Introduction

English medium instruction (EMI) is defined as “the use of the English language to teach academic subjects (other than English itself) in countries or jurisdictions where the first language of the majority of the population is not English” (Macaro et al., 2018, p. 37). Unlike CLIL—a content and language integrated educational model—language development is not the primary goal of EMI policy, but it can be assumed that students gain language skills while learning contents.

Owing to the hegemonic position of English language around the globe, HEIs in many contexts have adopted EMI policies to increase economic competitiveness and gain advantages from globalisation (Macaro, 2018) and internationalisation of higher education and university rankings (Rose & McKinley, 2018). Such presumed advantages of adopting EMI centre on the belief that education through the medium of English can enhance graduates’ English proficiency (although the primary goal of EMI is acquiring content knowledge rather than language learning), individuals’ mobility, and competitiveness in an international job market (Rose & Galloway, 2019). As a result, the provision of EMI programs has significantly expanded in HEIs across the globe.

The adoption of EMI policy in many multilingual sub-Saharan African (SSA) countries is related to post-colonial legacy that favours English over national and local languages because of its neutrality and to foster national unity among its diverse ethnic communities (Clegg & Simpson, 2016). However, the perception of EMI fostering development and international competitiveness has been contested by scholars based in SSA countries (e.g., Clegg & Simpson, 2016; Kamwangamalu, 2013, 2019). Kamwangamalu (2013, 2019), argues that in most SSA contexts students do not acquire a level of proficiency in English which impede the quality of education, consequently holding back development. Despite this, many postcolonial governments view English as the main means for modernisation, accessing knowledge and economic development.

EMI is a global phenomenon that has many potential benefits but also many challenges. A systematic review of EMI in higher education by Macaro et al. (2018a) highlighted the lack of empirical evidence on the effects of EMI on content and language learning. Thus, research is still inconclusive, as there are many factors that can influence EMI policy implementation and its outcomes. Studies indicate that learning academic content can be obstructed by students’ lack of competence in English. In contexts where the use of English is limited to educational institutions and limited resources exist to provide language support, EMI acts as a barrier to learning contents.

In Ethiopia, the recent expansion of HEIs under massification policies has impacted on an already scarce educational resources to provide additional language support for university students. Additionally, in multilingual SSA contexts (including Ethiopia) many students must learn local as well as national languages before switching to learning academic contents via the third language, English, which causes major challenges in knowledge acquisition contributing to poor educational achievement for many students in tertiary education (Romaine, 2015). These concerns may be left unaddressed, particularly in low-income countries, because EMI does not, by definition, target language learning and content teachers do not consider offering language support as a part of their responsibility.

Proficiency in English is considered to be a key factor to achieve success in EMI and economic development (Rose et al., 2019). To mitigate language related challenges, many HEIs provide language support for their students on EMI programmes in the form of pre-sessional or in-sessional English tutoring. In a review of the literature on the language-related challenges that EMI students experience, McKinley and Rose (2022) argue that English for academic purposes (EAP) and English for specific purposes (ESP) can provide targeted language support for students on EMI programmes to help them achieve academic goals. The current study aims to investigate whether language support policy is incorporated in higher education policy initiatives and universities’ mission statements to address language-related challenges experienced by students on EMI programmes.

EMI policy in Ethiopian higher education

Over the last two decades, EMI has become a dominant educational policy in many contexts, including settings where English had no previous official status, such as Ethiopia. Seeing English as a panacea for modernization and development (Negash, 2011), Ethiopia adopted EMI policy with the establishment of its first higher education institution, University College of Addis Ababa, in 1950 (Simie, 2022). The policy has been reaffirmed in subsequent higher education (HE) policy documents (MoE, 2003, 2009, 2019) that English shall be the medium of instruction in secondary and higher education, except in programmes that require the use of local or foreign languages. Since the introduction of the policy, English has been taught as a foreign language in primary schools but used as the sole medium of instruction in all public and private secondary and higher education as well as research publications.

A highly multilingual and multicultural country with over 85 languages spoken by its ethnically diverse society, the country adopted a multilingual education policy for primary education in 1994 (Ministry of Education, 1994). Since then, children across the country have received eight years of mother tongue medium education. At grade nine, the beginning of secondary school, the medium of instruction switches to English.

English was introduced as a medium of instruction by Emperor Haile Selassie (Yadete, 2017) to foster unity among Ethiopia’s diverse linguistic groups and to enhance a nation building strategy and modernise its education system. However, in the absence of past colonial influence, except the five-year attempt by Italy, adopting EMI policy has been considered a kind of self-colonization by some Ethiopian scholars (e.g., Ramadikela et al., 2020a, 2020b; Woldegiorgis, 2021; Woldeyes, 2022). Unlike postcolonial Anglophone sub-Saharan African (SSA) countries, English is not a dominant language in Ethiopia, and students’ opportunities to practise the language is limited to educational institutions (Simie, 2022; Woldeyes, 2022).

