An Empirical Investigation Into The Washback Of Language Testing On Language Learning

Washback refers to the influence of language testing on language teaching and learning. To date most washback studies have been focused on large-scale standardized language tests (e.g. TOEFL, IELTS, and CET). Besides, few of them have explored whether a language test may produce differential washback effects on different groups of test takers.The present study investigated the washback of the Fudan English Test (FET), a university-based English proficiency test, on students’English learning, and explored whether there were any differences in students’reported washback that were related to their gender, academic background and English proficiency level. This study addressed the following three research questions: 1) What are students’ perceptions and views of the FET? 2) What is the washback of the FET on students’ English learning in terms of learning attitudes, materials, motivation, and strategies? 3) Are there any differences in students’ reported washback on their learning that are related to their gender, academic background, and English proficiency level? Questionnaire survey was employed as the main research instrument, alongside with follow-up interviews. Altogether 335 students participated in this study. Various statistical analyses were performed, including exploratory factor analysis (EFA), descriptive statistics, t-tests, one-way ANOVAs, etc. In addition, qualitative interview data were transcribed and coded to shed additional light on the three research questions.EFA extracted one factor representing students’ perceptions of the FET and three factors representing the impact of the FET on students’ learning. Descriptive statistics indicated that students held the most positive perceptions of the test design and least positive of the computer-based speaking test. Besides, the FET had most affected students’ learning strategies and least affected their learning attitudes. ANOVAs indicated that while gender and academic background did not affect students’ perceptions of the FET and reported washback on their English learning, English proficiency level did affect students’ perceptions of the FET. Qualitative interview data indicated that the washback of the FET was more positive than negative. The findings of this study demonstrate that differential washback does exist, but only to a limted degree. Besides, this study has significant implications for the future development of the FET.