Investigating effects of considering mobile and desktop learning data on predictive power of learning management system (LMS) features on student success

Varshita Sher, Marek Hatala, Dragan Gaševic

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review


The research area of analyzing log file trace data to build academic performance prediction models has tremendous potential for pedagogical support. Currently, these learner models are developed from logs that are composed of one intermixed stream of data, treated in the same manner regardless of which platform (mobile, desktops) the data came from. In this paper, we designed a correlational study using log data from two offerings of a blended course to investigate the effects of the variables, derived from the use of varying platforms, on the prediction of students' academic success. Given that learners use a combination of devices when engaging in learning activities, it is apparent that weighing the logs based on the platform they originate from might generate different (possibly better) models, with varying priority assigned to different model features. For instance, our results show that the overall frequency of course material access is a less powerful indicator of academic performance compared to the frequency of course material access 'from mobile devices', probably due to the benefits associated with ubiquitous any-time access available to mobile learners. Thus, the primary goal of this study is to bring to light the potential for improvement of prediction power of models after considering the learner's platform of access, within the learning analytics community and the fields of user modeling and recommender systems, in general.

Original languageEnglish
Title of host publicationProceedings of the 12th International Conference on Educational Data Mining
EditorsCollin F. Lynch, Agathe Merceron, Michel Desmarais, Roger Nkambou
Place of PublicationMontreal Canada
PublisherPresses De I Universite Du Quebec
Number of pages4
ISBN (Electronic)9781733673600
Publication statusPublished - 2019
EventEducational Data Mining 2019 - Montreal , Canada
Duration: 2 Jul 20195 Jul 2019
Conference number: 12th


ConferenceEducational Data Mining 2019
Abbreviated titleEDM 2019
Internet address


  • Learner models
  • Learning analytics
  • Learning success
  • Mobile learning

Cite this