Abstract
In recent years, peer assessment has become of particular interest in educational assessment with the increasing emphasis on active learning. Although a considerable number of studies have shown that peer assessment is a good pedagogical method for teaching language, there is little agreement as to if it can be used to supplement teacher assessment as a formal assessment; hence, this study explores one way to lead toward more sophisticated peer assessment. In two English classes, university students gave two presentations: the first presentation (FP) and the second presentation (SP). In the experimental group (EG), after FP, peer assessments were analyzed using the FACETS computer program, and each student received comments from their teacher about the quality of their peer assessment. Meanwhile, in the control group (CG), students did not obtain any comments between FP and SP. This study concerned how those comments affected peer assessments of SP. The results indicated that, in EG, unexpected responses dramatically decreased and the wider variety of score categories were utilized in SP. Moreover, the bias interactions between raters and items of the EG also significantly decreased. Overall, the study indicated the effectiveness of indicating how students had undertaken their previous peer assessment.
