Despite the rapid growth of English Medium Instruction (EMI) programmes around the globe, the language goals in such programmes have not been clearly defined, making it difficult to evaluate their success. Accordingly, our paper reports on a study that attempted to identify the language goals, operationalised as the vocabulary expected of learners, by investigating the under-explored area of EMI assessment. We sampled a corpus of 16 sets of examination papers in two subject areas, namely, Biology and Geography, spanning eight years in Hong Kong. The examination papers, analysed for their vocabulary profiles, indicated that students were expected to learn beyond the K3 level (the most frequent 3,000 words in English) and master mid-frequency words (K4−K8) as well as academic words. Our findings suggest that the less frequent words were not often shared by the two subjects. Analysing academic words also enabled us to showcase how some words were common across the two subjects while others were more subject-specific. Pedagogical implications are discussed, highlighting the need for teachers and students to attend to target language goals in EMI, and to consider the subject-specific nature of vocabulary items in EMI. Copyright © 2023 Informa UK Limited, trading as Taylor & Francis Group.

This study takes a rarely employed learner's angle in examining how students are learning in the English Medium Instruction (EMI) classroom. Particularly, the study tapped into students' listening strategy use when comprehending the teacher input, a context that is underexplored for strategy research and yet highly important because teacher input predominates in the EMI classroom. An EMI class of Grade 11 Biology in Hong Kong was recruited. Through lesson observations and stimulated recall interviews with a purposively selected sample of eight students, a taxonomy of listening strategies in EMI was constructed. The taxonomy reveals the uniqueness of listening to the teacher in EMI context due to the non-existence of planning and prediction strategies, the identification of novel strategies, and the classification of content-mediated and language-mediated cognitive strategies. Factoring in students' linguistic knowledge and Biology achievement, the findings indicated the usefulness for students to deploy a range of language-mediated, content-mediated and metacognitive strategies in combination. This paper ends with implications for future research and pedagogy in order to make EMI learning and teaching more successful. Copyright © 2023 Elsevier Ltd. All rights reserved.

This paper investigates the language ideologies held by a group of international students in an English medium instruction (EMI) university in multilingual Hong Kong. The findings indicate that the participants' beliefs about English extend beyond its role as an instructional medium and encompass its use as a lingua franca and as a means of social inclusion. It is also found that the participants hold complex yet sometimes contradictory ideologies about the kinds of English to be used and/or accepted and about the monolingual and multilingual uses of English in the EMI university context. Taken together, the study suggests that the participants' ideologies with respect to the roles of English and the norms of English use in the EMI university context are much more complex than what language policy from above suggests. The findings also reveal the ideological tensions over the norms of English use in the EMI university context and the closely intertwined relationship between the participants' language ideologies and the maintenance of their interests, especially in relation to access to learning opportunities in the classroom. The study calls for attention to broader issues pertaining to social inclusion/exclusion, linguistic advantage/disadvantage and educational equality/inequality in the multilingual EMI university context. Copyright ©2021 Routledge.

Recent studies on classroom discourse have challenged the traditional classroom role set and emphasised equal contributions from the participants and the emergence of knowledge through active participation. Co-learning emphasises the process in which teachers and students attempt to adapt to one another's behaviour and learn from each other in order to produce desirable learning outcomes. Current research has paid little attention to the ways in which content teachers and students jointly negotiate new knowledge in bi/multilingual classrooms. Based on data collected from a linguistic ethnography in Hong Kong English-Medium-Instruction secondary mathematics classrooms, this paper uses translanguaging as an analytical perspective to analyse how the EMI teacher and students co-learn in the classroom. The data are analysed using Multimodal Conversation Analysis and triangulated with the video-stimulated-recall-interviews which are analysed using Interpretative Phenomenological Analysis. The paper argues that translanguaging creates a safe space for co-learning that emphasises equity in knowledge construction and challenges the hierarchical relationship between the teacher and the learner. Copyright © 2021 Routledge.

The role of the first language (L1) in second language (L2) learning has been widely discussed and researched in the second Language Acquisition (SLA) literature. Benefits of L1 use have been found for vocabulary acquisition, and in facilitating comprehension thereby alleviating student anxiety. Many teachers appear to consider L1 use as a necessary option in L2 pedagogy. Disadvantages of L1 use are argued to be that it reduces exposure to the L2 thereby not enabling more implicit forms of L2 acquisition. Considerably less research has been carried out in English medium instruction (EMI) classrooms, surprisingly, as these are predicated on the benefits of massive exposure to the L2. This study investigated the effect of L1 and L2 use on teacher question types and interaction patterns in science lessons in early EMI and late EMI schools in Hong Kong. Our findings show that in late EMI, where there was much greater use of the L1, teachers tended to use higher order questions and were more interactive. In early EMI, despite students having received four more years of greater exposure to English instruction than late EMI students through the use of English as the vehicular language, teachers asked lower order questions and the interaction was more monologic. The implications for policy and pedagogy are discussed.
[Copyright of International Journal of Bilingual Education and Bilingualism is the property of Routledge.]