Prior automatic MCQ generation approaches produce structurally simple questions (at most two stem entities) testing mainly recall¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2019 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? A review of ontology-based and medical question-generation approaches found generated questions restricted to at most two stem entities and mostly testing recall of learned information. → J. Leo 2019

Evidence¶

J. Leo 2019¶

J. Leo, G. Kurdi, N. Matentzoglu, B. Parsia, U. Sattler, S. Forge, G. Donato, W. Dowling. (2019). Ontology-Based Generation of Medical, Multi-term MCQs. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-018-00172-w

q2 · i? · design · r2

The authors' narrative review of prior approaches (Papasalouros et al., Alsubait et al., Wang et al., Karamanis et al., and others) reports that generated question "stem contains at most two stem entities" and mostly test recall. This is the review's own qualitative assessment, not a measured comparison.

"Most notable is the simplicity of the structure of auto-generated questions when compared to hand-crafted questions."

Discussion¶

Related Claims¶

• EMCQG generated more than 3 million distinct case-based questions for four physician specialities from the EMMeT ontology — related
• Ontology-based generation produced exam-appropriate case-based MCQs, with 30% of a stratified sample rated appropriate by both expert raters and 50% by at least one — related
• Case-based (vignette) MCQs are classified as testing higher-order thinking and discriminate better between low- and high-information students — related