Learning outcome can be predicted from overall BtBC via support vector regression¶

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

Subclaims¶

q2 i? SVR with nested cross-validation decoded learning outcome from overall BtBC, and the prediction held after controlling for instructional approach. → Yafeng Pan 2022

Evidence¶

Yafeng Pan 2022¶

Yafeng Pan, Suzanne Dikker, Yi Zhu, Cuirong Yang, Yi Hu and Pavel Goldstein. (2022). Instructor-learner body coupling reflects instruction and learning. npj Science of Learning. https://doi.org/10.1038/s41539-022-00131-0

q2 · i? · causal · r2

Prediction analysis on the combined personalized and nonpersonalized groups using support vector regression with outer leave-one-out and inner 10-fold CV. Controlling for instructional approach, the correlation remained significant (r = 0.41, p = 0.004, R² = 0.16).

"Support vector regression (SVR) analysis with nested cross validation (Fig. 4a) yielded a correlation coef ficient of 0.44 ( p = 0.002, R2 = 0.16) between the actual and predicted learning outcome"

Discussion¶

Related Claims¶

• Directional leader–follower BtBC patterns decode instructional approach better than individual motion, with peak performance at non-zero time lags — related
• Learner-leading BtBC is larger under scaffolding; instructor-leading BtBC shows only a trending effect, and instructor-leading BtBC fails to predict learning under explanation — related
• Instructional personalization did not significantly affect learning outcome or BtBC — related