Developing a Topic Analysis Instant Feedback System to facilitate asynchronous online discussion effectiveness

Abstract

Asynchronous online discussion is one of the widely used collaborative learning activities in digital learning that can facilitate learners to think of controversial problems, share ideas, ask questions, and give feedback without time and space constraints. However, if there is no appropriate and effective mechanism that can promote asynchronous online discussion performance, it may cause the contents of the discussion away from the discussion topic, too narrow, or not deep enough. Therefore, this research presents a Topic Analysis Instant Feedback System (TAIFS) that can explore and visualize the topics according to the contents of an asynchronous online discussion from the learners of different groups. The TAIFS was developed and integrated into Modular Object-Oriented Dynamic Learning Environment (Moodle) which is a free and open source digital learning platform that has been widely used around the world based on Latent Dirichlet Allocation (LDA) to help learners grasp the change of overall discussion topics in real-time and see the similarities and differences of the discussion among different groups, thereby effectively improving their asynchronous online discussion effectiveness. A quasi-experimental research method was adopted to examine the effects of the proposed TAIFS-assisted online discussion on learners' socio-scientific reasoning performance including the complexity and multiple perspectives. A total of 61 students from two classes in a high school in Taipei city, Taiwan were recruited as the research participants. One class of 31 students was assigned to the experimental group using the TAIFS-assisted discussion board for discussing the socio-scientific issue (SSI) entitled “Utilization of the Coastal Area,” while the remaining 30 students from another class were assigned to the control group using a general Moodle discussion board. Analytical results show that the experimental group was significantly better than the control group in the overall discussion performance and the complexity and multiple perspectives. Besides, TAIFS is more helpful to improve the overall discussion performance and multiple perspectives of female learners than male learners. The contribution of this study is to propose an innovative and effective TAIFS that can promote learners’ asynchronous online discussion effectiveness to bring the asynchronous online discussion into new ground.