Assessing Teachers' Language Proficiency

Abstract

The question of what language skills and competence are required of teachers, in order to be able to deliver lesson content appropriately and effectively, has been the subject of considerable debate given the growing numbers of non‐native teaching professionals employed in contexts where the medium of instruction is not their mother tongue. On the one hand, there is a view that a teacher should be native‐like or have complete control of the language that is the medium of instruction, the object of instruction, or both; on the other hand, it is argued that teacher language proficiency standards should be defined in relation to the particular context of instruction and in light of what is realistic given available training resources. The chapter demonstrates how these competing views play out by reviewing a variety of models for assessing the language proficiency of teachers as operationalized in the USA, the UK, Australia, Hong Kong, Brazil, and elsewhere. The models range from those that characterize the teacher proficiency construct in general linguistic terms to others that conceive of it as context specific, and as hence requiring performance‐based tasks and assessment criteria that mirror the target language use domain in very particular ways; for example, by requiring the test candidate to assume the role of the teacher when undertaking test tasks, or by assessing the teacher in the actual classroom situation. Contexts considered include higher education, with its demand for international teaching assistants capable of delivering high‐level subject content in a manner that is intelligible and acceptable to students, and school foreign language classrooms, where metalinguistic knowledge or knowledge about the target language can be considered part of the assessment construct. The chapter also considers issues of policy surrounding the delivery of teacher language tests or assessments and their impact on educational systems.

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 — 32 citations exist upstream.

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