The global rampancy of COVID-19 has caused profound changes in education sectors. Perhaps the most salient change is the shift of the instructional paradigm from face-to-face instruction to fully online learning. To address the challenges facing the education sector, researchers and educational practitioners have extensively investigated the transition in teaching mode under COVID-19, with a growing contribution to a range of topics in relation to online learning. Against this backdrop, it is necessary to gain a comprehensive understanding of the major hotspots and issues of online learning so as to develop appropriate and effective policies on strategic (re-)allocation of resources to more critical initiatives. This study aims to adopt bibliometrics and topic modeling to identify prominent research topics on online learning under COVID-19 from the large-scale, unstructured text of research publications. Specifically, structural topic modeling will be used to identify predominant topics concerned by scholars working in the field of online learning research. The non-parametrical Mann-Kendell trend test will also be applied to uncover the developmental tendency of each identified topic. In addition, the correlations among the key topics will be revealed and visualized by hierarchical clustering analysis. Based on the analytical results, suggestions will be made to facilitate educational policy formulation to promote the development and effective implementation of technological, scientific, and pedagogical activities of online learning. Copyright © 2022 IEEE.
|Title of host publication||Proceedings of the 2022 IEEE International Conference on Behavioural and Social Computing (BESC-2022)|
|Place of Publication||USA|
|Publication status||Published - 2022|