| Abstrakt: |
English-as-the-medium-of-instruction (EMI) and content-and-language-integrated-learning (CLIL) in English both involve teaching about course content beyond language and enhancing English proficiency. One area where they differ is classroom assessment: Assessment in EMI primarily focuses on learner understanding of course content, with English used as the language for learners to display their knowledge. This distinction has implications for EMI vs CLIL teachers in writing test items and assessing student responses. In CLIL tests, students' English usage of targeted language features can be an integral part of the construct assessed in a test item, and thus targeted vocabulary, grammar, etc., would be appropriately included. However, in EMI tests, English usage is not a part of the construct being assessed (course content), and thus test items should not introduce construct-irrelevance through inclusion of aspects of English grammar and vocabulary unfamiliar to or problematic for students due to differences between their first languages and English. This distinction between EMI and CLIL has implications for EMI teachers writing and interpreting tests about course content. This teacher narrative analyses two teachers' observations of instances of English language interference in EMI university tests in a second language acquisition class (Eastern Europe) and an Arabic heritage class (Middle East), identifying five sources of interference: Unfamiliar English words, familiar English words in unfamiliar syntactic and/or semantic context, course content vocabulary in a different syntactic context, sequential prepositional phrases, and item length. This teacher narrative offers insight into possible sources of language-related interference and describes steps taken to reduce such construct-irrelevant variance. [ABSTRACT FROM AUTHOR] |
