IMPLEMENTING LEARNING ANALYTICS IN A MASSIVE OPEN ONLINE COURSE FOR LANGUAGE LEARNING

In this paper we will describe the steps needed for the implementation of learning analytics tools in a massive open online course (MOOC) based on learning management system MOODLE and intended for language learning. MOOC in question is intended for learning of the Croatian language at the Common European Framework of Reference for Languages level A1, and it was started in 2017 in cooperation of the Faculty of Humanities and Social Sciences University of Zagreb and the Central State Office for Croats Abroad. It has over 4400 participants (November 2019) and the interface is in English and Spanish language. Since learning analytics tools have been introduced to MOODLE, that set of tools will be used for the analysis. Data collected by the learning management system will be combined with the information gathered through the MATOMO (ex. PIWIK) web analytics platform. We will also present the initial results, as well as predictive models that were tested (both pre-made and custom). Implementation will include stakeholders from both institutions: management, course teachers and mentors, digital content creators and instructional designers, and IT support personnel. The use of learning analytics tools is intended for the evaluation, analysis and prediction of the participant drop-out, as well future redesign of the current MOOC structure / delivery.