI courses (Rose et al., 2020), studies have shown that meeting a threshold level of English proficiency could not ensure the success of students' content learning through English (Galloway & Ruegg, 2020).Part of the reason is that such a threshold does not accurately reflect the required English skills for EMI (Hu & Lei, 2014;Wilkinson, 2013).Studies have revealed various linguistic challenges in EMI programs, including (1) using specialized vocabulary (Chan, 2015); (2) understanding lecturers' accents (Hellekjaer, 2010); (3) interacting with peers and delivering oral presentations in English (Kırkgöz, 2009;Pun & Macaro, 2019); (4) writing academic discourses (Abouzeid, 2021); and (5) reading academic texts (Uchihara & Harada, 2018).In non-Anglophone countries, these language-related difficulties have been found to affect students' performance in acquiring content knowledge, communicating content, willingness to ask questions, and requiring a longer time to complete EMI courses (Galloway et al., 2017).
Generally, many universities in Asia offer support services to tertiary students.In Japan, there has been an increasing number of writing support services (Johnston et al., 2008).Nevertheless, many Japanese students are unenthusiastic about taking up such support, although they often encounter linguistic difficulties in EMI classes (Ishikura, 2015).In Korea, many universities have also provided general English language courses, but these are considered insufficient in preparing students to enter EMI programs that require subjectdiscipline language skills (Chang et al., 2017).In China, students are reported to have insufficient command of English to engage in cognitively demanding content knowledge despite various language support mechanisms offered by the universities (Hu et al., 2014).These examples indicate two important issues: (1) there is still a lack of systematic support for EMI students, although its provision is available, and (2) the support provided does not effectively address students' English language challenges for EMI courses.
With regard to student support in EMI, Macaro (2018) has classified three models of support in EMI: (1) the preparatory year model, in which students take an EAP course prior to their academic studies; (2) the institutional support model, which offers in-sessional EAP or ESP courses integrated into EMI degrees; and (3) the pre-institutional selection model, which relies on English language entry requirements to select students.In some European contexts, one-to-one support has gained popularity among undergraduate students as these tutoring systems can cater to the specific needs of students (see Kelo, 2006).In addition, writing support is also available in many universities to enhance students' academic writing skills, such as making citations, paraphrasing, summarizing, synthesizing, and complying with academic ethics (Chang et al., 2017;McKinley, 2010).Such support, however, should be tailored to meet the needs of a specific context (McKinley, 2010), which includes facilitating both domestic and international students with diverse linguistic backgrounds to have equal access to such support.

Research on EMI in Indonesian Higher Education
There has been an increasing number of studies investigating the implementation of EMI in Indonesian universities.Most of the studies focused on investigating teachers' and/or institutional stakeholders' views of EMI (Coleman et al., 2023;Dewi, 2017;Floris, 2014;Hamied & Lengkanawati, 2018;Santoso & Kinasih, 2022;Simbolon, 2018Simbolon, , 2021)).These studies have shared similar findings showing the significant role of English, i.e., in increasing competitiveness, enhancing academic reputation, and providing greater access to employment, as the main driving forces of EMI in Indonesia.Many teachers in some of the studies, however, have been reported to face some difficulties in teaching content in English due to the lack of English proficiency and sufficient professional teacher training programs.
Scholars have also looked into the spread of EMI in Indonesia with a critical view.Floris (2014) argued that exposure to the general English language, rather than the language for specific purposes, during school years might pose linguistic difficulties for Indonesian students participating in EMI programs.As learning content knowledge involves the use of discipline-specific terminologies, students often find it difficult to master their field of study due to their low English proficiency.Dewi (2017) also highlighted the issue of Western imperialism associated with EMI in the EFL setting and emphasized that the government should provide equal access for all Indonesian students to empower themselves through EMI.Drawing on Indonesia's multilingual setting, Santoso and Kinasih (2022) argued that acknowledging English as an international language and, at the same time, considering the existence of multiple languages in Indonesia is required to balance the language policymaking in EMI.
Despite the different research contexts covering public and private universities in Indonesia, the above-mentioned studies underline a similar issue: students' low English proficiency, which could significantly affect their academic achievement.Therefore, it is crucial to explore what kind of support universities have offered to assist EMI students with necessary academic English skills.In addition, investigating the implementation of EMI from students' perspectives is deemed important to understand their expectations and experiences in studying through English.

