A BERT-based prediction task distinguishes between successful and unsuccessful math strategies in MATHia, yielding insights into student strategies¶

Claim · All claims
Evidence · 1 study · associational r? · q1
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person · recheck by 2028: tool and model accuracy changes quickly

Subclaims¶

q1 i? The authors learn a BERT model with embeddings for strategies and develop a prediction task distinguishing successful from unsuccessful strategies, analyzing its results for insights into student strategies. → Magar 2025

Evidence¶

Magar 2025¶

Magar, A.T., Fancsali, S.E., Rus, V., Murphy, A., Ritter, S., Venugopal, D. (2025). Analyzing Strategies in MATHia with BERT. In: Artificial Intelligence in Education. AIED 2025. Springer, Cham. https://doi.org/10.1007/978-3-031-98465-5_15

q1 · i? · associational · r?

Modeling contribution: the authors learn a BERT model for strategy embeddings and develop a prediction task to "distinguish between successful and unsuccessful strategies." No performance metrics or effect sizes are printed in the supplied text.

"Further, we learn a BERT model to learn embeddings for strategies, develop a prediction task to distinguish between successful and unsuccessful strategies, and analyze its results to reveal deeper insights into student strategies."

Discussion¶

Related Claims¶

• Students in MATHia find it hard to adapt their strategy choices to suit the problem, based on log data from over 600 schools — related
• Successful discoverers regulate learning more systematically, planning experiments and monitoring progress, while unsuccessful learners work unsystematically — related
• Process models of successful students come closer to self-regulated learning theory expectations than those of unsuccessful students, who do not evaluate their learning — related