Please use this identifier to cite or link to this item: https://repository.isls.org//handle/1/1640
Title: Longitudinal Analysis and Visualization of Participation in Online Courses Powered by Cohesion Network Analysis
Authors: Sirbu, Maria-Dorinela
Dascalu, Mihai
Crossley, Scott
McNamara, Danielle
Trausan-Matu, Stefan
Issue Date: Jun-2019
Publisher: International Society of the Learning Sciences (ISLS)
Citation: Sirbu, M., Dascalu, M., Crossley, S., McNamara, D., & Trausan-Matu, S. (2019). Longitudinal Analysis and Visualization of Participation in Online Courses Powered by Cohesion Network Analysis. In Lund, K., Niccolai, G. P., Lavoué, E., Hmelo-Silver, C., Gweon, G., & Baker, M. (Eds.), A Wide Lens: Combining Embodied, Enactive, Extended, and Embedded Learning in Collaborative Settings, 13th International Conference on Computer Supported Collaborative Learning (CSCL) 2019, Volume 2 (pp. 640-643). Lyon, France: International Society of the Learning Sciences.
Abstract: Online learning environments are increasingly used by students and instructors. Cohesion Network Analysis (CNA) can be employed by instructors to analyze discourse structure in terms of cohesive links in order to model student participation and interactions in collaborative learning environments. This paper presents an extended longitudinal analysis together with corresponding visualizations of participation generated for an online math course, powered by CNA. Multiple interactive views centered on the evolution of participation and of interactions between students clustered into three layers are generated using the CNA indices provided by the ReaderBench framework. Two types of sociograms are used to show the interactions between learners in the two course weeks that exhibited extreme conditions, namely: a) week 6, when a dramatic decrease of participation was identified, and b) week 16, when the highest number of participants and contributions were logged. In addition to the views centered on participants, we introduce a heatmap depicting the evolution of keyword relevance over time, as well as a Chord diagram for visualizing concept maps based on semantic relatedness. Our analytics dashboard can be used by tutors to monitor students throughout the term and to better ascertain the correlation of course material, schedule, and deadlines with the participation of students, as well as their interactions among themselves and with the tutor.
URI: https://doi.dx.org/10.22318/cscl2019.640
https://repository.isls.org//handle/1/1640
Appears in Collections:CSCL 2019

Files in This Item:
File SizeFormat 
640-643.pdf895.41 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.