Preliminary Exploration of Learning Analytics in Virtual Reality: A Space-Themed Approach

Abstract

In this study, we present an end-to-end implementation of learning analytics for a virtual reality (VR) learning environment. The proposed solution includes a VR environment featuring a virtual International Space Station, an Experience API (xAPI) schema based on Actor–Verb–Object triples, a Learning Record Store (LRS), and a dashboard. The current dashboard presents summary metrics, task completion, accuracy, and the distribution of logged action verbs, providing an initial view of learner participation, task progression, and spatial exploration in the VR environment. To demonstrate feasibility, we conducted a workshop with 28 elementary school students using a scaffolded learning activity, during which 1,052 behavioral statements were collected. This preliminary work also revealed several technical challenges, including (1) device-sharing constraints, (2) verb–object semantic ambiguity, and (3) multi-level data granularity. Future studies will extend the analytical depth of the dashboard and explore sequence-based approaches for better understanding learners’ behavior in VR environments.