Exploring Sources of Heterogeneity in Supplemental Interventions for Students with Mathematics Difficulties within Multi-tiered Systems of Support: A Meta-Analysis

About This Grant

Many students across the country face ongoing challenges in learning mathematics, which can significantly limit their academic progress and restrict future opportunities in higher education and careers, particularly in high-demand fields such as science, technology, engineering, and mathematics (STEM). To address these difficulties, schools frequently implement supplemental instructional programs, such as tutoring, small group interventions, or specialized curricula, aimed at helping struggling learners to increase their mathematics proficiency. Critical questions remain unanswered despite these efforts: Which instructional approaches are most effective? For which types of learners do they work best? And under what classroom or school conditions do they produce the strongest outcomes? This project seeks to answer these questions by synthesizing decades of educational research focused on mathematics interventions. By analyzing patterns across a wide range of studies, the project aims to identify the specific teaching strategies, learning environments, and implementation practices that consistently support positive student outcomes in mathematics. The ultimate goal is to equip educators, school leaders, researchers, and policymakers with clear, evidence-based guidance on designing and delivering effective mathematics support for struggling students. This work intends to strengthen the foundational understanding about students and their ability to engage with complex mathematical concepts. To achieve this, the research team will conduct a comprehensive meta-analysis of group-design (i.e., experimental and quasi-experimental) and single-case design studies evaluating supplemental mathematics interventions. The analysis will examine how malleable instructional features, such as intervention dosage (e.g., duration and frequency), alignment with core content (e.g., fractions and algebra), and student characteristics (e.g., area of academic risk, such as mathematics or reading), moderate the effectiveness of interventions. The study will apply traditional meta-regression techniques and advanced machine learning methods, specifically Random Forest Regression via the MetaForest package in R. This approach uses decision trees to identify which study features most strongly influence outcomes, allowing the team to detect complex, non-linear moderator effects that may not be visible through traditional models. Risk of bias and study quality will be rigorously assessed using frameworks such as the What Works Clearinghouse design standards and the Council for Exceptional Children quality indicators. Students with mathematics difficulties include learners from a wide range of educational backgrounds, making it essential to identify effective instructional approaches that work across varied settings and student populations. By integrating multiple analytic strategies with a strong theoretical foundation, the project aims to produce nuanced and actionable insights that improve intervention effectiveness and support the development of a more STEM-ready student population. This project is supported by the EDU Core Research (ECR) program. ECR supports fundamental research that generates foundational knowledge to advance research literature in STEM learning and learning environments. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.