The association of English with modernity and the Westernization of Ethiopian education have been regarded as repressive of the indigenous epistemic system, implying that both the content of the curriculum and the language of instruction disengage students. Negash (2006, p. 33) argued that teaching all academic subjects in English at the cost of one’s native language ‘is tantamount to the wholesale adaption of the culture that the English language represents. Although many ministries of education in SSA countries perceive learning in English as gaining the linguistic capital that provides access to global economy, Romaine (2015, p. 263) argues that adopting English as the sole medium of instruction “will not guarantee the supposed benefits of participation in the global economy to the majority of African students.” Conversely, other scholars in Ethiopia (Negash, 2011; Yadete, 2017) believe that English is key for development and modernisation of society in Africa.

With the presence of EMI for the entire 70 + year history of HE in Ethiopia, it would seem that these universities should function comfortably in English without problems concerning quality experienced in other countries adopting EMI practices (see e.g. Hu et al., 2014). Reports indicate that the quality of education in higher learning institution has been declining; university graduates’ poor academic achievements, inability to read and write basic reports in English, skills deficits and employers’ dissatisfaction have been cited as indications for the decline in the standard of education in Ethiopian HEIs (International Institute of Education, 2012; Teferra et al., 2018). Students’ and teachers’ lack of competence in the language of instruction has largely been blamed for the deteriorating quality of learning outcomes (Bishaw & Melesse, 2017; Teferra et al., 2018). Moreover, compared to many inner-city students, students from rural and lower socio-economic backgrounds are less likely to succeed in their education (Spolsky, 2019), as they often do not have access to English language learning resources and are more likely to face challenges in learning through English at HEIs. In short, owing to limited linguistic ability, students struggle to cope with the demand of learning academic subjects through English. Reported over a decade ago, students and teachers’ lack of competence in the English language had reached “an alarming proportion regarding the quality of education and the graduates in Ethiopia” (Muchie, 2010, p. 156). And more recently, it was reported that as “students [in Ethiopian HEIs] have insufficient proficiency in English, learning remains at a surface level” (Simie, 2022, p. 62).

A need for English language support

Language support for students studying through L2 programmes (English in this case) is believed to provide opportunities for academic success (Galloway & Ruegg, 2020). For students with limited English proficiency, EAP or ESP language support encompasses the provision of additional language assistance in the form of academic writing, communication, or presentation skills (McKinley & Rose, 2022).

Three main models of EMI language support have been identified in the literature: preparatory year, the institutional support, and the pre-institutional selection model. In the preparatory model, students take longer intensive language courses before embarking on EMI programmes (Macaro et al., 2018). The institutional or concurrent support model offers EAP or ESP courses alongside content courses, but language support can be reduced over time. The pre-institutional model uses English language proficiency levels as a requirement to select students, so limited language support is provided on EMI programmes. Research reports indicate that students on EMI programmes often require specific language support (e.g., see Zhou et al., 2022), such as discipline-specific writing skills, as writing is perceived as one of the most challenging language skills for students on English-taught programmes.

In the Ethiopian EMI context, there is no language proficiency requirement for students to transition from secondary to tertiary education (Sahan et al., 2021). Transition to EMI programmes is regulated by the national university entrance exams, which is administered at the end of secondary education. Students who achieve above the fifty percent threshold in their national exams can be assigned a university place by the MoE. Policymakers assume that students who pass their national exams can cope with university education through English. However, owing to exam malpractices including cheating during the national university exams (Teferra et al., 2018), many students pass the exams with poor academic knowledge and language skills.

Several studies have indicated that first-year university students from emerging regional states entering higher education under the government’s affirmative action policy in particular, struggle to cope with learning content through English. A qualitative case study conducted at Addis Ababa University (Shimekit, 2018, p. 39) reported that first-year undergraduate students from lower socioeconomic backgrounds have challenging experiences “due to absence of a language support system in the university”. Despite the positive discrimination policy of admitting students of lower grades from marginalised regional states to university education, inequality of opportunity is still prevalent in the Ethiopian education system (Molla & Gale, 2015; Tesema & Braeken, 2018).