METHOD
This study used both quantitative and qualitative approaches to collect the data.The former used a questionnaire survey to gather data from students based on the pre-determined themes, including (1) students' motives for enrolling in EMI, (2) students' views towards teachers and language use in EMI, and (3) academic language support; whereas the latter used interviews to obtain data specifically regarding student support from the perspectives of students and teaching staff.Although this study was framed within "a fixed design" through the use of questionnaires, the latter phase allowed "a flexible design" to take place (Robson & McCartan, 2016, p. 46), which might facilitate further elaboration and clarification from the participants' responses.This design shift from the philosophical assumptions of postpositivist to constructivist is thus expected to cover a wide range of participants and produce indepth descriptions of multiple views towards the phenomena under scrutiny.In addition, convenience sampling was used to select the participants for the data collection process.Therefore, the choice of the participants was largely based on their willingness and availability to participate in this study.With this sampling method, it is imperative to acknowledge that the representativeness of the population being studied might not be fully achieved, thus affecting the potential biases and the generalizability of the findings.To address these limitations, this study recruited participants from different study programs in both state and private Indonesian universities located in various geographical areas across Java Island to better represent the target population.The convenience sampling method is beneficial in this study for identifying trends and collecting initial data, particularly related to student support.Given the lack of research focus in this area, the findings of this study could potentially inform larger-scale studies involving a more representative sample.
In this study, the questionnaire items, consisting of 17 close-ended items using the Likert scale and four open-ended items, were adapted from Galloway and Ruegg (2020) and designed to elicit relevant information related to students' attitudes towards EMI and academic language support.The questionnaire was translated into Indonesian and initially distributed via Google Form to EMI students from six Indonesian universities in several provinces on the island of Java.Moreover, the pilot study was also conducted to test and refine the questionnaire to ensure the clarity and relevance of its items.Based on the results of the pilot study, several open-ended questions were added to ask about how student support had been implemented in the respective universities.
The interview guideline was developed to examine the types of language support available for students.Focus group discussion was conducted with six students (S1, S2, S3, S4, S5, S6) from three universities (University A, University B, University C), which provide other models of language support in addition to requiring students to meet the minimum English language requirement.The students were selected since they had experience utilizing academic English language support; thus, they could provide useful information about their attitudes toward university services.Also, three university teachers (T1, T2, T3) who teach language support were also interviewed to obtain detailed information about the types of student support available at each university.
The data analysis for quantitative and qualitative data was conducted separately.To analyze quantitative results, this research used descriptive statistics to obtain "a summary picture of a sample" regarding key themes being researched (Gray, 2014, p. 626).This study focused on describing frequency distribution in percentage and presented the data using tables and diagrams.The percentage was then interpreted using descriptive analysis, which refers to the following categorization using the framework proposed by Riduwan (2019) (see Table 2).Additionally, this research used thematic analysis to analyze the qualitative data.It adopted the stages of thematic coding analysis consisting of (1) transcribing, reading, and re-reading the data; (2) constructing initial codes; (3) categorizing codes into potential themes; (4) describing and interpreting patterns reflected in the data (Robson & McCartan, 2016).This study constructed the themes deductively with reference to the relevant literature, which includes the main theme in the qualitative data, i.e., the student support models, following Macaro's (2018) models of support in EMI: (1) the preparatory year model, ( 2) the institutional support model, and (3) the pre-institutional selection model.The qualitative data was then coded and categorized into these pre-set themes.

FINDINGS AND DISCUSSION
This section presents both questionnaire and interview data, which was categorized based on the main themes: (1) students' motives for enrolling in EMI, (2) students' attitudes towards EMI teachers and exposure to English, and (3) student language support.

