The Evidence on Teaching

A Meta-Analysis of the Efficacy of Teaching Mathematics with Concrete Manipulatives

Carbonneau, K. J., Marley, S. C., & Selig, J. P. · 2013

grade Cmeta-analysisindependentreplicated
Sample
55 studies / N=7,237, K-college
Population
Typically developing students, K through college.
Design
Manipulatives vs abstract-symbols-only. Outcome classes coded separately; moderators: guidance, perceptual richness, age, topic. Exact pooled d paywalled (~0.37 commonly cited, unverified).
Key findings
Manipulatives help — conditionally: with high instructional guidance, perceptually bland objects, ages ~7-11, and especially fractions. The effect is a guided-representation effect, not a 'hands-on learning' effect; rich/toy-like objects hurt transfer.
Genetic confound
Mostly experimental comparisons.
Replication notes
Categorical pattern (guidance matters; bland > rich distally) consistent with Kaminski/Sloutsky and Peltier 2020.
DOI / URL
10.1037/a0031084

Effects

OutcomeMetricValueMeasureTimingVsHorizonClass
Retention of taught contentdsmall-to-moderate overall (exact value unverified)mixedmostly immediatebusiness-as-usualend-of-treatmentdomain-skill
Problem solving / transferdsmallmixedend of treatmentbusiness-as-usualend-of-treatmentnear-transfer
Moderator: instructional guidancecontrastHIGH guidance > low for retention/problem-solving (transfer direction unresolved — one summary reports a low-guidance transfer advantage)mixedend of treatmentactive-alternativeend-of-treatmentdomain-skill
Moderator: perceptual richnesscontrastBLAND objects beat rich ones on distal outcomes (rich manipulatives distract)mixedend of treatmentactive-alternativeend-of-treatmentnear-transfer
Moderator: age/topiccontrastages 7-11 > 12+; fractions strongest topicmixedend of treatmentbusiness-as-usualend-of-treatmentdomain-skill

Cited by

A Meta-Analysis of the Efficacy of Teaching Mathematics with Concrete Manipulatives · The Evidence on Teaching