The Evidence on Teaching

The Effect of Testing Versus Restudy on Retention: A Meta-Analytic Review of the Testing Effect

Rowland, C. A. · 2014

grade Cmeta-analysisindependentreplicated
Sample
159 effect sizes
Population
Mostly undergraduates in lab; verbal materials.
Design
Cleaner control than Adesope (testing vs restudy only). Lab verbal materials, proximal tests.
Key findings
Cleanest testing-vs-restudy meta: g=0.50, 81% favor testing, driven by effortful recall, delay, and feedback. Lab verbal materials with proximal tests — external validity to authentic content not established here.
Genetic confound
Retention outcome only.
Replication notes
Converges with Adesope and classroom Yang 2021.
DOI / URL
10.1037/a0037559

Effects

OutcomeMetricValueMeasureTimingVsHorizonClass
Retention: testing vs restudyg0.50 [0.42,0.58]; 81% favored testingresearcher-designedmixed; concentrated on delayed tests; labactive-alternativeuncleardomain-skill
Moderator: final-test formatdirectioneffortful RECALL > recognitionresearcher-designedmixed labactive-alternativeuncleardomain-skill

Cited by