Thus, instead of a ‘one-size-fits all’ language policy, provision of targeted language support to improve students’ general English language proficiency and academic language skills are important as knowledge of English language appears to be significant predictors of students’ academic performance (Macaro, 2020; Rose et al., 2019). Lower-level proficiency students on EMI programmes require more disciplinary targeted language support (McKinley & Rose, 2022) to increase their likelihood of success in their university education. It was based on these reports that we decided to explore governments’ and universities’ positions on language support for students experiencing difficulty accessing contents due to language-related challenges.

Context of the study

Ethiopia, located in the horn of Africa, is the second most populous nation in Africa after Nigeria, and the fastest growing economy in the region (World Bank, 2022). However, the population increase in Nigeria, a former British colony until 1960, has resulted in a massive labour force proficient in English. Ethiopia’s increasing workforce does not have such English proficiency, despite the implementation of EMI policy in all public and private secondary and higher education.

Over the last two decades, MoE has produced several policy documents to help raise the standard of education in higher learning institutions. The main policy initiatives were the 1994 education and training policy (Ministry of Education, 1994), three higher education proclamations (Ministry of Education ), six Education Sector Development Programmes (ESDP) I–VI, and the higher education policy and strategy (Ministry of Education, 2020). Education Sector Development Programme (ESDP) is a policy document that contains core educational priorities and targets to be achieved at all levels of education over a period of five years (Ministry of Education, 2021). HE proclamations are documents that provide a legal framework through which HEIs deliver top-down policies at the institution level. It is through these documents that the MoE has established a language-in-education policy to guide institutions in delivering EMI programmes. However, there appears to be a gap between policy intention and its implementation in teaching and learning disciplinary subjects (Molla, 2013), and there is little research that examines the impact of these policies on the implementation of EMI policy in this context.

With the increased provision of EMI in HE institutions across the globe, substantial studies have only been conducted in Europe and Asia, but “Africa is poorly represented at tertiary level” (Macaro et al., 2018, p. 7), in particular. EMI research in SSA countries, according to Briggs et al., (2018, p. 4), is “exclusively confined to pre-tertiary education”. This study serves to expand the currently limited scope of prominent research in higher education policy, providing insights into interpretations and implementations of EMI in this unique, non-colonial and highly multilingual context of Ethiopia.

Drawing on the investigations of EMI policy in Japanese higher education by Rose and McKinley (2018) and EMI policy implementation at Chinese universities (Rose et al., 2020), this study analyses language support for EMI in Ethiopian higher education policy and universities’ mission and vision statements. Publicly available policy documents were retrieved from MoE and university’s official websites to explore the contents of the policies in relation to language support for EMI programmes.

The study is guided by the following two research questions.

RQ1:

How are Ethiopian higher education English-medium instruction policy objectives stated in universities’ mission and vision statements?

RQ2:

How do these universities position themselves to address language concerns in the implementation of English medium instruction policy?

Each question dealt with a different level of policy analysis. The first question requires analysis of macro level policy from a top-down perspective, as dictated by the Ministry of Education. The second question examines how universities plan at the meso (institutional) level to address language related challenges in the implementation of EMI policy.

Theoretical framework

Language policy encompasses linguistic behaviours, perceptions, and decisions of a community (Spolsky, 2004). These elements, while interconnected, can be distinctively described. Language beliefs, or ideology, reflect deep-rooted views on appropriate language choices, influenced by the prestige of specific languages (Spolsky, 2004). Language practices denote actual linguistic behaviours, serving as real language policies that provide frameworks for learning (Spolsky, 2009). Language management, meanwhile, represents deliberate efforts by authorities to regulate language practices or modify beliefs (Spolsky, 2009).

Framed by this understanding of Spolsky’s tripartite definition of language policy, our study draws on Bourdieu’s theories as conceptualised in critical EMI research conducted by Sah (2022). To analyse the phenomenon of English medium instruction policy and practices in multilingual schools in South Asia, Sah (2022) uses Bourdieu’s (1993) concepts of linguistic capital and linguistic marketplace as a theoretical framework. Linguistic capital refers to the value attached to a language or variety of languages, and linguistic marketplace being the social space where languages compete for recognition can create language hierarchy in society. Using this framework as a lens, Sah examines how English in the context of EMI is perceived as a valuable linguistic resource that can open doors to the global economy and social mobility on one hand, and how EMI policy creates inequality of opportunities for different social groups in multilingual contexts, on the other hand. Based on these two concepts he explores the development of EMI policy and its ideological and pedagogical motivations, EMI policy models in multilingual contexts, and the social justice concerns that arise from such policies and practices.

In the current study, we adopted a similar framework to analyse the role of English medium instruction policy in achieving Ethiopian higher education objectives. Focusing on EMI, we examine language support policy to enhance the effective implementation of EMI policy and achieve HE policy initiatives.

