MultiLA - An Authoring and Learning Analytics Tool for e-Learning Applications in Data Science Education

Interactive e-learning applications (eLAs) are proving to be a useful addition to teaching material for data science courses, as they allow to combine mathematical theory, interactive visualizations, coding and other exercises in a single environment. In this paper, we present our tool MultiLA, which comprises an authoring tool to build eLAs and a backend for data collection. The eLAs can track learner behavior from mouse clicks and pointer movements up to the success in completing exercises, all being stored in the backend. Using learning analytics learning behavior and success can be analysed aiming to improve eLAs and support learners. eLAs can easily be adapted to different learner groups by [de]selecting their individual learning blocks in the backend, also providing an easy way to run A/B tests for comparing different versions of an eLA.