TrackThinkDashboard: Understanding Student Self-Regulated Learning in Programming Study

2025-03-25

AI Summary

Supporting and assessing students’ self-regulated learning (SRL) in programming education requires interpretable, actionable tools. Method: We propose a multimodal learning behavior visualization dashboard that uniquely integrates and time-aligns web browsing and programming log data. Through a mixed-methods empirical study (N=33), we identify distinct SRL strategies—including trial-and-error and search-oriented approaches—and demonstrate the significant influence of domain knowledge on behavioral patterns. Contribution/Results: The dashboard enables students to reflect on problem-solving processes and pinpoint knowledge gaps, while supporting teachers in identifying at-risk learners with precision. Results show significant improvements in students’ metacognitive awareness and teachers’ data-driven intervention capabilities. The framework provides a scalable, reproducible, evidence-based approach for SRL assessment in computing education.

Technology Category

• Machine Learning: Multimodal Learning
• Search and Optimization: Metareasoning and Metaheuristics
• Application Domains: Humanities & Computational Social Science

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

In programming education, fostering self-regulated learning (SRL) skills is essential for both students and teachers. This paper introduces TrackThinkDashboard, an application designed to visualize the learning workflow by integrating web browsing and programming logs into one unified view. The system aims to (1) help students monitor and reflect on their problem-solving processes, identify knowledge gaps, and cultivate effective SRL strategies; and (2) enable teachers to identify at-risk learners more effectively and provide targeted, data-driven guidance. We conducted a study with 33 participants (32 male, 1 female) from Japanese universities, including individuals with and without prior programming experience, to explore differences in web browsing and coding patterns. The dashboards revealed multiple learning approaches, such as trial-and-error and trial-and-search methods, and highlighted how domain knowledge influenced the overall activity flow. We discuss how this visualization tool can be used continuously or in one-off experiments, consider associated privacy implications, and explore opportunities for expanding data sources to gain richer behavioral insights.

Problem

• Visualizing student learning workflows in programming education
• Identifying knowledge gaps and self-regulated learning strategies
• Enabling teachers to detect at-risk learners effectively

Innovation

• Visualizes learning workflow via integrated logs
• Helps students monitor problem-solving processes
• Enables teachers to identify at-risk learners