5. Discussion and conclusion
This study implemented an online video peer assessment system to help nursing students to develop their professional communication skills with simulated psychiatric patients. In it, the peer assessment process consisted of two rounds. Each student received comments and feedback from three to four of their peers in each round. The analysis of peer feedback showed that students received numerous amounts of feedback from their peers (an average of 17.4 and 19.6 comments per student for the first round and the second round respectively). Such high interactions among learning peers were hardly likely to happen in traditional classroom settings, where class time is limited and the class size is large. Furthermore, this study found that online peer assessment could enhance nursing students' communication performances significantly. It was inconsistent with Kruijver et al. (2000)'s finding: the outcomes of many communication training programs are very limited. We believe this situation was due to the fact that in the process of peer assessment, the students gained a variety of feedback which had a positive influence on their learning. As described above, Gibbs and Simpson (2004) indicated such feedback should have several conditions, including being sufficient, focused, timely, appropriate, attended and acted. As shown in Section 4.3, the feedback in the present study was not only sufficient in frequency and detail, but also focused on peers' performance and their learning. As described in Section 4.4, most students read the peers' feedback and tried to make some changes in order to improve their future performance. It indicated that students could accept their peers' suggestions and act upon them. This means that the feedback had a positive impact on students' learning. It also led students to make changes when they went onto their next assignment or when tackling any future tasks. Apart from rating schemes (ICAS), peer feedback played an important role in improving students' communication skills in this study. This result is consistent with Xiao and Lucking (2008)'s finding. They compared the effects of two peer assessment methods on university students' academic writing performance and their satisfaction with peer assessment. Their results indicated that students in the experimental group (using rating-plus-qualitative-feedback) demonstrated greater improvement in their writing than those in the comparison group (using rating-feedback-only). Xiao and Lucking also found that students in the experimental group exhibited higher levels of satisfaction with the peer assessment method both in peer assessment structure and peer feedback than those in the comparison group.