Teaching to the test in the English language classroom: Development and validation of a measurement instrument

Abstract

In the last few decades, the assumption that the implementation of standardized tests might cause washback effects on language teaching and learning has become certainty. Researchers and practitioners report an increased use of ‘teaching to the test’ (TTT) practices, however, in current literature, consensus about which practices this includes is missing and operationalizations in studies vary greatly. Further, even though TTT is assumed to impair meaningful learning, no study so far examined students’ perspective. The current study therefore used a multi-method approach to theoretically and empirically define TTT and develop a uniform measurement instrument, focusing on students’ perspective of their English language teachers’ practices. Results suggest that TTT is a multidimensional construct consisting merely of test preparation practices, but not changes in teaching style. Multilevel analyses showed adequate fit for two equally plausible models, good psychometric properties, and full measurement invariance of the instrument across gender, first language, and school type. The theory-based and empirically supported instrument therefore represents a valid and reliable measure for uniformly assessing perceived TTT in the language learning context. Our study paves the way for future research studying the effects that this specific type of washback might have on students’ learning outcomes.