We have commenced a new project in the area of learning analytics exploring how we can better understand how students engage in student discussion forums, with the aim of building models that describe and inform educator interventions. Early work in this new project has been accepted as a Work in Progress paper for the upcoming Learning @ Scale conference.
N. Falkner, C. Szabo and K. Falkner, Formal Forum Triage: Towards the Strategic Selection of Responses to Student Discussion Forums, Accepted for Learning @ Scale 2017.
Small classroom environments allow teachers to effectively monitor student understanding and levels of engagement by observing the ‘front of the classroom’. In larger class- rooms with hundreds of students, or for on-line courses with thousands, this degree and ease of monitoring is infeasible without the addition of automated assistance. It can be very difficult for an educator with limited resources to decide which queries require immediate attention when a high volume of questions arises on the discussion forum. This becomes increasingly complex as the educator aims to obtain the best outcome across all threads in a timely fashion. Existing approaches to automated forum analysis, including sentiment or social network analysis, provide a useful grouping of messages and identify common discussions, but require additional instructor attention towards effective intervention. Research has shown that the timing of messages relative to associated deadlines is a key indicator of priority, when compared to topic and structural analysis. In this pa- per, we propose the formal representation of events within a discussion forum to facilitate the definition of potential and existing intervention strategies. We extend forum events with information about teaching activities, such as assignment deadlines, and discuss subsequent intervention strategies, focusing on resource constrained environments such as large-scale discussion forums.