Roles of English in English-Medium Instruction (EMI) University Settings in Latin Americaa transnational multiple-case study

Abstract

By adopting a qualitative approach, the thesis, which was framed as a transnational multiple-case study conducted in two international universities, one in Colombia and one in Brazil, aimed to gain an understanding of how English as a medium of instruction or EMI affects the roles that English plays in international university settings in Latin America and which specific roles, or functions, English performs in these settings. The study followed the principle of replication, so data collection included in-depth interviews with EMI professors and institutional documents in both universities. Also, national education and language policies were examined in order to gain a deeper understanding of the cases under study. The dimensions of the ROAD-MAPPING framework, and more specifically Roles of English (RoE) (Dafouz & Smit, 2017, 2020), served as a starting point for data analysis. After the iterative processing of the data, a codebook built upon a mixture of theory-driven and data-driven elements became the prime method for conducting a systematic and rigorous analysis process. The data drawn from each case we, first, analyzed individually; then, a contrastive analysis across cases was performed. Findings suggest that the ways in which EMI affects RoE in international university settings in Latin America are highly dependent on five factors: societal, communicational, institutional, pedagogical, and individual or personal. The first four factors were proposed by Dafouz and Smit (2017, 2020); the fifth factor emerged from this transnational multiple-case study, thus, providing evidence that RoE in EMI settings are situated and context-sensitive. In addition, seventeen (17) different RoE shaped by the meaningful discursive elements drawn from the rich conversational and documentary data emerged in this study. While there were a few differences (i.e., certain roles were present in one university, but not in the other), thirteen out of seventeen RoE (13/17) were found in both cases. Both the national context (e.g., the role historically assigned to the national language) and the institutional context (e.g., one university was private while the other one was public) were found to exert an influence on both the differences and the similarities identified in the findings. The study also yielded to a set of pedagogical implications which may be of potential use in other EMI settings including but not limited to Latin America. First, international universities adopting EMI should recognize the value of plurilingualism from the beginning so that professors and students would not have to be under the pressure of an English-only policy. In fact, the use of the L1 in the EMI classroom has proved to enhance student learning and motivation. Second, professors teaching through English should be aware that EMI may bring about changes to their professional identities; if universities are careful enough to inform professors of this prior to the adoption of EMI (e.g., through professional development), they would be preventing feelings of inadequacy, anxiety, or frustration among professors. Last, universities which adopt EMI should offer opportunities for professors to engage in professional communities of practice (CoPs) with other colleagues who are teaching EMI courses, thus, ensuring spaces for research collaboration and peer support among professors.