What Do Educational Data, Generated by an Online Platform, Tell Us About Reciprocal Web-Based Peer Assessment?

Abstract

Peer Assessment (PA) is a promising evaluation strategy in the educational context, not only due to its effectiveness to reduce instructor’s evaluation loading, but mainly due to its benefit towards student development e.g., teamwork, in-depth thinking. In this exploratory study we sought to explore how do educational data, as generated by an online platform (i.e., Peergrade) and displayed in teacher’s and students’ Learning Analytics Dashboard (LAD), can potentially inform us of the PA process and the peer interactions, as they take place. Participants in the study were 21 undergraduate teacher-students who attended a science course (electrical circuits topic) following the inquiry-based approach. Students were asked to reciprocally and individually assess the responses of a peer in a given task. The findings of this study have implications towards the establishment of new theoretical frameworks and developments for bridging educational theory, design process and data science, in the field of assessment.