Adding a Prior Error Valence (PEV) feature to a population-parameter model reduces systematic mastery-timing error in simulation¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2022 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? In the simulation, the error-sensitive model EPFA reduced systematic error toward zero, halving the median absolute difference in predicted mastery trial (9 to 4). → Eglington 2022

Evidence¶

Eglington 2022¶

Eglington, L.G., Pavlik, P.I. (2022). How to Optimize Student Learning Using Student Models That Adapt Rapidly to Individual Differences. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-022-00296-0

q2 · i? · design · r2

Numerical simulation comparing PFAP (population parameters) with EPFA (PFA plus PEV) on the same 1000 simulated students; the article reports EPFA "reduced systematic error towards zero as practice accumulated". No effect size printed.

"In this simulation, the median absolute difference between predicted and actual trial of mastery (i.e., predicting when correctness probability >=.95) for the PFAP was 9 vs. 4 for EPFA."

Discussion¶

Related Claims¶

• EPFA estimates fewer trials to mastery for above-average students and more trials for below-average students relative to standard PFA — related
• Learner models with individual-difference features (PEV, student-level proportional correctness) improve cross-validated fit on six educational datasets — related
• Adding PEV or student-level proportional correctness significantly reduces the absolute value of student-level signed error in all six datasets — related
• Learner models with population-level parameters systematically misestimate the trial of mastery for most students whose learning rates vary — related