Students' motives for enrolling in EMI
This section presents the results of questionnaire items requiring the participants to state their stance toward possible reasons for taking EMI (Table 3).
The survey results demonstrated that increasing employment opportunities and improving English proficiency were the students' main reasons for taking EMI, 100% (62.5% strongly agreed and 37.5 agreed) and 99.3% (57.4% strongly agreed and 41.9% agreed), respectively.The other statements also gained significant agreement from most of the participants, i.e., interest in the content of EMI courses (96.3%), quality of universities offering EMI (95.6%), opportunities to study abroad (94.1%), interest in learning English (91.2%).The aforementioned findings correspond to the empirical research conducted by Galloway et al. (2017) in Japan and China, showing that improving English has become one of the major factors affecting students' decisions to take EMI programs.While the reasons related to employment opportunities, opportunities to study abroad, and interest in learning English were found to be insignificant in the prior study, these factors were significant in this current study.Some of the participants also mentioned other reasons for taking EMI in the open-ended question, e.g., studying through EMI could provide access to better salaries at work (n= 6), better employment (n= 6), global competitions (n= 3), and academic knowledge (n= 1).Similar findings regarding the significance of English in these domains were also identified in numerous research investigating teachers' and students' perspectives and governmental-level EMIrelated policies (Lamb et al., 2021;Dearden & Macaro, 2016;Hu et al., 2014).
The inevitable demand for EMI cannot be separated from the two buzzwords: globalization and internationalization.In this sense, globalization has triggered the spread of English as the global language, with the aim of fostering international cooperation, trade, and economy across nations (Santoso & Kinasih, 2022;Tupas, 2018).EMI is thus seen as a major tool to internationalize universities, mainly aiming to attract foreign students and strengthen global competitiveness (Macaro et al., 2018).In addition, the importance of internationalization is reflected in the participants' major motives for taking EMI programs.The implementation of EMI as part of internationalization has become a strategic agenda in many Indonesian universities as it is seen as a way to provide graduates with better access to employment and education (Santoso & Kinasih, 2022;Simbolon, 2021).However, as Lin and Lo (2018) argue, EMI provision should be critically evaluated.If not thought through, the adoption of EMI would potentially result in an injudicious implementation that lacks educational principles informed by empirical research in applied linguistics.Asian countries, like Indonesia, may have their own unique contexts along with their rich local diversity that may be slightly or considerably different from other countries where EMI has been successfully implemented.A critical view towards EMI is thus needed to ensure that institutional conditions are in place to avoid universalizing EMI.
The findings of this study also indicate that the switch in MoI from Bahasa Indonesia to English seems to have generated an over-optimistic view of EMI.Although EMI does not explicitly aim for English language improvement, it is often assumed that studying through English will result in better English language proficiency (Chin & Li, 2021).Some studies focusing on teachers' and students' attitudes have revealed this similar trend (see Lamb et al., 2021;Lei & Hu, 2014;Li & Wu, 2017;Santoso & Kinasih, 2022).Nevertheless, prior research investigating the effectiveness of EMI in improving English language proficiency has revealed disparate findings (Huang, 2015;Li, 2017).This can be caused by several factors that might affect students' English language proficiency, such as the length of the English learning experience and the diverse methods of implementing EMI in different contexts.

Students' attitudes toward EMI teachers and exposure to English
This section discusses the results of the questionnaire regarding students' attitudes towards EMI teachers and exposure to English.As shown in Table 4, all of the participants strongly agreed or agreed that EMI teachers should have the pedagogical skills to teach EMI courses.Similar to Galloway et al.'s findings (2017), the students' perceptions of EMI as a means of learning English, in contrast to learning through English, seemed to influence their expectations towards EMI teachers in this study.For example, 98.5% of the participants believed that teachers should have good English proficiency.80.5% of them also expected that EMI teachers should speak English with native speaker accents, with only 16.2% choosing disagree or strongly disagree.Interestingly, over 90% of the participants strongly agreed or agreed that teachers should obtain EMI certification, which still becomes an absent attribute in the current EMI provision (see Macaro & Han, 2020).Regarding students' attitudes towards English in EMI (Table 5), there seems to be an inconsistent finding resulting from two questionnaire items.While 80.5% of the participants thought that EMI should be delivered in English exclusively, none of them expected the exclusion of other language(s) in EMI.This result could be explained by a number of responses throughout the obtained data.Referring to the questionnaire items requiring the students to select the frequency of using English on a 5-point scale (see Figure 3), 61.1% of the participants sometimes or rarely used English to interact with the teachers, whereas 27.9 % of them chose often and only 8.8% chose always.When talking to their peers, almost 60% of the participants rarely or never used English, and only 14.8% of them chose always or often.It is also apparent that the choice of students' language use was quite distinct from how stakeholders, such as the teachers and the institutions, actually practiced language use.For example, the students were always or often received examinations (97%), materials (82.3%), and lectures (72%) in English.
The results of this study may indicate what sort of 'English' the participants are in favor of based on their expectations towards EMI teachers.As can be seen in Table 4, many of them preferred EMI teachers who have good English proficiency and English native speaker accents, as in British and/or American English.Their views could be affected by the long-held assumption that the 'correct' way to use English is the way it is used by its native speakers (Jenkins, 2019;Rose et al., 2022).It is thus important that 'E' in EMI should be reconceptualized in relation to the kind of English used in EMI in different settings.Such reconceptualization could potentially affect how English could be appropriately realized in EMI programs without a strong adherence to English native-speaker norms.Nevertheless, this issue might not become a popular consideration in the implementation of EMI in Indonesia since universities tend to focus on the strategies of recruiting both local and international students through the use of English as MoI (Santoso & Kinasih, 2022).Considering the Indonesian multilingual context, current EMI research has put forward the shift from Eurocentric views of EMI to taking into account multilingual language practices, particularly in the wider multilingual ecology of Asia Pacific (Heugh et al., 2017;Kirkpatrick, 2014).

