Accuracy of students' self‐assessment and their beliefs about its utility

The purpose of the two studies presented here was to evaluate the accuracy of students' self‐assessment ability, to examine whether this ability improves over time and to investigate whether self‐assessment is more accurate if students believe that it contributes to improving learning. To that end, the accuracy of the self‐assessments of 3588 first‐year students enrolled in a post‐secondary institution was studied throughout a semester during which each student made approximately 80 self‐assessments about his or her own learning process. These self‐assessments were then compared with multiple judgements by peers and tutors. The overall correlations between the scores of self‐, peer and tutor assessments suggest weak to moderate accuracy of student self‐assessment ability. The findings also reveal an ability effect; students judged as more academically competent were able to self‐assess with higher accuracy than their less competent peers. Comparing the accuracy of student self‐assessment averaged over four consecutive periods indicates that the accuracy does not improve over time. In a second study, a questionnaire aimed at eliciting student's beliefs about the effects of self‐assessment on their learning was administered to 936 first‐year students. Based on their responses, sub‐groups of students were identified: those who either believed in the usefulness of self‐assessment or did not. Results suggest that there is no significant association between student beliefs about the utility of self‐assessment and the accuracy of their self‐assessments.