Assessing speaking skills in English: an approach to test and rating scale design in a university context

Abstract

Assessment is a key feature and frequently the culminating point of the majority of foreign language teaching programmes, therefore it seems obvious that its analysis and validation should constitute an important and interesting area in the field of Second Language Acquisition research. As test users and designers, we need to be able to justify the implementation oftests and the validity oftheir scores since we have a responsibility in responding both to the challenges of defining language constructs and to developing validity arguments that we can apply to testing practice. The tests we use need to be appropriate to the context in which they are used and provide scores that are as accurate a reflection as possible of learners' linguistic performance, underlying language ability, and of the principies behind the approach and content ofthe teaching and learning programme.