Figure 3 Students' Exposure to English in EMI
This shift entails the role of English as a contact language in multilingual communication, where other languages are also present in the interaction (Jenkins, 2015).This conceptualization is arguably compatible with the context of current research where other language(s) of the students may be preferred in any particular circumstances (see Figure 3).The majority of EMI students in many Asian contexts, including Indonesia, are local students who share the same language as the teacher (Sahan, 2020;Wang & Curdt-Christiansen, 2019).This does not mean that the communicative need of international students is overlooked, but the multilingual realities need to be acknowledged in order to take into account the sociolinguistic circumstances of a particular context in language planning and policy (Rose et al., 2022;Dewi & Goebel, 2023).This call to action highlights the importance of avoiding the pragmatic implementation of EMI.If not carefully considered, the spread of EMI could result in the reproduction of a neoliberal agenda (Gu & Lee, 2019), which has consequently commercialized English and Westernized students and undermined the potential of other languages in knowledge production (Sah, 2022).

Student Support
This section presents the research results which specifically address academic language support for EMI students.This study revealed that University D and University E only applied the pre-institutional selection model, which requires official certificates of standardized tests such as TOEFL and IELTS prior to enrollment, with little support available throughout the studies.Moreover, University A, University B, and University C provided other language support in addition to standardized English language tests.In this study, the participants stated that they took TOEFL, IELTS, or English tests administered by the university, 57.40%, 24.30%, and 18.3%, respectively (see Figure 4).Regarding the TOEFL test, the participants mentioned the required scores ranging from 475-550 for the paperbased test and 60-80 for the internet-based test.Meanwhile, the scores varied between 5.5-6 for the IELTS test.It is also reported that University A also had interviews in English to assess students' speaking abilities.