Research methods

In this study, we analysed the contents of both HE EMI policy and universities’ mission statements to understand language support policy designed to address language-related challenges that students on EMI programmes experience. As Bayrak (2020) points out, mission statements are good sources of information to understand universities’ purposes, values and objectives. Accordingly, the purpose of this study was to determine how universities position themselves in their mission and vision statements in relation to HE EMI policy statements.

The study used qualitative content analysis (QCA) for searching repeated patterns that appeared from the data. Selvi (2020, p. 442) explains that although it is qualitative, QCA allows for “subjective interpretation of the contents of both qualitative (assignment of categories) and quantitative (the use of text passages and analysis of frequencies of categories) steps in a systematic and context-dependent manner.” In this study, government produced HE documents were retrieved and salient concepts in the policy statements were identified. These terms were identified based on their keyness (i.e., the statistical significance of the frequency of keywords in the corpus. Analysis was conducted following methods used in Rose and McKinley (2018) EMI language policy study, which was interpretive and qualitative rather than corpus-like; and an initial review of the texts revealed subtle nuances and differences in terms of how language-related issues were discussed. As the lead author of the current study has lived experience as a student and teacher in the Ethiopian education system, subjective interpretation has been reflected in the data analysis.

Data collection and analysis

For the main data source, we identified governmental HE policies on MoE websites via document data collection (Rose & McKinley, 2018) and mission statements from 42 (out of 50Footnote 1) Ethiopian universities. Google was used to search for and identify relevant higher education policy documents on the official websites of the MoE and the public universities in the country. We limited our search to documents published since 2015, as major HE documents introduced after 2015 are the updated versions of older documents. For instance, HE proclamation 2019 is an updated version of proclamations 2003 and 2009. Additionally, this limit made our data more manageable and applicable to our research questions. Accordingly, first, we retrieved four documents related to HE policy initiatives from MoE websites, which include HE policy and strategy 2020 (Ministry of Education, 2020), HE proclamation 2019 (Ministry of Education, 2019), two education sector development programmes (ESDP) V 2015/16–2019/20 (Ministry of Education, 2015) and ESDP VI 2020/21–2024/25 (Ministry of Education, 2021). Then, we retrieved a complete list of universities’ names and their corresponding web address from the Ethiopian national educational assessment and examination agency’s website (see Neaea, 2019). We used the universities’ official web address to retrieve textual data related to mission and vision statements as well as language policy wherever it was available. When any universities’ web address was inaccessible, we used the key words including the institution names, mission statements, English medium instruction, English language improvement programme, and higher education policy to search for relevant policy data.

Based on their year of foundation, public universities were categorised into first, second, third and fourth generations. The 8 oldest universities are called first generation universities (hereafter 1st GUs). The second and third generation universities consist of 13 and 12 universities respectively. The 11 newly established universities are categorised as fourth generation universities (4th GUs). In total, there are 50 public universities in Ethiopia of which 8 universities including universities in the war-torn region of Tigray do not have publicly accessible websites to retrieve mission statements at the time of data gathering. As a result, we only scanned the 42 universities’ website in the search for data pertinent to our research questions.

To respond to RQ1, data were analysed in two phases. In the initial phase, we used the broader higher education objective statements to identify and analyse key contents pertinent to EMI policy. The data for this study come from documents produced by the Ministry of Education, which includes higher education policy and strategy, higher education proclamation 2019, education sector development programme (ESDP) V and V1. The wording of the documents was coded in line with how they positioned key terms related to higher education objectives including language support policy, but not limited to, global competitiveness and internationalisation. NVivo 2020 was used to code the data. Open coding was employed for coding the documents by reading through relevant parts of the texts line by line to identify patterns of salient contents.

In the next stage of the research, we analysed how universities at the meso (institutional) level communicated the top-down HE policy in their mission statements. To achieve this, data collected from 42 universities’ mission and vison statements were imported to NVivo for coding and analysis. By reading and annotating the texts of the mission statements, initial codes and categories were generated. Prior to coding, each website was searched using Google for the keywords mission or vision together with the respective names of the institutions. Then, textual documents identified from each university’s website were coded and analysed in line with key concepts in HE policy with particular focus on EMI policy implementation. To be more specific, we used key terms and concepts identified in the relevant part of HE policy statements as a frame of reference to code the mission statements. We coded the entire mission statements retrieved from the universities’ websites.

