Learning Analytics in Supporting Student Agency: A Systematic Review

Student agency, or agency for learning, refers to an individual's ability to act and cause changes during the learning process. Recently, learning analytics (LA) has demonstrated its potential in promoting agency, as it enables students to take an active role in their learning process and supports the development of their self-regulatory skills. Despite the growing interest and potential for supporting student agency, there have yet to be any studies reviewing the extant works dealing with the use of LA in supporting student agency. We systematically reviewed the existing related works in eight major international databases and identified 15 articles. Analysis of these articles revealed that most of the studies aimed to investigate student or educators' agency experiences, propose design principles for LA, and to a lesser extent, develop LA methods/dashboards to support agency. Of those studies developing LA, none initially explored student agency experiences and then utilized their findings to develop evidence-based LA methods and dashboards for supporting student agency. Moreover, we found that the included articles largely rely on descriptive and diagnostic analytics, paying less attention to predictive analytics and completely overlooking the potential of prescriptive learning analytics in supporting agency. Our findings also shed light on nine key design elements for effective LA support of student agency, including customization, decision-making support, consideration of transparency and privacy, and facilitation of co-design. Surprisingly, we found that no studies have considered the use of LA to support student agency in K-12 education, while higher education has been the focal point of the LA community. Finally, we highlighted the fields of study and data visualization types that the studies mostly targeted and, more importantly, identified eight crucial challenges facing LA in its support of student agency.