Posted in January 2024

Abstract of pitch presentation of the TU Delft workgroup that won first place with a process-mining case study in the Leadership Challenge with Data Analytics (LCDA) organized by Erasmus Centre for Data Analytics in collaboration with SURF.

Abstract

Online evaluation surveys increasingly suffer from declining and selective response rates. Hence, MOOC Learning developers/ Lecturers have limited and insufficient insights to enable fact-based course improvement. In order to provide more accurate insights in actual learner behaviour we performed pattern analysis on event-log data in two TU Delft Extension School MOOCs. We applied Process Mining as an explorative method combined with clustering techniques to compare intended vs. actual paths followed in the MOOCs. We find that higher performers show patterns of an iterative learning strategy compared to more linear learning paths of lower performers and non-passing learners. This corroborates the theories of Self-regulated Learning and Metacognition. Implications for data driven course improvement such as learning path analysis and other applications for Process Mining are discussed for TU Delft.

Keywords

Learning Analytics, MOOCs, course evaluation, process mining, learning paths.

Reference

Gherghiceanu, A., van Huik, B., Hunte, Z., Vriend, A., de Vries, N. (2024) DATA DRIVEN COURSE IMPROVEMENT., Abstract of pitch presentation on team findings during the 2024 Leadership Challenge with Data Analytics (LCDA) organized by Erasmus Centre for Data Analytics in collaboration with SURF.

Interested?

You can contact us for more information on research-es@tudelft.nl

License

This is an Open Access abstract, distributed under the terms of the Creative Commons Attribution 4.0 International licence (https://creativecommons.org/licenses/by-nc-sa/4.0/), which permits reusers to distribute, remix, adapt, and build upon the material in any medium or format, so long as attribution is given to the creator, the purposes are non-commercial and distribution of remixed, adapted or build upon material should be released under the same license.