A new study by Liubov Tupikina and Marc Santolini, Fellows at CRI Research, and their collaborator Sasha Poquet (University of South Australia and National University of Singapore), was accepted for publication and is available here! This work has been accepted in the Proceedings of the 10th International Conference on Learning Analytics and Knowledge (LAK ’20), March 23–27, 2020, Frankfurt, Germany (https://lak20.solaresearch.org/) LAK is very selective conference at the core of the learning analytics community, aimed at deriving behavioral and performance metrics of individual learners from digital data.
Learning analytics (LA) strives to improve learner experiences using insights from digital learner traces. Capturing learner interactions as network representations then helps visualize and analyze social dynamics in online course forums, making them a popular element in learning analytics dashboards.
In this paper, authors explore how online forum networks used by students during classes can be used as tools to construct social networks from which we can derive insights about collaborative learning. In this construction, the inferred social networks carries properties from the forum network, so that a high degree can merely be reflecting a high posting activity. The paper presents a methodology to disentangle forum activity from social connectivity through several generative models. The authors build a methodological framework to robustly infer such social networks from online forum data while controlling for the natural diversity of classroom setups. This work is relevant to scientists interested in social interactions and learner networks in digital learning, and more generally to researchers interested in deriving informative social network models from online forums.
The graphs included here represent visualisation of social networks obtained from the online forum data of classrooms in Australia.
