Li, H., Chen, Y., Liao, W., & Wang, X.
Grouping learners based on complementary knowledge strengths, tracked dynamically, significantly improves collaborative learning outcomes.
Researchers developed and tested a five-step collaborative grouping model that uses a deep learning algorithm to dynamically diagnose students' mastery of specific learning objectives. Using these diagnostics, a clustering algorithm groups students with complementary knowledge profiles, which are then manually refined by instructors. The experimental results show this systematic, data-driven approach leads to more equitable groups, better peer interactions, and increased overall learning efficiency.
When designing collaborative tasks, group learners by balancing complementary knowledge profiles rather than using random assignment or self-selection. Use diagnostic assessments to identify individual strengths on specific sub-topics, then pair learners who can teach and learn from one another. Always allow instructors to manually adjust these data-driven groups to account for real-world interpersonal dynamics.
Change point detection in learning analytics reveals exactly when and how different gamification strategies shift student behaviors.
Engaging graduate students in co-creating design-based research transforms them into active, networked practice-based scholars.
Building digital textbooks using a crowdsourced model of professionals, students, and AI produces highly rated, deep learning resources.
Researchers have validated a reliable 30-item scale to measure how university instructors adopt and use AI in their teaching.
Summaries are generated automatically from the publisher's abstract or article text. Always check the original source before citing.