Figure 4 Types of English tests required for EMI students
The finding showed that University B offered several models of language support for the students.S1 stated that students who could not meet the required English score should attend a compulsory English course prior to their studies, although this course was not part of the credits that students were required to take.His response was confirmed by the teaching staff at the university: T1 further stated that the materials were mostly about English grammar and its usage, which were considered beneficial to help students achieve a high TOEFL score.Students were also expected to be able to differentiate sentences and clauses; verbs, nouns, and adjectives; and analyze grammatical mistakes in sentences.In addition, University B also had several compulsory academic English courses embedded in the curriculum: S1: "There are at least 2 required academic English courses that we need to take… For example, we are required to read and write several types of essays such as, cause-effect and expository essays… I think the focus of these courses is to equip us with necessary academic skills such as, summarizing, referencing, and paraphrasing."(University B) S2: "I think if we do not pass these [courses], we cannot proceed to the thesis writing… [These courses] are useful because we can learn about English more comprehensively, for example, analyzing different types of essays, organizing ideas… We also have some speaking tasks such as presentations for professional purposes."(University B) T1 stated that the purpose of these courses was to enhance students' abilities to produce written and spoken academic discourse for general and specific purposes.She also added that these courses emphasized the teaching of academic writing skills such as, paraphrasing, summarizing, and referencing.
The findings of this study also demonstrated that University A and University C provided a similar type of student support integrated into the university language center.This support was voluntary and available to all students taking either regular or international programs.Joining classes in the language center would require students to pay depending on the type of course they would take: S3: "Students usually take TOEFL tests that are held regularly by the language center… I think it [the language center] offers TOEFL preparation program for students as well."(University A) S4: "I think there is no specific language support for the students in the international program.The language center is for everyone, I think.It just offers a general English course focusing on reading, writing, listening, and grammar… We need to seek assistance by ourselves if we need additional language support related to our disciplines."(University A) S5: "The language center usually only offers EPT [English Proficiency Test] and general English to students… No specific support for students taking the international program, such as English for specific purposes that are related to our majors."(University C) S6: "It [the language center] also provides courses related to cultures where we can learn cultures from English speaking countries.But, there is no course that specifically address our needs to understand English for related disciplines."(University C) These students' responses were confirmed by the teaching staff.
T2: "We offer several programs for our students, for example, English conversation, general English, and TOEFL preparation program.Basically, we aim to improve our students' English abilities, including their reading, writing, listening, and speaking skills… All students [enrolled in regular or international programs] and the public can take these [language] program if they are interested in improving their English."(University A) T3: "Some of the programs include EPT, English language course, and cross-cultural understanding programs."(University C) In this study, the students expressed their views towards the available language support.The students from University B were quite satisfied with such support as they could learn how to write academic essays with regard to their academic discipline.Nevertheless, they felt that too much time was allocated to general reading and writing tasks and expected to have more ESP-related tasks.The others also expected to have academic language support that can enhance their English abilities in relation to their disciplines: S4: "Sometimes I have difficulties in reading academic texts because the vocabulary is very complicated and unfamiliar… It is very hard for me to comprehend materials without sufficient English vocabulary related to law." (University A) S6: "My major [business management] requires me to be able to speak English actively for… presenting data, discussing business plans… We need teachers who understand specific terminologies relevant to our major."(University C) When asked about how these students overcame their problems, some of them stated that they looked up the meaning of an unfamiliar word in specific dictionaries designed for their academic disciplines.One participant stated that he preferred reading many business texts to acquire businessrelated terminologies useful for his study.The others did self-study to support their learning through English.
The above findings revealed that the universities provided different types of academic language support for EMI students, as categorized by Macaro (2018).In this study, only University B offered both preparatory and concurrent support models in addition to the pre-institutional selection model.While University D and University E only offered the pre-institutional selection support model, Universities A and C used this model along with the provision of language centers where students can voluntarily seek academic assistance.The establishment of a language center for student support, however, can be problematic.As Galloway and Rose (2021) argue, it is often managed separately from the language department and other academic support, resulting in structural challenges for teachers to provide an appropriate type of support for EMI students.This separation could possibly hinder the collaboration between language practitioners and content specialists in a concerted attempt to incorporate language support into EMI programs (Lin & Lo, 2018).The research results also demonstrate that the language support in this context mainly focuses on EAP, which emphasizes the training of students' general academic skills.Following recent research (e.g., Lamb et al., 2021;Galloway & Rose, 2021;McKinley & Rose, 2022), this study suggests the movement from EAP to English for Specific Academic Purposes (ESAP), which is concerned with the specific disciplinary needs of students (Flowerdew, 2016), to better cater specific academic needs including technical vocabulary and writing genres.
Overall, the results of this study offer several implications for policymakers and educational practitioners.First, this study suggests that a critical evaluation of EMI provision in Indonesia is required to avoid a one-size-fits-all approach to EMI.Exploring best practices of EMI in Indonesia is crucial to ensure that contextual factors such as Indonesia's sociolinguistic landscape, the readiness of university teachers to teach content through English, and the availability of student support are taken into account.In this respect, Richards and Pun's (2021) typology of EMI could serve as guidance for curriculum planners and for content and English language teachers in identifying both similarities and differences of EMI across contexts and designing an EMI model suitable for their own teaching contexts.In this typology, Richards and Pun (2021) have classified the criteria to identify the characteristics of EMI, including purposes of EMI, curriculum models, assessment in EMI, EMI teachers and students, and instructional materials in EMI.The typology provides a framework to profile the features of EMI and to set appropriate parameters to implement EMI by considering sociocultural and classroom settings.Second, this study suggests that the re-conceptualization of the 'E' in EMI entails the need to raise awareness among both EMI teachers and students regarding how English is practiced within multilingual contexts.This requires a deliberate effort to challenge the monoglossic orientation by acknowledging and legitimizing multilingual practices in EMI classes (Santoso & Kinasih, 2022).Third, the fact that the available type of student support focused on EAP highlights the necessity for universities to provide ESAP courses to better meet students' disciplinary needs.Providing support through ESAP, however, could be challenging for ELT practitioners as it requires a sufficient understanding of subject knowledge and specific competencies for the subject discipline (Flowerdew, 2016).It is also important for teacher education programs to provide pedagogical training for prospective teachers to teach in ESAP classes.Lastly, universities can design a curriculum model that facilitates the collaboration between content and language teachers, such as the adjunct/linked course model (see Lin, 2016).This could be a possible alternative to equip students with the language demands of the content subject.In this model, students take an adjunct/linked ESAP course that covers specific language features in the content subject that are needed to do various classroom tasks such as reading academic texts, writing specific academic genres, and participating in the discussions in the linked content course (Lin, op.cit.).

