Improving Students’ Self-awareness by Analyzing Course Discussion Forum Data

The growing demand for self-paced online learning (SPOL) courses lead to post-secondary institutions exploring how information technology can be used to improve the quality of SPOL courses by evaluating teaching and learning strategies, methods, activities, and the way students engage with their studies. One of the main barriers in SPOL is that students may feel isolated as they learn in an individualized mode and collaborative learning could be difficult to realize. The isolation may be lessened when students interact in a course discussion forum. To improve students’ self-awareness, it is needed to collect and analyze the forum data and visualize students’ sentiments to provide feedback to the students. This paper presents a model for sentiment analysis using natural language processing techniques to visualize students’ affective states towards the course as a strategy to enhance students’ self-awareness. We used the Stanford MOOC Posts dataset, from Stanford University’s eleven public online courses to test the proposed model. Finally, the paper presents a method to visualize the insights gained from the analysis in a student-facing intelligent learning dashboard (SF-iLDs) to support students’ self-awareness of their sentiment towards the course to encourage adjustments to the level of engagement.