Collecting Experience Data from Remotely Hosted Learning Applications.

TL;DR: In this article, a learning system architecture is proposed to apply advanced Learning Analytics techniques on experience data collected from remotely hosted learning applications through a seamless integration between LTI and Tin Can API.

Abstract: The ability to integrate multiple learning applications from different organizations allows sharing resources and reducing costs in the deployment of learning systems. In this sense, Learning Tools Interoperability (LTI) is the main current leading technology for integrating learning applications with platforms like Learning Management Systems (LMS). On the other hand, the integration of learning applications also benefits from data collection, which allows learning systems to implement Learning Analytics (LA) processes. Tin Can API is a specification for learning technology that makes this possible. Both learning technologies, LTI and Tin Can API, are supported by nowadays LMS, either natively or through plugins. However, there is no seamless integration between these two technologies in order to provide learning systems with experience data from remotely hosted learning applications. Our proposal defines a learning system architecture ready to apply advanced LA techniques on experience data collected from remotely hosted learning applications through a seamless integration between LTI and Tin Can API. In order to validate our proposal, we have implemented a LRS proxy plug-in in Moodle that stores learning records in a SCORM Cloud LRS service, and a basic online lab based on Easy JavaScript Simulation (EjsS). Moreover, we have tested our implementation using resources located in three European universities.