Dusting for fingerprints: Tracking online student engagement

Hybrid learning strategies blend face-to-face instruction with online components, using Learning Management Systems (LMSs) as key platforms for educational resources. In strategies like the flipped classroom, students need to follow a specific learning pathway to complete certain activities in the L...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Computers and education. Artificial intelligence 2024-06, Vol.6, p.100232, Article 100232
Hauptverfasser: Armas-Cervantes, Abel, Abedin, Ehsan, Taymouri, Farbod
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Hybrid learning strategies blend face-to-face instruction with online components, using Learning Management Systems (LMSs) as key platforms for educational resources. In strategies like the flipped classroom, students need to follow a specific learning pathway to complete certain activities in the LMS before class. This includes watching videos and completing readings and quizzes, to prepare for hands-on exercises during classroom time. Consistent student engagement with this approach is vital for success – but in large subjects with hundreds of enrolments, monitoring that engagement is a complex task. Using the data collected in an LMS, this paper presents an approach for detecting significant changes in student engagement in hybrid learning environments. The approach uses Process Mining (PM), a family of tools and techniques to analyze data through a process lens, to compare students' learning pathways between pairs of learning windows (e.g., weeks in a semester). Using a real-life event log containing more than 26,000 interactions of 194 students over a full semester, the findings demonstrate the approach's ability to detect changes in student engagement over time. These insights can be used by educators to refine their instructional design, deliver targeted interventions, and ultimately improve the overall effectiveness of hybrid learning. •Process mining can be used to discover student learning pathways in hybrid learning environments.•Variants analysis can be used to compare pairs of learning windows to detect differences in student engagement.•Mutual fingerprints can detect significant differences between learning windows, even with thousands of student actions.
ISSN:2666-920X
2666-920X
DOI:10.1016/j.caeai.2024.100232