An interpretable mathematical model based on saccade lengths and angles performs well, is transparent, and is theoretically meaningful for identifying students' histogram strategies¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2024 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? The IMM is transparent in how it classifies strategies and its performance is comparable to the MLA baseline. → Boels 2024

Evidence¶

Boels 2024¶

Boels, L., Garcia Moreno-Esteva, E., Bakker, A., & Drijvers, P. (2024). Automated Gaze-Based Identification of Students' Strategies in Histogram Tasks through an Interpretable Mathematical Model and a Machine Learning Algorithm. International Journal of Artificial Intelligence in Education, 34, 931-973. https://doi.org/10.1007/s40593-023-00368-9

q2 · i? · design · r2

The IMM was constructed based on attentional anchors found in a prior qualitative study, using saccade lengths and angles to classify gaze patterns as predominantly horizontal or vertical. The authors state it "is transparent, performed well, and is theoretically meaningful."

"The IMM (2) is transparent, performed well, and is theoretically meaningful."

Discussion¶

Related Claims¶

• Students' horizontal or vertical gaze patterns on the graph area of single histograms are indicative of their task-specific strategies for estimating the mean — related
• The machine learning algorithm and the interpretable mathematical model identified the same task-solving strategies from students' gaze data — related
• Decision trees achieve competitive accuracy with black-box models (LR, RF, SVM), which performed slightly better but lack interpretability — related