‘Test in Swedish for Universities and Higher Education (TISUS)’: On the use of conversation as an oral proficiency test situation
The current paper presents results from an investigation of interaction between the examiners of TISUS (Test in Swedish for Universities and Higher Education), and test takers during their oral proficiency assessment of Swedish as a second language. More specifically, the study deals with the interactional procedure of the test, which is usually carried out in a group of two or three test takers. Even though the test takers are informed beforehand about the topics discussed during the exam, within such tests, various questions and topics are also exchanged that may change the trajectory of the talk and impact the test takers’ performances as well. The encouragement of keeping the conversation natural between the test takers may be constrained by the formal nature of the test, i.e. the distribution of topics and turns by the examiners. Furthermore, due to the contingencies of talk-in-interaction and co-construction of the context, participants are faced with many challenges. In this presentation, we will demonstrate, by means of conversation analysis, how participants orient to problems related to the fact that the conversation is neither totally natural, nor is it open-ended. The test takers’ preparation for thematic talk may, for example, lead to ‘mediated turn distributions’, which in turn, negatively affects the performances and also the administration of the test situation.
The results show that despite the possible benefits of having conversation as a test situation, problems arise in interaction; thus, assessing the test takers’ performance, such interactional context and its construction may be taken into account.
EARLI 2015 (16th biennial conference of European Association for Research in Learning and Instruction), Limassol, Cyprus, 25-29 August 2015