The testing effect for mathematics learning (testing versus restudy, g = 0.22) is not robust, as its 95% confidence interval crosses zero.¶

Claim · All claims
Evidence · 1 study · quant-synthesis r2 · q3
Currency · newest study 2023 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? Across 6 studies with 20 effect sizes, testing versus restudy yields a weighted mean of g = 0.221, but the 95% confidence interval [-0.091, 0.532] crosses zero. → Murray 2023

Evidence¶

Murray 2023¶

Murray, E., Horner, A. J., & Göbel, S. M. (2023). A Meta-analytic Review of the Effectiveness of Spacing and Retrieval Practice for Mathematics Learning. https://osf.io/

q3 · i? · quant-synthesis · r2

Meta-analysis of the retrieval-versus-restudy subset (6 studies, 20 effect sizes), mostly university settings. The weighted mean was "g = 0.221 (se = 0.108)" and the authors state the confidence interval crossed zero, so the effect is not robust.

"the overall weighted mean effect size is g = 0.221 (se = 0.108) (95% confidence interval [-0.091, 0.532]). Note that the 95% confidence intervals crossed the zero line."

Discussion¶

Related Claims¶

• Retrieval interval length may moderate the testing effect in mathematics: restricting to intervals over one day raises the effect to g = 0.378 (exploratory). — related
• Testing benefits retention over a conservative restudy control, establishing the testing effect as robust — a broader claim this one bears on