Beyond class size: Understanding student academic achievement through student, school, and national contexts

Educational systems now operate in a globalized and highly competitive environment, making international assessments like PISA central to policy discussions. This study investigates how variables at the student level (i.e., student socioeconomic status (SES) and educational track), the school level (i.e., classroom structural conditions of class size and student-teacher ratio, and school selectivity), and country level (i.e., GDP per capita) contribute to variation in student academic achievement in PISA. Using data from 77 countries in PISA 2022 and applying three-level multilevel modeling, the study examines the relative contributions of these variables to student academic achievement in PISA assessment in mathematics, reading, and science. Findings show that the student-level variable of student SES and the school-level variables of class size are positively associated with student academic achievement in PISA assessment across domains, while the student-teacher ratio is not significant. In addition, results showed that the country-level variable of GDP per capita is positively associated with PISA assessment in reading. An additional analysis examined a cross-level interaction to test whether GDP per capita moderates the relationship between class size and student academic achievement; however, no significant direct or moderating effects were found. Overall, the results indicated that the variance in student academic achievement in the PISA assessment is similarly distributed across all three levels, suggesting a multi-level structure of influences. The study offers meaningful insights for improving educational outcomes globally.