Easy Java/JavaScript Simulations as a tool for Learning Analytics

Abstract: In this paper we introduce the new and planned features of Easy Java/JavaScript Simulations (EJS) to support Learning Analytics (LA) and Educational Data Mining (EDM) research and practice in the use of simulations for the teaching and self-learning of natural sciences and engineering. Simulations created with EJS can now be easily embedded in a popular Learning Management System using a new plug-in that allows creation of full-fledged instructional units that also collect and record fine-grained, instructional-savvy data of the students interaction with the simulation. The resulting data can then be mined to obtain information about students performance, behaviors, or learning procedures with the intention to support student learning, provide instructors with timely information about student performance, and also help optimize the pedagogic design of the simulations themselves. We describe the current development and architecture, as well as future directions for testing and extending the current capabilities of EJS as a modelling and authoring tool to support LA and EDM research and practice in the use of simulations for teaching science, in particular in the context of the increasingly popular online learning platforms.