%0 Journal Article %1 Rizvi2019 %A Rizvi, Saman %A Rienties, Bart %A Rogaten, Jekaterina %A Kizilcec, René F. %D 2019 %J Journal of Computing in Higher Education %K MOOC actionresearch assessment educationaldatamining engagement intervention learninganalytics learningdesign %R 10.1007/s12528-019-09231-0 %T Investigating variation in learning processes in a FutureLearn MOOC %U https://doi.org/10.1007/s12528-019-09231-0 %X Studies on engagement and learning design in Massive Open Online Courses (MOOCs) have laid the groundwork for understanding how people learn in this relatively new type of informal learning environment. To advance our understanding of how people learn in MOOCs, we investigate the intersection between learning design and the temporal process of engagement in the course. This study investigates the detailed processes of engagement using educational process mining in a FutureLearn science course (Nþinspace=þinspace2086 learners) and applying an established taxonomy of learning design to classify learning activities. The analyses were performed on three groups of learners categorised based upon their clicking behaviour. The process-mining results show at least one dominant pathway in each of the three groups, though multiple popular additional pathways were identified within each group. All three groups remained interested and engaged in the various learning and assessment activities. The findings from this study suggest that in the analysis of voluminous MOOC data there is value in first clustering learners and then investigating detailed progressions within each cluster that take the order and type of learning activities into account. The approach is promising because it provides insight into variation in behavioural sequences based on learners' intentions for earning a course certificate. These insights can inform the targeting of analytics-based interventions to support learners and inform MOOC designers about adapting learning activities to different groups of learners based on their goals.

@article{Rizvi2019, abstract = {Studies on engagement and learning design in Massive Open Online Courses (MOOCs) have laid the groundwork for understanding how people learn in this relatively new type of informal learning environment. To advance our understanding of how people learn in MOOCs, we investigate the intersection between learning design and the temporal process of engagement in the course. This study investigates the detailed processes of engagement using educational process mining in a FutureLearn science course (N{\thinspace}={\thinspace}2086 learners) and applying an established taxonomy of learning design to classify learning activities. The analyses were performed on three groups of learners categorised based upon their clicking behaviour. The process-mining results show at least one dominant pathway in each of the three groups, though multiple popular additional pathways were identified within each group. All three groups remained interested and engaged in the various learning and assessment activities. The findings from this study suggest that in the analysis of voluminous MOOC data there is value in first clustering learners and then investigating detailed progressions within each cluster that take the order and type of learning activities into account. The approach is promising because it provides insight into variation in behavioural sequences based on learners' intentions for earning a course certificate. These insights can inform the targeting of analytics-based interventions to support learners and inform MOOC designers about adapting learning activities to different groups of learners based on their goals.}, added-at = {2019-06-23T10:34:42.000+0200}, author = {Rizvi, Saman and Rienties, Bart and Rogaten, Jekaterina and Kizilcec, Ren{\'e} F.}, biburl = {https://www.bibsonomy.org/bibtex/2a414425205fe1762ac9cbd65f755452f/ereidt}, day = 14, doi = {10.1007/s12528-019-09231-0}, interhash = {82915f84535575e1fca48a108644351a}, intrahash = {a414425205fe1762ac9cbd65f755452f}, issn = {1867-1233}, journal = {Journal of Computing in Higher Education}, keywords = {MOOC actionresearch assessment educationaldatamining engagement intervention learninganalytics learningdesign}, month = jun, timestamp = {2019-06-23T10:34:42.000+0200}, title = {Investigating variation in learning processes in a FutureLearn MOOC}, url = {https://doi.org/10.1007/s12528-019-09231-0}, year = 2019 }