The purpose of this mixed-methods study was to investigate whether blended learning increased student achievement in middle schools at a public charter school system. Qualitative data was gathered through classroom observations (N=8), teacher interviews (N=8), and student focus groups (N=6). Quantitative analysis involved a difference-in-difference (DID) regression analysis. Data was gathered from seven schools and 44 classrooms. Level of blended learning implementation was recorded based on school principal rating on a 5-point scale. Data showed that a 1-point increase in blended learning was estimated to lead to a 0.05 standard deviation increase on the Measures of Academic Progress (MAP) math assessment on average, by 0.10 standard deviations for boys, 0.20 standard deviations for African American students, and 0.23 standard deviations for students with 504 plans (p < 0.05). Qualitative data suggests that blended learning was most impactful when teachers used data from adaptive digital content for differentiation. Limitations to this study as well as recommendations for future research are discussed.