Washback effects of speaking assessment of teaching English in Sri Lankan schools

Abstract

Washback is a concept commonly used in applied linguistics to refer to the influence of testing on teaching and learning. The purpose of this study was to investigate the washback effect of a new system of English language speaking assessment in Sri Lanka. The new assessment was introduced with the intention of promoting the teaching and learning of English speaking skills in schools as part of a Presidential educational initiative called the English as a Life Skill Programme. The study examined the washback effect of the introduction of speaking assessments at both National and school levels from the perspectives of participants at three levels of the education system: the decision making level, intervening level (teacher trainers and in-service advisors), and implementing level (teachers and students). For this purpose, a mixed methods research approach was employed. Semi-structured interviews were conducted with participants at the decision making level and intervening level to examine whether there were any important gaps in translating policy intentions to the implementing level participants (teachers and students). A questionnaire survey was conducted with teachers and students to investigate their perceptions of the assessment change and its effects on teaching and learning speaking in the classroom. Classroom observations were conducted to gain insights into actual classroom practices in relation to teaching and learning speaking, along with follow-up interviews to seek teachers’ accounts of their classroom practices. The study found that the assessment change did influence teachers’ and students’ perceptions of teaching and learning speaking in the classroom, as well as teachers’ instructional practices. Therefore, some of the policymakers’ intended aims were achieved. However, the intensity and direction of washback were shown to be influenced by several mediating factors such as teachers’ training and contextual factors such as the availability of classroom resources. The findings of this study suggest that assessment reforms can be used to promote change both in what is taught in the classroom and how it is taught, but to different degrees. The study indicated that washback does occur in this context, but it operates in a complex manner associated with many other variables besides the assessment itself. The findings of this study have implications for the improvement of future assessment policies in Sri Lanka, highlighting the importance of timely implementation of reforms and of monitoring them. The findings suggest that it is especially important to listen to key stakeholders’ (teachers’ and students’) voices in the initial planning and feasibility study phases of reform.