Our analysis of the contents of each university’s mission statements involved comparing key identified terms with different groups of universities and official HE policy statements, which enabled us to identify commonalities and differences in the institutions’ areas of focus and to understand underlying issues regarding EMI policy. We used quantitative data (percentages) to measure the prevalence and frequency of the terms in the mission statements. As Selvi (2020) explains, there can be quantitative elements to qualitative content analysis where the quantitative refers to ‘the use of text passages and analysis of frequencies of categories’ (p. 442). Additionally, Kaefer, et al. (2015) stress that QCA can include quantitative elements such as word frequency analysis allowing a focus on quantitative aspects of the research data that are most relevant to the research question. In the current study, by integrating the qualitative content analysis with quantitative data, we aim to provide a clearer understanding of the extent to which the universities’ and HE policy statements align with each other, most importantly in terms of providing language support for EMI programmes.

To address RQ2, we searched each university’s website for information related to language concerns in relation to the implementation of higher education policy initiatives through English medium instruction. Scanning lists of programmes offered by the institutions, we identified two language programme types through which universities position themselves in addressing language related challenges. The programmes, which include one mandatory and the other optional, are intended for first-year students on undergraduate programmes.

Findings

RQ1:

How are Ethiopian higher education English-medium instruction policy objectives stated in universities’ mission and vision statements?

To respond to RQ1 we conducted two stages of data analysis. First, we analysed HE policy documents and then universities’ mission statements, as presented below.

Analysis of HE policy documents

In the first stage of data analysis, initially we retrieved four publicly available HE documents from MOE websites (see data collection and analysis section), and coded and analysed the data in line with key themes in the higher education policy statements.

Shown below are items clustered by code similarity generated from NVivo, which illustrates key coded concepts in higher education policy documents.

As evident from Figure 1, our analysis of higher education documents illustrates core educational priorities, which include:

  • Producing competent graduates

  • Offering quality education

  • Providing community services

  • Conducting research

  • Fostering economic growth

  • Enancing technology

  • Global competitiveness

  • International collaboration

Figure 1
figure 1

Ethiopian universities' core educational priorities

Full size image

The language policy, which stipulates English as the medium of instruction through which these policy goals can be achieved, has also been identified in the higher education policy documentation. As our research questions centred on language related challenges and subsequent implications for the implementation of EMI policy in the process of achieving higher education policy goals, we identified English language improvement centres and language training in the higher education policy documents. Stated in the ESDP V (2015) policy document, “the development of English language skills among teachers will be given high priority through the English language improvement programme for teachers” (p.58). This implies that policymakers recognise the need for language support for subject teachers to overcome linguistic challenges they experience in the implementation of EMI policy at the micro level. However, students are expected to be multilingually proficent, as stated in the ESDP V1 (2021) policy document:

‘Every learner will be proficient in his/her mother tongue, the official national language(s) of communication, the interregional languages(s) and English as a language of international communication’ (p. 29).

Although these documents clearly state English as the only medium of instruction for all curriculum area, except for Amharic, it does not provide a guidance for students’ language support to achieve the required level of proficiency necessary for successful educational outcomes. Here it is worth noting that to achieve the HE policy proiroities identified above, students’ mastery of the language of instruction appears to be essential.

Also linked to these are the concepts of international collaboration and attracting international students and staff, which are strategies identified in the HE policy statements for internationalisation of universities. As English has become a lingua franca in global HEIs, competency in the language is vital for universities to benefit from internationalisation. Although internationalisation has been a focus of Ethiopian HE policy, only a very few of its universities are visible on African academic plaforms, such as African university rankings.

Analysis of universities’ mission and vision statements

In the subsequent phase of our data analysis, we explored how various university groups interpret government higher education policy statements within their mission and vision statements. While the majority of the content from these policies has been integrated into the universities' mission statements, there are discrepancies in the inclusion and omission of certain policy terms. Notably, language support for students, a significant policy term, is often overlooked in these mission statements. Table 1 presents the results, which have been coded and quantitatively analyzed, from the four university groups.

Table 1 Crosstab query
Full size table

In our analysis of the universities’ mission and vision statements, providing community services, conducting research, and producing competent graduates were the most coded key terms in the university’s documentations, which implies that provision of local focused service, research outputs and producing a skilled workforce are the three major priorities for most Ethiopian universities. Also, we found that different groups of universities have emphasised different key contents of higher education policy in their mission statements. For instance, the key concepts conducting research and leading in Africa (vision to be one of the premier universities in Africa) were the most coded term (87.5%) in the mission statements of the 1st GUs. As shown in the table above, global competitiveness, a concept inextricably linked with internationalisation and most importantly with the use of English in EMI universities in Ethiopia, did not appear as key term in 2nd and 3rd GUs.

