Date of Graduation

Spring 5-21-2026

Document Type

Dissertation

Degree Name

Doctor of Education (Ed.D.)

College/School

School of Education

Department

Learning and Instruction

Program

Learning & Instruction EdD

First Advisor

Patricia L. Busk

Second Advisor

Helen Maniates

Third Advisor

Nicola A. McClung

Abstract

Mastery learning has been one of the most extensively researched instructional strategies in education, studied continuously since the 1960s. Prior meta-analyses have consistently found positive effects of mastery learning on student learning outcomes, however, none have examined its effectiveness on student achievement explicitly within science, technology, engineering, and mathematics (STEM) disciplines.

This comprehensive meta-analysis investigated the effectiveness of the mastery-learning instructional strategy on student achievement across STEM disciplines in K-12 and higher-education contexts. A systematic literature search yielded 85 studies with combined sample sizes of 10,244 participants that met the inclusion criteria. Heterogeneity was assessed using Q,I2, and T2. Using a random-effects model, this meta-analysis found a statistically significant, very large positive effect of mastery learning on student achievement across STEM disciplines (g = 1.53, 95% CI [1.35, 1.70], I2 = 94.11%). Funnel plot asymmetry and Egger’s regression tests indicated substantial publication bias, and after trim-and-fill adjustment for 29 imputed studies, the effect size remained statistically significant, with a large positive effect (g = 0.92, 95% CI [0.72, 1.12]. Cumulative analyses showed the evidence was robust regardless of study sample size.

Moderator analyses indicated that STEM disciplines significantly influenced effectiveness, with technology showing larger effects (g = 1.68) than science (g = 1.60) and mathematics (g = 1.49). In the science discipline, Biology yielded the largest effect size (g = 2.13), whereas in mathematics, Geometry produced the largest effect size (g = 1.95). For the educational level, elementary-school students produced the largest effect size (g = 2.60). Moreover, meta-regression showed that mastery learning produced larger effect sizes at lower-grade levels than at higher-grade levels, suggesting that younger students benefit most from mastery-learning applications. Studies that implemented mastery learning for 2 to 4 weeks produced a statistically significant, very large effect (g = 1.65), suggesting the intervention is very effective over this duration in enhancing student achievement. For mastery-level criteria, the results revealed that a mastery level of 80%-89% produced the largest effect size (g = 1.72), indicating that the mastery-level criterion used to define student proficiency significantly influenced the effects. For assessment frequency, the study indicated that a one-test retake to demonstrate mastery had a larger effect (g = 1.65) than two or more retakes (g = 1.52). Both teacher-developed assessments (g = 1.55) and constructed-response questions (g = 1.64) yielded statistically significant, very large positive effects on student achievement across STEM disciplines.

Download

Share

COinS