Accuracy of Peer Assessments Performed by Elementary Physical Education Students

Background/Purpose: The importance of peer assessments within physical education has been increasingly recognized and endorsed within the international physical education community (Hay & Penny, 2009; Johnson, 2004). Despite the recommendation for the increased use of peer assessments in physical education, a lack of research exists that examines the accuracy with which elementary students can perform peer process assessments. Therefore, the purpose of this study was to examine the degree of accuracy with which elementary physical education students performed a peer process assessment of an overhand throwing task. 

Method: 28 elementary students, in first through third grade, were asked to conduct an assessment of their peers’ performance of two skill components (i.e., ‘side to target’ and ‘step with opposite foot’) during an overhand throwing task. Students were placed in pairs; the first participant performed five trials of the task while the partner assessed each trial. After five trials, the partners switched roles so that each participant had the opportunity to complete the throwing task as well as the assessment. Sessions were conducted once a week for a total of 4 weeks. All trials were video recorded; videos were later viewed by the researchers who conducted an assessment of each trial to be used as the criterion for comparison. The participants’ assessments were then compared to the researchers’ and evaluated for accuracy.

Analysis/Results: A one-way ANOVA revealed no significant difference in participants’ assessment accuracy across weeks (F(3,142) = 0.392, p = .759), indicating the participants’ accuracy did not change over time. There were, however, significant differences between grades overall (F(2,143) = 9.182, p < .001). Post-hoc analyses showed these differences were between 3^rd (M = 91.77 %, SD = 13.97) and 1^st (M = 78.95, SD = 22.76) grades (p = .005), and between 3^rd and 2^nd (M = 77.17, SD = 22.28) grades (p < .001). 

Conclusions: Despite older students performing the assessment significantly better than their younger counterparts, results still indicated that students as young as first grade can still perform a peer process assessment with a relatively high degree of accuracy (i.e., nearly 80% accurate). These results indicate that physical education teachers may be able to implement peer process assessments with lower elementary students with the assumption that they will be completed accurately.