Title: Time to focus on the temporal dimension of learning: a learning analytics study of the temporal patterns of students' interactions and self-regulation

Authors: Mohammed Saqr; Jalal Nouri; Uno Fors

Addresses: Department of Computer and System Sciences (DSV), Stockholm University, Borgarfjordsgatan 12 PO Box 7003, SE-164 07, Kista, Sweden ' Department of Computer and System Sciences (DSV), Stockholm University, Borgarfjordsgatan 12 PO Box 7003, SE-164 07, Kista, Sweden ' Department of Computer and System Sciences (DSV), Stockholm University, Borgarfjordsgatan 12 PO Box 7003, SE-164 07, Kista, Sweden

Abstract: In this learning analytics study, we attempt to understand the role of temporality measures for the prediction of academic performance. The study included four online courses over a full-year duration. Temporality was studied on daily, weekly, course-wise and year-wise. Visualising the activities has highlighted certain patterns. On the week level, early participation was a consistent predictor of high achievement. This finding was consistent from course to course and during most periods of the year. On course level, high achievers were also likely to participate early and consistently. With a focus on temporal measures, we were able to predict high achievers with reasonable accuracy in each course. These findings highlight the idea that temporality dimension is a significant source of information about learning patterns and has the potential to inform educators about students' activities and to improve the accuracy and reproducibility of predicting students' performance.

Keywords: learning analytics; temporality; time; problem-based learning; collaborative learning; social network analysis; self-regulation.

DOI: 10.1504/IJTEL.2019.102549

International Journal of Technology Enhanced Learning, 2019 Vol.11 No.4, pp.398 - 412

Received: 11 May 2018
Accepted: 07 Jul 2018

Published online: 12 Apr 2019 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article