Investigating shared agency in student Knowledge Building discourse using epistemic network analysis

In computer-supported collaborative learning, students engage in both synchronous online and face-to-face discussions to enhance engagement and learning from peers. Recent research highlights the potential of quantitative modelling techniques, such as those used in Epistemic Network Analysis, to deepen our understanding of student collaboration through discourse, including the concept of shared epistemic agency in Knowledge Building. However, these efforts are still in the early stages. This study aims to advance this research by analysing student contributions in synchronous discussions (online and verbal) within a Knowledge Building framework. Data was collected from a Grade 5 class during a 2.5-hour Social Studies lesson designed to foster real-world problem-solving and collaborative discussions. Students participated in discussions on Knowledge Forum (online) and in face-to-face small groups. Transcripts of group discussions and Knowledge Forum notes were coded for semantic types and analysed for weighted connections between coded contributions using Epistemic Network Analysis. The analysis of overall student contributions revealed distinct patterns of weighted connections between codes and varying degrees of idea development in online versus oral discussions. Additionally, the study identified different engagement patterns within groups based on their epistemic networks. The findings offer insights into student engagement during collaborative talk and have implications for future research and practice.