Learning Analytics and Big Data Regulation
E-learning offers huge possibilities for the delivery of affordable high quality education irrespective of temporal and geographical constraints, however, many students suffer from the high demand such learning makes on the student’s ability to self-regulate or co-regulate.
Empirical evidence shows that humans at all levels are unable to accurately evaluate their own learning, a key requirement for regulated learning, thus leading prominent researchers to predict that without external support a very large proportion of learners will fail to advance their skills effectively.
Without opportunities for intervention that the traditional classroom offers, e-learning students must trudge through courses generally on their own. The frustrations and low achievement of these students currently leads to high dropout rates.
In order to leverage the opportunities that the flexible format of e-learning offers of reaching currently disadvantaged students with high quality education, a means for remotely infusing regulation in learning tasks is highly desirable.
This research proposes an agent-facilitated regulation model to assist learners in their tasks. It will refine current understanding of self-regulation and co-regulation to include the interaction between a learner and an anthropomorphic pedagogical agent.
The general objective is to instil self-regulation and co-regulation in these agents so that they can participate not only in regulating the behaviour of learners but also to infuse regulation transparently into learning tasks so as to gradually improve the learner’s regulatory competencies.