The key terms in the data such as leading in Africa, (part of universities’ internationalisation agenda), and research outputs, (both key concepts implicitly linked with English language skills), appeared most (87.5%) in the mission statements of 1st GUs. For example, Addis Ababa, Bahir Dar, Jimma and Haramaya universities (all 1st GUs) aspire to become one of the top ten premier research universities in Africa by 2025. However, ranking in African universities does not seem to be a priority for most other institutions in this study as the code leading in Africa, referenced only 52.38% overall in our analysis. Similarly, the key term global competitiveness, which is linked with internationalisation, was referenced in 25% of the statements by the older universities (1st GUs) but only 11% overall. As this finding reveals, there is variability of policy focus in the mission statements of different groups of institutions. Notably, global competitiveness / internationalisation—key contents in HE policy statements—is the least endorsed concept in the universities mission statements.

RQ2:

How do these universities position themselves to address language concerns in the implementation of English-medium instruction policy?

In response to RQ2, we scanned the official websites of 42 universities and managed to identify two types of language programmes—the communicative English language skills course and English language improvement programme (ELIP)—adopted by the universities to address language related challenges. The details of this finding are discussed below in turn.

Communicative English language skills course

By scanning universities’ websites, we found that a communicative English language skills course is a mandatory common course offered to all first-year students on undergraduate EMI programmes. The course is divided into two modules, communicative English language skills I and II, offered concurrently along other disciplinary subjects over two semesters during the first academic year. The communicative English skills modules and all common academic courses intended for first-year students can be accessed on most universities’ websites (e.g., see Addis Ababa University, n.d.; Arsi University, n.d; University of Gonder, n.d). Module one of the communicative English courses, given course code Enla1011, is scheduled for semester one. Module two (also known as communicative writing course) has the course code Enla1012. We also noted that other institutions as well as the module writers use the code FLEn1011- instead of Enla1011- FLEn being the short form for foreign language English (e.g., see Ambo University, n.d.; Ferede et al., 2019; University of Gonder, n.d.). The intended goal of both Enla1011 and 1012 modules is to develop first-year students’ general English language proficiency in all four language skills including vocabulary knowledge for the students’ academic and social needs (Ferede et al., 2019; Yigzaw, 2020). Although universities offer Enla1011 and Ena1012 to improve students’ academic writing and reading skills, including speaking and listening, with the assumption that skills acquired from the courses will help students study disciplinary subjects through the medium of English, research shows that students still struggle to learn through English (Simie, 2022).

In many EMI settings, admission to EMI programmes often requires specific English proficiency levels (McKinley & Rose, 2022). There are no language proficiency requirements for students to transit from secondary to tertiary education in the context of Ethiopia. Entry to higher education is regulated by national university entrance exams (Sahan et al., 2021). However, as is the case in many EMI contexts, it is quite common for students to pass the entrance exam with poor English language proficiency due to limited exposure to the language in secondary education. Hence, Enla1011 and Enla1012 are deemed vital language support courses to help learners cope with learning disciplinary subjects. According to Sahan et al. (2021), this form of support offered to address language concerns is the most common form of support available to students on undergraduate programmes in many EMI contexts. However, the extent which this concurrent language support courses meet students’ language need is beyond the scope of current study.

English language improvement programme (ELIP)

Launched by the government in collaboration with the British Council based in the capital, the purpose of the ELIP (also known as ELIC—English language improvement centre) was created to provide language support for students and teachers in schools and public universities in the country. Based on individual language needs, institutions provide English language trainings to students outside normal teaching hours, usually in the evenings or at the weekends. ELIP is a self-access model where students receive support in writing centres. However, scanning the 42 institutions’ websites, we found only a limited number of universities have established ELIP centres.

Six out of eight (75%) of 1st GUs indicate on their websites that they provide language support (e.g. see Bahir Dar University, n.d.; Haramaya University English Language Improvement Centre Department, n.d.; Jimma University, n.d.) for their students and academic staff through ELIP centres on their campuses. However, out of thirteen 2nd GUs only three state that they provide language support to students with poor English proficiency. For instance, at Adama Science and Technology University, according to information published under Division of Freshman Programme (see Adama Science and Technolgy University, 2017), first year students with poor English language skills and those who joined the university from remote rural high schools were identified at the beginning of term and assigned a tutor for English language support. Additionally, through scanning the official websites of the 3rd and 4th generation universities we found that only two of the twelve newly established universities offer English language support to students and teachers through ELIP centres in their universities. In scanning all 42 universities’ websites, we found that only 11 (26%) have such provision. This finding aligns with the MoE report (Ministry of Education, 2021) that acknowledges inconsistency in implementing ELIP due to a shortage of English language training centres.

