A Study of Using GenAI to Identify Key Course Factors of 25 Online Courses with High Learning Engagement at an Open University in China

With the rapid advancement of open education and the increasing demand for high-quality learning experiences, understanding the course-related factors that contribute to high learning engagement has become a critical research area. Traditional methods of analyzing course characteristics often face numerous challenges, such as time-consuming processes and subjective biases. GenAI offers new approaches to overcome these difficulties. Using GenAI models to analyze the texts of 25 online courses with high learning engagement at an Open University in China, this study attempts to discover common characteristics of these courses to provide insights into best practices for designing and organizing online courses. The findings show that clear guidance, automatic feedback test, and practical material might be more important factors when catering to learning engagement in open universities. However, social interaction and teacher support at the course level were found to be less important than previously thought, and their actual value for learning engagement needs to be further explored in different situations. The study also validated the possibility of applying GenAI models to analyze textual data according to a given framework. It provides a new pattern for detecting and assessing online course quality on large scales: real-time, automatic analysis of real-world data in learning management systems (LMS).