CONCLUSION
The study aimed to investigate students' attitudes towards EMI and language support in Indonesian universities.The findings showed that increasing employment opportunities and improving English proficiency became the main reasons for students taking EMI courses.The students also had numerous expectations towards EMI teachers, which seemed to be affected by their perspective of EMI as a tool for learning English.Although they preferred the exclusive use of English, none agreed with the elimination of other language(s) as MoI in EMI classes since the other language(s) frequently occurred between teacher-student and studentstudent interactions.In addition, this study reported various models of language support available for EMI students in this context.All of the universities involved in this study implemented the preinstitutional selection support model, which requires students to meet a certain English language proficiency before the start of the academic year.Two out of five universities also developed a center for language services where students can get academic assistance voluntarily, and only one university provided the preparatory and concurrent support models in the form of enrichment programs and compulsory English courses embedded in the curriculum.The results of this study not only enriched the current scholarly discussion on the multifaceted nature of EMI students' attitudes but also shed light on the evidence of the different kinds of student support in Indonesia, which has mainly focused on EAP rather than ESAP.
This study has contributed to expanding the existing literature by offering insights into students' attitudes towards EMI in the Indonesian context.First, the exponential growth of EMI in HEIs should not reproduce a neo-colonial mentality that propagates 'colonization of the minds of those working and studying' within the systems of Western countries (Rose et al., 2022, p. 167).Thus, the notion of EMI and how it is practiced in these educational settings should be critically adapted and evaluated with reference to the needs of the local context.Second, the fact that there is room for other languages in EMI classrooms raises the question of whether EMI should mean English-only, particularly in multilingual contexts such as Indonesia.The mismatch between official policy and actual practice may inform that there needs to be a re-orientation of the institutional aim to accommodate multilingual language policies to better reflect how English is practiced within multilingual contexts (Kirkpatrick, 2014).Third, this study calls for a range of ESAP support mechanisms to address EMI students' specific needs.It also suggests that intensive collaboration between content and language teachers could be beneficial in developing a curriculum model that links both content and language courses tailored to meet the required language demands.This study has several limitations.Since it only focused on universities located on Java Island, the generalizability of the findings is uncertain as there is also a growing number of universities offering EMI in other parts of Indonesia.Future studies could cover a broader range of contexts and investigate more stakeholders' perspectives.Furthermore, this study did not consider students' content areas as a variable that might influence their attitudes toward EMI and language support.Further investigation may take into account aspects of students' majors to explore the role that language plays in different academic fields.Lastly, students' proficiency levels in English and prior academic backgrounds, e.g., whether they used English as MoI in high schools, could also affect their attitudes towards EMI.These variables could also be further considered in future research.

Figure 1
Figure 1 Distribution of Gender

Figure 2
Figure 2 Distribution of Age T1: "All freshmen whose English proficiency tests [administered by the university] are below 457 must attend the English course.It [this course] is part of for new students and it focuses on the core language skills."(University B)

Table 1
Information about the Research Setting

Table 2
Framework for Interpretation of Descriptive Statistics

Table 3
Motives for Students Enrolling in EMI Programs

Table 4
Students' Attitudes towards EMI teachers Notes: SA = Strongly Agree; A = Agree; N = Neutral; D = Disagree; SD = Strongly Disagree

Table 5
Students' Attitudes towards English in EMI Notes: SA = Strongly Agree; A = Agree; N = Neutral; D = Disagree; SD = Strongly Disagree