The project aims to discuss the effect of class size on students’ academic performance. In this project, the author explores and aggregates the data distributions of 1st-grade students in the STAR program. To further investigate the connection between class type and students’ academic performances, specifically their math grades, the author sets up two Two-Way ANOVA models to analyze average and median math scaled scores across different classes. However, some assumptions in the models are violated. To address this issue, a mixed-effect model is applied. Consequently, the mixed-effect model confirms the effect between class size and median math grade with no assumption violations.
- Two-Way ANOVA Model with mean aggregation
- Two-Way ANOVA Model with median aggregation
- Mix Effect Model