Effect of three network visualizations on students' social‐cognitive engagement in online discussions

Abstract Social learning analytics (SLA) tools are designed to visually demonstrate online discussions with a goal to foster student engagement. However, empirical studies indicate controversial results of the effect of SLA tools on student engagement. This design‐based research designs a student‐facing SLA tool to demonstrate discussions from three perspectives and further uses mixed methods to investigate the effects of the tool. Results indicate the tools have positive influences on increasing student social‐cognitive engagement. The social network visualization has positive influences on both socially active and inactive students; it particularly increases peripheral students' social engagement. The topic network visualization improves all students' perspective expressions, indicating that demonstrating students' interested topics may increase cognitive engagement. The cognitive network visualization triggers students' information sharing, which is considered as the beginning of online engagement development. Based on the results, this research proposes integrated implications by considering learning theory, pedagogical supports and tool development. Practitioner notes What is already known about this topic SLA tools are designed to turn learning data into actionable insights of student engagement. Network‐oriented, content‐oriented and process‐oriented analytics are three major approaches. Extant empirical studies indicate controversial results of the effect of SLA tools on student engagement. What this paper adds This research designs a SLA tool to demonstrate student engagement via three network visualizations. This research uses mixed methods to examine students' social‐cognitive engagement. Results show that the designed SLA tools have positive influences on increasing students' engagement. Implications for practice and/or policy Design of SLA tools needs to consider learner agency, disposition and characteristic in the local context. Instructors should deliberately guide students reflect on the information demonstrated in the SLA tool. Future SLA design should make integrate multiple analytic results to reflect engagement in varied ways.