Profiling Student Learning from Q&A Interactions in Online Discussion Forums

• Main Authors: ONG, De Lin, SHIM, Kyong Jin, GOTTIPATI Swapna
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2021
• Subjects: Profiling student learning; Online discussion forum; Q&A interactions; Slack; Computer Sciences; Educational Assessment, Evaluation, and Research; Higher Education
• Online Access: https://ink.library.smu.edu.sg/sis_research/6897; https://ink.library.smu.edu.sg/context/sis_research/article/7900/viewcontent/Profiling_Student_Learning_from_Q_A_Interactions_in_Online_Discussion_Forums.pdf
• Institution: Singapore Management University

The last two decades have witnessed an explosive growth in technology adoption in education. Proliferation of digital learning resources through Massive Open Online Courses (MOOCs) and social media platforms coupled with significantly lowered cost of learning has brought and is continuing to take education to every doorstep globally. In recent years, the use of asynchronous online discussion forums has become pervasive in tertiary education institutions. Online discussion forums are widely used for facilitating interactions both during the lesson time and beyond. Numerous prior studies have reported benefits of using online discussion forums including enhanced quality of learning, improved level of thinking beyond the classroom, collaborative knowledge building, and enhanced participation by shy or intimidated students. By monitoring and analyzing students’ activities in online discussion forums, instructors can intervene and manage students’ learning. For the instructor to employ appropriate intervention measures, both quantitative and qualitative analyses of students’ participation are important. To mitigate the challenge of the sheer volume of conversation threads in online discussion forums, we present a text mining approach to profiling student learning based on Q&A interactions. Firstly, we perform text classification to categorize conversations into two categories: non-programming-related and programming-related. Secondly, from the programming-related conversation threads, our method categorizes students into four participation proficiency types based on their Q&A activities. Next, our method determines whether a student adopts more explicit or implicit expression behavior in Q&A activities. We evaluate our approach on the second-year computing course, Web Application Development II. Finally, we share the lessons learned in this teaching process.