This finding is significant because, despite mandated requirement of universities through higher education policies (as we found in response to RQ1 in this study), few higher learning institutions position themselves to address language concerns. Although the government policy initiative (Ministry of Education, 2020, p. 40) to “establish English Language Improvement Centres in higher learning institutions that aim at improving language capacity and proficiency in academic staff and students” is in place, failing to implement the policy across the HEIs may have an adverse effect on students’ academic success.

In response to universities’ failure to provide language support, the MoE has recently agreed to collaborate with the American embassy in Ethiopia (U.S. Embassy Ethiopia, 2022) to expand English language improvement centres in HEIs in the country. To sum up, in our analysis of publicly available data, we identified English language courses and ELIP centres as institutional/concurrent support intended to meet language-related challenges for successful implementation of higher education policy through the medium of English in Ethiopian universities.

Discussion

We analysed the contents of both HE policy and universities’ mission statements to understand language policy to address language related challenges and enhance the successful implementation of EMI policy at the university level. To our best knowledge, this is the first study (at least in the context of Sub-Saharan Africa) that explored higher education policy interpretation in institutional mission statements formulated for the implementation of EMI programmes.

University mission statements vs. practices

Our findings were consistent with previous studies that revealed relationship between the use of English and producing competent graduates as one of the most prominent concepts in the mission statements of educational institutions (Holosko et al., 2015). Directly linked to the government agenda of economic growth and poverty alleviation, producing skilled work force proficient in English (proficiency in English in this case is assumed, not stated) was a major HE policy initiative, as identified in the current study. However, although more than 60% of the universities mapped the term producing competent graduates to its mission statements, studies in Ethiopia (e.g., Rekiso, 2019; Woldegiyorgis, 2017) reported that HEIs have continued producing graduates that cannot verbally express themselves or write reports in English in workplaces. These poor educational outcomes are attributed to low-quality instruction as well as poor proficiency in English (Tadesse et al., 2018; Tessema, 2009). Our study reveals, despite the presence of sufficient evidence about the adverse impact of students’ limited language ability on successful implementation of EMI programmes, language support policy has been an important missing link in most universities’ mission statements.

Only a few 1st GUs located in major cities provide language support (as stated on their official website) outside normal teaching classes to help students improve their English proficiency. This provision seems to be even more important for students from disadvantaged backgrounds especially in the newly established universities located in the marginalised regional states. Owing to the expansion of universities, enrolment has increased, but the gap in academic success among university students seems to be widening as students from better socio-economic backgrounds have better access to English learning resources (Tesema & Braeken, 2018) contributing to their chance of success in education and employment.

Additionally, graduates’ poor educational outcomes have been extensively reported in the literature despite the fact that it was stipulated as a priority HE objective in the government policy documents (Ministry of Education, 2020). Despite being one of the key themes in HE policy, fostering quality education was covered by less than half (42.86%) of the mission statements across the universities, and its link with the use of English again was assumed rather than stated. Poor educational achievement has been associated with “students’ poor command of the English language” (Akalu, 2016, p. 269) which aligns with the literature on EMI studies in Ethiopian HEIs (Simie, 2022). For instance, there is a lack of clearly defined language support policy for students from underrepresented regional states entering HEIs with their university entrance exam results below the 50% threshold based on special admission (affirmative action) policy.

Research is another key concept we explored in our data analysis. Studies indicate that owing to a dearth of scholars engaging in sound educational research in English, and lack of sufficient funding, research outputs in HEIs appeared to be very low (Simie, 2022). Published research outputs are in English appearing in both national and international journals of varying prestige. According to Salmi et al. (2017), most research outputs in Ethiopia are produced by postgraduate students and a few academics based in older universities such as Addis Ababa, Haramaya, Jimma and Mekelle Universities. Unsurprisingly, according to Kahsay (2017), the impact of universities’ academic research on the country’s economic development has been very minimal. Surprisingly, however, in our study, the term research outputs were the second most (66.67%) prominent theme after community services in the universities’ mission statements. This raises questions whether contents in mission statements turn to educational outcomes, which in turn raises questions about faculty members’ research skills, but most importantly, about their command of English, a language of research publication. As McKinley and Rose (2018) highlight, competence in English language is considered vital for research publication as well as internationalisation.

In a similar vein, in our analysis of universities’ mission statements, 87% of 1st GUs aspire to be top research universities in Africa, which is linked to internationalisation. However, even Addis Ababa University, the flagship university in the country, has a long way to go to achieve the status of being a pre-eminent research university in Africa, as “the quality and quantity of research publications by staff members has been deteriorating for the last decades” (Sisay & Moges, 2014, p. 299). This vision of the university is linked to researchers’ command of English which is essential to produce and publish in high-ranking journals. However, to achieve international recognition and other major higher education objectives, the role of the English language has not been clearly stated in any of the universities’ vision and mission statements, which we identified as an important missing link. The diminishing research outputs can also be justified, by the fact that research resources allocated for HEIs accounted for only 1% of their total budget (Ministry of Education, 2015). However, the MoE avoids reflecting explicitly on the impact of faculty members’ English language ability on research outputs.

