Examining Washback of Language Assessment within an Educational Context: Voices from Teachers and Examiners

Teaching enriches students’ learning contents, while assessment evaluates students’ learning results. In recent decades, assessment has gained increased attention in EFL education. One of the important issues in the practice of assessment is the washback effect of testing on teaching and learning (Alderson and Wall, 1993; Buck, 1988; Hughes, 2003). Washback effect on teaching and learning can be positive (beneficial) or negative (harmful). Much research has been focused on the washback effect of assessment on students’ learning or large-scale, standardized tests (Watanabe, 1996). Assessment has powerful influence on teaching, too. Wall (1998) claimed that high-stakes tests might induce the impact on teaching methodology and content.This study explored the washback effect (Alderson and Wall, 1993) of a high-stakes exam, TOEFL iBT. It focused on the teachers’ perceptions of the test and its washback effect on teaching. The purpose of this qualitative study was to describe the washback behaviors of teachers and, to a lesser extent, students in the high-stakes testing environment of TOEFL iBT. The study followed a group of five teachers, teaching TOEFL iBT preparation courses. To explore the washback phenomenon, this study employed various methodological techniques, including questionnaire surveys, classroom observations, and in-depth interviews. The results of this study have shown a multi-layered account of the washback phenomenon. Teachers’ different perceived levels of awareness of high-stakes exams and perceived students’ learning attitudes have a crucial influence on teachers’ perceptions of the impact of high-stakes exams on their curricular planning and instruction. However, several discrepant findings from this study further support that washback is quite context-oriented and complex. The results imply that simply examining one factor without a covariance analysis or examining the phenomenon in one context is not capable of explaining critical washback issues, such as how and why washback phenomenon influences some teachers but not others. It is recommended that longitudinal studies, such as long-term classroom observations, should be conducted in order to explain to what extent washback actually occurs to influence classroom teaching.