A hybrid DEMATEL and social network analysis model to identify factors affecting learners' satisfaction with MOOCs

Abstract

Massive Online Open Courses (MOOCs) offer free access to training in various topics in all fields. However, the low percentage of course completion by learners is a significant challenge for these platforms. Previous studies on this challenge have investigated user behavior and concerned topics in discussion forums, but these data are mostly momentary and cannot be used for long-term improvement. Thus, this study aimed to address this gap by analyzing learners' comments to identify the factors affecting user satisfaction and prioritize them to improve MOOC platforms. The purpose was to analyze the feedback and actual experiences of users shared through their comments on MOOC online platforms to explore factors affecting user satisfaction to optimize MOOC platforms. To achieve this, sentiment analysis and topic modeling techniques were applied to the user feedback on courses with popular topics, such as Skills for Data Science Teams and Data-Driven Decision Making, available on Coursera.com. The study used DEMATEL analysis, which uses a relation matrix of factors to rank them based on their interrelationships, and network analysis to prioritize the factors that should be improved to achieve the highest user satisfaction. The effect of the proposed approach was investigated through a case study on a course from Coursera. The findings demonstrate that the suggested method has the potential to assist MOOC platforms in several ways. Firstly, it enables the identification of course strengths and weaknesses. Secondly, it allows for the identification of factors that influence learner satisfaction by analyzing user feedback. Lastly, it aids in prioritizing the factors that should be enhanced to attain optimal user satisfaction, thus leading to overall improvement in the status of the MOOC platform.