Smaller, underpowered studies report larger flipped classroom effect sizes, indicating small-study and publication bias¶

Claim · All claims
Evidence · 1 study (2 entries) · quant-synthesis r3 · q3
Currency · newest study 2019 · checked against its sources 2026-10-31 (automated, 2 of 2 entries) · not reviewed by a person

Subclaims¶

q3 i? Meta-regression shows study sample size exerts a fairly strong influence on effect sizes, with smaller studies having larger effect sizes. → Torstein Låg 2019 q3 i? Funnel-plot inspection and trim-and-fill indicate clear small-study or publication bias for continuous-learning measures. → Torstein Låg 2019 (2)

Evidence¶

Torstein Låg 2019¶

Torstein Låg, Rannveig Grøm Sæle. (2019). Does the Flipped Classroom Improve Student Learning and Satisfaction? A Systematic Review and Meta-Analysis. AERA Open. https://doi.org/10.1177/2332858419870489

q3 · i? · quant-synthesis · r3

Meta-regression on continuous-learning outcomes (sample size range 13–1,554). The article reports a negative sample-size coefficient (Z = −2.46, p = .01), consistent with small-study effects inflating the summary estimate.

"study sample size seems to exert a fairly strong influence on effect sizes, with smaller studies having larger effect sizes."

Torstein Låg 2019 (2)¶

Torstein Låg, Rannveig Grøm Sæle. (2019). Does the Flipped Classroom Improve Student Learning and Satisfaction? A Systematic Review and Meta-Analysis. AERA Open. https://doi.org/10.1177/2332858419870489

q3 · i? · quant-synthesis · r3

Publication bias assessment using funnel plots, Kendall's Tau, Egger's regression intercept, cumulative meta-analysis, fail-safe N, and trim-and-fill across the three outcomes. The article reports clear indications of bias for learning measures and less clear-cut but present bias for pass rates and satisfaction.

"For the continuous-learning measures outcome, there are clear indications of a small-study or publication bias."

Discussion¶

Related Claims¶

• Publication bias tests indicate the pooled flipped classroom effect is unlikely to be inflated by publication bias — reports the opposite
• Flipped classroom interventions produce a small positive effect on continuous learning outcomes, around one fifth of a standard deviation after correcting for small-study bias — related
• Publication bias is present in the Turkish flipped learning literature, and correcting it lowers the pooled effect from 0.895 to 0.713 — related
• The meta-analysis shows signs of publication bias, with smaller studies tending to give more positive results than larger studies — possibly the same claim (merge candidate)
• Small sample sizes and publication bias against non-significant findings may inflate meta-analytic effect size estimates in the chemistry teaching-method categories — related
• Funnel-plot asymmetry ran contrary to small-study effects because larger-sample studies reported larger effect sizes, likely reflecting longer training programs — reports the opposite
• Publication-bias checks indicate the pooled effect is unlikely to result from publication bias — related
• Testing student preparation is associated with slightly higher learning gains in flipped classroom implementations — related
• Publication year significantly moderates flipped classroom effect sizes in meta-regression — related
• No evidence of publication bias was found in the flipped classroom literature, though dissertations showed a larger (non-significantly different) effect size than journal articles — related
• No evidence of publication bias in the SDL meta-analysis — related