Effect of classroom assessment stakes on English language learners’ oral performance

Abstract

High‐stakes tasks may prevent language learners from performing successfully due to factors of “communicative stress” (Skehan, 1996, p. 52). Little research, however, has directly examined phonological patterns in learners’ performances on low‐stakes, medium‐stakes, and high‐stakes classroom tasks. The current study investigated the oral performance of 37 intensive English program students on a low‐stakes classroom presentation, a medium‐stakes classroom achievement assessment, and a high‐stakes end‐of‐term exit assessment. Suprasegmental measures of rate, fluency, prominence, and pitch were analyzed instrumentally. The researchers conducted mixed factorial analyses of variance and regression analyses to determine the effect of the stakes of the task on speech performance and scores. Findings reveal that students gave prominence to more words, had a decreased mean length of run, and had a narrower pitch range on the high‐stakes exit assessment. Furthermore, frequent phonological features significantly predicted task scores. The researchers discuss implications with regard to how teachers can implement a variety of tasks of different stakes in the classroom to allow opportunities for learners to improve their speech performance inside and outside of the classroom.