Key points are not available for this paper at this time.
Language tests are frequently criticized for having negative impact on teaching — so-called 'negative washback'. Some writers believe that it is possible to bring about positive change in language teaching by changing tests. However, neither positive nor negative washback on classrooms has been established empirically by observation of classrooms. This article seeks to redress this situation by reporting on an innovative study of the impact of a new English examination in Sri Lanka on language teaching. Although impact is demonstrated on the content of teaching, no evidence was found for any influence of the test on how teachers taught. It is argued that studies of washback need to relate teachers' attitudes to and understanding of exams to observations of classrooms in order to understand why teachers teach the way they do, and why tests might not have the impact that is frequently asserted. It is concluded that the supposition of washback as currently formulated is an oversimplified account of the relationship between tests and teaching and it is suggested that the complexity of that relationship, and of curricular innovation more generally, needs further exploration.
Wall et al. (Mon,) studied this question.