Dosage significantly predicts intervention effects in a linear model (β = 0.03 per hour), but the dosage main effect is no longer significant after controlling for moderators¶

Claim · All claims
Evidence · 1 study (2 entries) · quant-synthesis r2–r3 · q3
Currency · newest study 2025 · checked against its sources 2026-10-22 (automated, 2 of 2 entries) · not reviewed by a person

Subclaims¶

q3 i? For every 1-h increase in intervention dosage, predicted effect sizes increased by 0.03 on average in the two-predictor model. → Anna H. Miller 2025 q3 i? When moderators were added (Model 3), the dosage main effect was no longer significant (β = 0.02; 95% CI [−0.29, 0.33]). → Anna H. Miller 2025 (2)

Evidence¶

Anna H. Miller 2025¶

Anna H. Miller, Daniel R. Espinas, Daniel McNeish, & Marcia A. Barnes. (2025). Dosage Response in Intensive Mathematics Interventions for Early Elementary Students With or At-Risk for Mathematics Learning Disability. Educational Psychology Review, 37, 91. https://doi.org/10.1007/s10648-025-10070-y

q3 · i? · quant-synthesis · r3

Linear random-effects meta-regression with robust variance estimation (Model 2), dosage centered at the median of 20 h, across 164 effect sizes. The authors report "For every 1-h increase in overall intervention dosage, predicted effect sizes increased by 0.03, on average"; the pooled effect at 20 h was 0.73 (95% CI [0.50, 0.96]).

"The slope of dosage (e.g., total hours of intervention) on effect sizes was significant and positive, β = 0.03 (95% CI [0.01, 0.06]). For every 1-h increase in overall intervention dosage, predicted effect sizes increased by 0.03, on average."

Anna H. Miller 2025 (2)¶

Anna H. Miller, Daniel R. Espinas, Daniel McNeish, & Marcia A. Barnes. (2025). Dosage Response in Intensive Mathematics Interventions for Early Elementary Students With or At-Risk for Mathematics Learning Disability. Educational Psychology Review, 37, 91. https://doi.org/10.1007/s10648-025-10070-y

q3 · i? · quant-synthesis · r2

Model 3 of the same meta-regression added instructional focus, outcome proximity, implementer, and group size as moderators with dosage interactions. The authors report "total hours of intervention is not a significant predictor of effect sizes at the median of dosage (i.e., 20 h)" in this full model.

"The main effect of dosage decreased and was no longer significant (β = 0.02; 95% CI [− 0.29, 0.33]), indicating that holding the effects of the other predictors constant, total hours of intervention is not a significant predictor of effect sizes at the median of dosage (i.e., 20 h)."

Discussion¶

Learner Variables¶

• Time and Continuity — moderator: an instructional effect differs with it

Related Claims¶

• Instructional focus (number and operations vs. word-problem solving) and intervention implementer (researcher vs. school personnel) did not significantly predict effect sizes or moderate the dose–response relation — related
• Multi-session game interventions show significantly larger learning benefits than single-session interventions, while total play duration shows no association with effect size — related
• Tier 2 dosage and scheduling findings are mixed: more frequent kindergarten sessions helped, but first-grade studies found no outcome differences across 16-hour and 32-hour schedules — related
• The dose–response relation for K-3 intensive mathematics interventions is linear, with no support for a nonlinear (quadratic) curve — related
• Group size moderates intervention effects: 1:1 interventions had effect sizes about 0.5 points higher than small-group interventions at 20 h of dosage — related
• Outcome proximity moderates the dose–response relation: distal outcomes outperform proximal outcomes at dosages at or below 5.43 h, while proximal outcomes exceed distal outcomes at or above 22.6 h — related
• Mathematics outcomes benefit from PA with academic instruction of 6–10-week duration at moderate or moderate-to-vigorous intensity, and language outcomes from enriched PA programmes — related