InitialD converges faster toward the final diagnosis only for small question counts; a standard CAT converges faster beyond about 14 questions¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2018 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? For k ≤ 9 questions InitialD converges faster towards the final diagnosis, while for k ≥ 14 the CAT baseline converges faster, indicating adaptation benefits emerge only in later stages. → Jill-Jêenn Vie 2018

Evidence¶

Jill-Jêenn Vie 2018¶

Jill-Jêenn Vie, Fabrice Popineau, Éric Bruillard, Yolaine Bourda. (2018). Automated Test Assembly for Handling Learner Cold-Start in Large-Scale Assessments. Int J Artif Intell Educ. https://doi.org/10.1007/s40593-017-0163-y

q2 · i? · design · r2

Numerical cross-validation result on the Fraction dataset (Results section): the diagnostic advantage of the non-adaptive DPP sampling is limited to the early cold-start stage; adaptive selection takes over once several ability estimates exist.

"For k ≤ 9, InitialD converges faster towards the final diagnosis, while for k ≥ 14, CAT converges faster, showing a benefit in adaptation in later periods of the process."

Discussion¶

Related Claims¶

• On the TIMSS dataset, InitialD reaches the same estimation error with fewer questions than random sampling or a traditional adaptive test — related
• On the Fraction dataset, InitialD outperforms random sampling, uncertainty sampling and a normal adaptive test in cold-start response-pattern reconstruction — related
• On the TIMSS dataset, random sampling performs well compared to a normal adaptive test, possibly because the test is already balanced — related
• Cold-start initialization converges faster than starting from an inconsistent student model — related