Dosage Response in Intensive Mathematics Interventions for Early Elementary Students With or At-Risk for Mathematics Learning Disability

Abstract

We examined the effect of intervention dosage on mathematics outcomes for students with or at risk for mathematics learning disability in kindergarten through third grade. Using linear and nonlinear models, we meta-analyzed 164 effect sizes from 24 published experimental and quasi-experimental studies. Results supported a linear model and an overall pooled mean effect size of β =.61 (95% CI [.29,.92]). Mixed-effects meta-regression analyses revealed that dosage (i.e., total hours spent in intervention) significantly predicted intervention effects such that for every 1-h increase in dosage, there was an increase of.03 effect size units (95% CI =.01,.06). However, the dosage main effect was no longer significant after controlling for moderators. Outcome proximity did significantly predict the dose–response relationship, such that effect sizes of proximal outcomes increased at a higher rate as a function of dose than did distal outcomes. Intervention effects were also moderated by group size such that 1:1 interventions had higher effect sizes than small-group interventions. Recommendations for future research and implications for mathematics intervention are discussed.