K* models outperform step regression for detecting transfer-shallowness, with no-interactions and interaction models each best on different metrics¶

Claim · All claims
Evidence · 1 study · associational r1 · q2
Currency · newest study 2013 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? The K* no-interactions model achieved better cross-validated Kappa and precision than all other models, while the K* multiplicative-interactions model achieved better A′ and recall than all other models for no-transfer-shallowness. → Gowda 2013

Evidence¶

Gowda 2013¶

Gowda, S. M., Baker, R. S., Corbett, A. T., & Rossi, L. M. (2013). Towards Automatically Detecting Whether Student Learning is Shallow. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-013-0006-4

q2 · i? · associational · r1

Comparison of four models (K* and step regression, each with and without multiplicative interactions) in Table 2, evaluated with 10-fold student-level cross-validation on A′, Kappa, precision and recall. The authors conclude the overall success of the K* models suggests the relationships are non-linear.

"the K* no-interactions model achieves a better cross-validated Kappa and better precision than all other models, and the K* multiplicative interactions model achieves better A ′ and better recall than all other models."

Discussion¶

Related Claims¶

• K* detectors identify no-PFL-shallow learners with cross-validated A′ of 0.790 and kappa 0.55 — related
• K* detectors identify no-transfer-shallow learners at student-level cross-validated A′ of 0.766–0.791 — related
• Shallowness detectors transfer between the transfer-shallow and PFL-shallow constructs with minimal A′ degradation — related