Testing L2 Vocabulary: current test formats in English as a L2 used at Swedish universities

Författare

Henrik Gyllstad

Summary, in English

Recent literature on L2 vocabulary testing points to a move away from discrete vocabulary testing towards more embedded and integrative approaches. This paper investigates if the way vocabulary is currently tested at English departments at Swedish universities follows this trend. The survey shows that a great majority of universities use discrete and selective vocabulary tests and that the multiple-choice format is the most widely used format. The paper takes a closer look at this format, in particular a 120-item version used by a great number of departments. A small-scale study of the results on this test, by university learners of English in their first, second and third term of study, is presented. The study shows that learners on the most advanced level of study scored significantly better than the two less advanced levels, but that no significant difference could be observed between scores of first term and second term full-time learners. It also shows that learners’ scores are highly scalable, lending validation support to the underlying assumption of the test design.

Avdelning/ar