Including typical student solutions in explicit instruction showed no significant effect on conceptual knowledge in the Bayesian-reasoning replication¶

Claim · All claims
Evidence · 1 study · causal r2 · q3 · i0 negligible
Currency · newest study 2024 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i0 Conditions with typical student solutions in instruction did not significantly outperform conditions without them on conceptual knowledge. → Loibl 2024

Evidence¶

Loibl 2024¶

Loibl, K., & Leuders, T. (2024). Can failure be made productive also in Bayesian reasoning? A conceptual replication study. Instructional Science. https://doi.org/10.1007/s11251-024-09670-y

q3 · i0 · causal · r2

Same 2×2 experiment (N = 75): the ANOVA on conceptual knowledge found no significant effect of the factor type of instruction, i.e., instruction with vs. without typical student solutions (F(1,71) = 0.35, p = .56, ηp² = 0.005). Descriptively, conditions with typical solutions averaged 2.5 points vs. 1.3 without, but the difference was not significant.

"the factor type of instruction (hypothesis 2: F (1,71) = 0.35, p = .56, ηp 2 = 0.005)"

Discussion¶

Related Claims¶

• In a Bayesian-reasoning replication, problem-solving prior to instruction (PS-I) showed no significant advantage over instruction-first (I-PS) on conceptual knowledge — related
• PS-I effects did not replicate in all domains: null sequence effects in Bayesian reasoning, javelin throwing, and inquiry-skill sequencing — a broader claim this one bears on
• No significant interaction between order of phases and type of instruction on conceptual knowledge in the Bayesian-reasoning replication — related
• Students' generated solutions rarely matched the erroneous solutions included in the explicit instruction phase — related
• PS-I sequence outperforms I-PS on conceptual knowledge only when instruction builds on student solutions — related