Peer Assessment by Groups of Group Work

Group work can be used to encourage deeper learning, promote student autonomy by transferring some of the responsibility for teaching and learning to students and simultaneously reduce academic time in feedback and marking. While these aims can also be achieved with peer assessment of student output (Williams, 1992; Hughes Somervell, 1993), its usage is often limited to assessing the contribution or input of each member to a group project (Goldfinch, 1994). This paper reports the results of a peer assessment experiment with 210 final-year undergraduate business students. Students formed 41 groups or teams, whose membership was constant for the duration of the subject, to complete two of the four assessable tasks. One of the team tasks was an oral class presentation worth 25% of a student's overall grade. Teams of peers rated the quality of the presentation, using a 22-point guide, in terms of content and presentation. Their mark was compared with a staff score also rating the presentation as a team. While there is no significant difference in the student and staff averages, there is a significant difference in the standard deviations which suggests that the two populations are different. However students scores were significantly closer to staff scores in the second half of the semester at least in terms of the presentation component. Although the peer-assessed marks are not significantly different from those assigned by staff, the correlation between them is only moderate.