Self- and Peer Assessment of Oral Presentations: Investigating Correlations and Attitudes

Despite the proven advantages of alternative assessment including self- and peer assessment, there is still uncertainty about the accuracy of these techniques especially in Iran since little research has conducted in this area. To this aim, a total of 2261 evaluation sheets were collected to investigate 118 peer assessments and 48 self- assessments of oral presentations in comparison with their teacher assessments. The data were collected during four semesters in general English classes held in Sharif University of Technology. Students were supposed to make an oral presentation for at least twenty minutes on a topic of their interest, and were then rated by their teacher, themselves, and their peers based on ten pre-established assessment criteria. The number of peer assessments for each presentation was within the range of 8-27. In addition, in order to delve into the students' attitudes toward the new techniques, 64 students expressed their opinions on five-point Likert scale surveys before and after the implementation of self- and peer assessment. It was indicated that the results of the peer assessment were more consistent with those of teacher assessment and the resemblance grew when the number of peers was within the range of 16-19. On the other hand, it was observed that the students' attitudes significantly changed in a positive way toward both techniques. All in all, the findings of the present study suggested that peer assessment closely resembled teacher assessment while the students' attitude changed more positively toward self-assessment in comparison with peer assessment. The findings of this study could have implications for language teachers, material designers, and educational policy makers.