Constructivist learning models have a medium-level overall effect (ES=.78) on K-12 student outcomes across 19 meta-analyses¶

Claim · All claims
Evidence · 1 study · quant-synthesis r3 · q3
Currency · newest study 2022 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? The second-order meta-analysis of 19 meta-analyses yields a mean effect size of ES=.78 (LL=.70, UL=.87) for constructivist learning models on student outcomes, with substantial heterogeneity (I2=81.68). → Demir 2022

Evidence¶

Demir 2022¶

Demir, M., & Kaya, M. (2022). Analysis of constructivist learning model's effects on student outcomes: A second order meta-analysis. Journal of Theoretical Educational Science, 15(4), 938-957. http://doi.org/10.30831/akukeg.1122136

q3 · i? · quant-synthesis · r3

Second-order meta-analysis under a random effect model combining 19 first-order meta-analyses of constructivist learning models. The authors report "The mean impact size is ES=.78 LL=.70 UL=.87", with individual effect sizes ranging from ES=.45 to ES=1.20.

"The dataset is made of n=19 independent me ta-analysis research. Meta -analysis research consists of a total of 623 independent basic research. k=21 impact size is obtained from the meta -analysis research constituting the dataset . Impact sizes vary between ES=.45 and ES=1.20. The mean impact size is ES=.78 LL=.70 UL=.87."

Discussion¶

Related Claims¶

• The effect of constructivist learning models on student outcomes varies significantly by location, with Turkey (ES=.87) and Indonesia (ES=1.07) showing high-level effects and China (ES=.58) and mixed-country samples (ES=.67) medium-level effects — a narrower finding that bears on this claim
• Constructivist learning models show high-level effects on thinking skills (ES=.88) and academic success (ES=.81) but a medium-level effect on attitudes (ES=.51) — a narrower finding that bears on this claim
• Cooperative learning models have a moderate average effect on student learning outcomes across 23 first-order meta-analyses — related
• The effect of cooperative learning differs by the location of the meta-analysis studies, with single-country studies showing higher effects than mixed-country studies — related
• Publication bias in the dataset is low and unimportant: Egger's test is non-significant and the trim-and-fill adjusted effect size (ES=.76) differs from the observed value by only about .02 — related
• Mean effect sizes across the 13 reviewed meta-analyses ranged from -.11 to .57 — related