Marking L2 written exam answers in an EMI context: no effect of ELF and no foreign language effect on grading by university lecturers

Abstract We aim to contribute to the discussion on whether the use of English in higher education in an international setting affects study outcomes. Previous research comparing instruction in English versus instruction in the students’ L1 yielded contradictory results. Our concrete interest was how the reader-marker of exams evaluates the content knowledge of a student-writer who expresses themself in English as a lingua franca. The experiment consisted of grading tasks in two language conditions complemented by an online questionnaire. We presented L1 Dutch professional university lecturers with different versions of answers originally written in English and Dutch both as L2, by German students. The versions varied in L2 CEFR levels. We investigated whether the versions would be assessed differently and/or would influence the grades. We found no effect of (a) the condition “marker reads in their first language versus in their second language” when assessing whether an exam answer met content criteria, or (b) differences in CEFR-based proficiency (lexical and grammatical accuracy). When grading written answers to an exam question in an EMI high education setting, lecturers weigh content criteria and are not distracted by poor formulations, due to students’ lower proficiency in English. This outcome points out that ELF is an acceptable communicative variety in an EMI context.