Abstract:
In an era when online education is becoming increasingly prevalent, and many institutions are transitioning onto online platforms, it has become crucial to understand onl...Show MoreMetadata
Abstract:
In an era when online education is becoming increasingly prevalent, and many institutions are transitioning onto online platforms, it has become crucial to understand online learning behaviors. The user data kept by online learning platforms has supported relevant research that has made great strides in analyzing both engagement and dropout rates in online courses. In online programming courses, learners struggle to understand the code and even code by themselves, while instructors barely have access to check everyone's code and give advice in real-time. It becomes critical for instructors to quickly grasp how learners are progressing and encountering difficulties in their online programming learning. This study combines learners' programming activity data, click stream data from video learning, and textual data such as accompanying notes and questions. A visual representation that combines multiple learning data is proposed, and learners' Learning Trajectories Map in an online programming course is created from experimentally collected data. The results show a strong correlation between learners' programming activities and their engagement with the learning material. In addition, by comparing the learners' learning trajectories, it was possible to observe their different learning strategies, which provided more detailed clues for studying behavioral patterns and learner characteristics in online learning. This study introduces a new tool for educational evaluation, provides a more comprehensive perspective, and addresses the limitations of relying only on a single indicator for evaluation. It is of great significance to educational technology and computer-assisted education.
Date of Conference: 18-20 March 2024
Date Added to IEEE Xplore: 05 June 2024
ISBN Information: