The Effects of Training in Peer Assessment on University Students\u27 Writing Performance and Peer Assessment Quality in an Online Environment

This study was designed to examine the effects of peer-assessment skill training on students\u27 writing performance, the quality of students\u27 feedback, the quality (validity and reliability) of student-generated scores, and the students\u27 satisfaction with the peer assessment method in an online environment. A quasi-experimental design was employed to test group differences on the dependent variables. Four hundred and seventy-three sophomore and junior undergraduate students who were enrolled in a Foundations of Education course were selected by convenience sampling at a Large East-Coast Urban University. Students enrolled in Spring and Fall semesters of 2008 were assigned to the two experimental groups that received principle-based peer-assessment skill training or target-criteria-based peer-assessment skill training, while students enrolled in Fall semester of 2007 were assigned to the comparison group and did not receive structured peer-assessment skill training. The results of the study indicated that students who had peer assessment skill training in the experimental groups outperformed their counterparts in the comparison group on writing performance, and provided higher quality written feedback to their peers than those in the comparison group. The findings revealed that students in experimental groups generated more reliable assessment scores than those in the comparison group in the second round of peer assessment. The findings also revealed that students in the target-criteria-based training groups exhibited a higher level of satisfaction with peer feedback than those in the other groups. In addition, the results indicated that use of the target-criteria-based training method had no apparent superiority to use of principle-based training method on students\u27 writing performance, and peer-assessment skill training had no apparent positive impacts on the validity of student-generated assessment scores during peer assessment