Peer and Self-Assessment for English Oral Performance: A Study of Reliability and Learning Benefits

This paper reports a study which investigated the reliability and potential benefits of peer and self-assessment for English oral training at the tertiary level. Participants were 40 second-year students taking English oral training at a national university in southern Taiwan. Data were collected from pre- and post-study questionnaires as well as student and teacher assessments of oral performance. The study found that peer and self-assessment was highly correlated with teacher assessment, though students' markings were more centrally distributed. Students' analytical scorings were likely to agree with the teacher's on the criteria of language and delivery most and manner least. After the practice, students were more certain about student participation in assessment and felt more comfortable about peer assessment, but the proposed ratio for student assessment was reduced 2.4%. Peer and self-assessment was perceived less objective than teacher assessment. In terms of learning effects, students claimed gains in English speaking skills, critical thinking, awareness of strengths and weaknesses in the assessed skills, and confidence in evaluation. Pedagogical implications and suggestions for future research are also provided.