Self-assessment first or peer-assessment first: effects of video-based formative practice on learners’ English public speaking anxiety and performance

Although formative assessment has been recognized as an effective way for improving learning, scant attention has been paid to the specific design on the sequence of applying formative assessment practice in computer-assisted language learning (CALL). Even less emphasis has been devoted to the cognitive and affective outcomes of different orders of formative assessment strategies on building students' public speaking competence. This mixed-method research employed multi-source data to explore and explain the effects of two different ways of sequencing video-based formative assessment on English language learners' public speaking anxiety and their performance. Two intact classes in a 16-week English public speaking course were randomly assigned to two groups, 25 in the self-assessment-initiated (SAI) group and 26 in the peer-assessment-initiated (PAI) group. Following two different sequences of formative practice, both groups were required to complete three formal English public speeches. Both quantitative and qualitative findings indicated that learners in the SAI group showed significantly lower public speaking anxiety than those in the PAI group, whereas learners in the PAI group showed significantly better public speaking performance. Moreover, the findings showed that SAI formative practice helped the learners to pay more attention to their delivery and anxiety levels in public speaking, while PAI formative practice may have led the learners to improve their proper use of English language. This research indicated that the sequence of implementing different modalities of assessment matters. Supported by video technology and CALL tools, different sequences in obtaining varying sources of feedback can affect students' public speaking anxiety and competence. This study provides refined understandings about the sequences of using video technology in formative practice for alleviating the learners' public speaking anxiety and improving their performance. The findings suggest that self-assessment should be arranged first for learners with a higher-level of anxiety while peer-assessment should be conducted first for learners with a lower-level of English language proficiency.Supplemental data for this article is available online at https://doi.org/10.1080/09588221.2021.1946562 .