Prompting in Pseudonymised Learning Analytics: Implementing Learner Centric Prompts in Legacy Systems with High Privacy Requirements

摘要

Learning analytics have become a well-considered aspect of modern e-learning environments. One opportunity of learning analytics is the use of learning process data enabling lecturers to analyse students' learning progression as well as to identify obstacles and risks. With this analytics knowledge, lecturers may want to scaffold students' learning activities to improve the learning progress and overcome obstacles or risks. Prompts are known to be a possible solution for such scaffolding mechanics. However, implementing prompts into existing legacy systems in learning environments with high data privacy concerns is quite a challenge. This research shows how a prompting application has been implemented into an existing university environment by adding a plugin to the local e-learning platform which injects user centric prompts to specific objects within their e-learning environment. The prompts are dynamically loaded from a separate learning analytics application which also collects the students' learning trails and progress. The system is evaluated in two courses in the fall semester 2017 with more than 400 students altogether. The system collects up to two thousand student events per day. An in-depth empirical investigation on how various prompts influence students' learning behaviours and outcomes is currently conducted.