Without initial instruction, repeated worked examples produce better generalization than repeated retrieval practice¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r2 · q3 · i2 medium
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i2 In Experiment 2 (no initial instruction), participants studying worked examples scored significantly higher on the final test than those doing retrieval practice overall (ηp² = .11). → Meng Cao & Paulo F. Carvalho 2026 q3 i? The repeated-items simple effect drove this: repeated worked examples (M = 0.69) beat repeated retrieval practice (M = 0.49). → Meng Cao & Paulo F. Carvalho 2026 (2)

Evidence¶

Meng Cao & Paulo F. Carvalho 2026¶

Meng Cao & Paulo F. Carvalho. (2026). Striking the Balance: How Variability Shapes Retrieval Practice and Worked Examples for Transfer Learning. Educational Psychology Review. https://doi.org/10.1007/s10648-026-10169-w

q3 · i2 · causal · r2

Experiment 2 replicated Experiment 1's procedure with the initial instruction omitted (74 participants after exclusions). The mixed-effects ANOVA on final-test accuracy showed a worked example advantage with printed effect size ηp² = .11 ("F(1,72) = 8.72, p = .004, ηp 2 = .11") — the opposite pattern from Experiment 1.

"participants who studied worked examples performed significantly better on the final test than those who engaged in retrieval practice, F(1,72) = 8.72, p = .004, ηp 2 = .11."

Meng Cao & Paulo F. Carvalho 2026 (2)¶

Meng Cao & Paulo F. Carvalho. (2026). Striking the Balance: How Variability Shapes Retrieval Practice and Worked Examples for Transfer Learning. Educational Psychology Review. https://doi.org/10.1007/s10648-026-10169-w

q3 · i? · causal · r2

A post hoc simple-effects analysis within the repeated-items condition of Experiment 2 found repeated worked examples (M = 0.69) outperformed repeated retrieval practice (M = 0.49), "t(72) = −4.34, p = < .001"; no standardized effect size was printed for this contrast.

"Participants who studied repeated worked examples (M = 0.69, SE = 0.04) performed better on the final test than those who engaged in repeated retrieval practice ( M = 0.49, SE = 0.04), t(72) = −4.34, p = < .001."

Discussion¶

Related Claims¶

• Without initial instruction, varied retrieval practice matches varied worked examples and beats repeated retrieval practice for generalization — related
• When learners receive initial instruction, retrieval practice yields better generalization than worked examples regardless of item variability — reports the opposite
• Retrieval Practice Improves Transfer — reports the opposite