Our assessment of policy content in the mission and vision statements reflected identifiable disparities as well as commonalities across the institutions. Mission statements are important to communicate HE objectives to their stakeholders. However, as some scholars argue, the contents of the statements do not always seem to reflect the educational outcomes that the institutions claim (e.g., Holosko et al., 2015), which is the case in some of our findings. As a result, the findings of this study need to be interpreted with caution. We also acknowledge that our study is limited to textual data analysis to reach a comprehensive conclusion about the effects of HE policy and mission statements on the implementation of EMI policy. A future study that encompasses all stakeholders (policymakers, teachers, students, and university officials) may be able to explore policies and strategies on how to raise the standard of English language to achieve higher education policy objectives.

What is at stake with issues of competence and proficiency in English?

In the implementation of EMI policy in multilingual Ethiopian HEIs, proficiency in English holds profound implications for both students and teachers. For students, proficiency challenges could potentially impede their comprehension of complex disciplinary subjects, limiting their ability to actively engage in academic contents and resources, hindering overall academic performance. In this highly multilingual context, students from diverse linguistic backgrounds grapple with a double linguistic burden as they are neither proficient in English nor in Amharic (the national language) to comprehend cognitively challenging academic contents. The consequences extend beyond academic settings, impacting students' future employability and professional success in an increasingly globalised job market. In Ethiopian higher education, both students’ and teachers’ lack of competence in the language of instruction can compromise the effectiveness of EMI, leading to a compromised educational experience and academic achievements for the graduates.

Recognising variations in English language proficiency among students and teachers, policymakers need to consider the benefits of translanguaging practices, which are already happening in most Ethiopian university classrooms, except where teachers have advanced levels of proficiency (i.e., Addis Ababa University). The use of major local languages (e.g., Tigrigna, Afan Oromo, Somali, and Amharic) alongside English needs to be introduced into policy to allow students and teachers to express themselves effectively and facilitate better comprehension of academic content. This may be done by considering more locally grounded and flexible forms of multilingual education that may address the needs of concerned teachers and learners. Policymakers and teachers alike can then pay more attention to local forms of knowledge that may be emerging and which may otherwise be ignored.

The inclusion of local languages in Ethiopian higher education curriculum, as is done in early years EMI education in Sub-Saharan Africa (see Milligan & Tikly, 2016), can enhance local knowledge and mitigate educational inequalities contributing to a more equitable and accessible education (see Clegg & Simpson, 2016). In short, as Ethiopian universities are highly multilingual, reimagining a flexible language policy that fosters linguistic diversity, and draws on the benefits of the range of linguistic repertoires available in Ethiopian educational contexts, can promote a more inclusive educational environment and mitigate the unintended consequences of linguistic hierarchies.

Conclusion

Generally, for Ethiopian higher education students to have equal opportunity of accessing learning through the medium of English, and for the graduates to generate knowledge through research and take part in international academic activities, competency in English language is vital. Furthermore, Ethiopia’s development has been very much dependent on university education through the English language to produce a workforce with skills, knowledge, and abilities that the economy requires to grow. Indeed, the internationalisation of higher education through EMI is assumed to promote a dynamic and interconnected learning environment, preparing graduates to navigate the complexities of a globalised economy. As Ethiopia seeks to enhance its standing in the international academic community, an effective EMI policy serves as a cornerstone, facilitating the integration of Ethiopian higher education into the broader global discourse and contributing to the country’s economic development through a more globally competitive workforce. Therefore, the MoE and the respective universities should collaborate in establishing and implementing industry informed ELIP centres across the higher learning institutions that address students’ specialised academic and social needs (see e.g., Al Hilali & McKinley, 2022).

In conclusion, education plays an important role in promoting development in a society, and language, especially English, is essential to facilitate it. Therefore, Ethiopian HEI expansion to increase access and participation in university education alone brings about neither equality and quality in education, nor does it take the country out of poverty unless there is a change in the standard of the language of instruction. For Ethiopian universities to achieve higher education objectives through EMI policy, encouraging translanguaging practices, offering language support to help students cope with the academic and social demand of EMI, and providing professional development for teachers to enhance their pedagogical and linguistic skills are important factors. Instead of a one-size-fits approach, policymakers may also need to adopt a multilingual approach that encourages translanguaging practices that allow the use of student’s linguistic resources that may help to achieve a better educational outcome for the graduates.