English learning and use in Hong Kong's bilingual education: Implications for L2 learners’ development of interactional competence

Abstract

Abstract The use of English in international communication has focused scholarly attention on L2 learners’ development of interactional competence (IC), that is, the ability to co‐construct interactions purposefully and meaningfully in specific contexts. Under Hong Kong's bilingual education policy, students mainly learn and speak English in two classroom contexts: (1) English language and (2) English‐medium‐instruction (EMI) content‐area subject lessons, which may contribute to their IC development. This study compared the English‐speaking/learning experiences of students in these two classroom settings in three Hong Kong secondary schools. Task‐based language teaching (TBLT) was the recommended teaching approach in the English language subject in all these schools, but they adopted different medium‐of‐instruction arrangements in content‐area subjects. Their students also had different academic attainments. The investigation adopted a case study approach consisting of a questionnaire and semi‐structured focus group interviews ( N = 160). Our findings suggest that students generally lacked opportunities to speak English, regardless of the specific school‐based policies. The higher‐proficiency students were given more opportunities to speak English in group/pair activities both in English language and EMI content‐area subjects. In contrast, the lower‐proficiency students received more explicit teaching of the linguistic forms/structures required in English examinations. The paper concludes by highlighting several critical issues concerning students’ IC development in bilingual education, namely, the effects of English‐medium education on students’ speaking ability, implementation of TBLT, the influence of examinations on teaching practices, and teachers’ role in promoting students’ interactional competence through classroom interactions.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 16 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.