One-year mathematics achievement growth and its determinants: Evidence from the TIMSS Longitudinal study

This research project examines determinants of student learning growth using data from TIMSS Longitudinal (TIMSS-L), a new international panel study that follows students from Grade 4 to Grade 5 and from Grade 8 to Grade 9. The data was collected in 2023 and 2024. The project comprises four coordinated studies, each focusing on a distinct set of predictors of achievement growth: (1) student characteristics, (2) teacher characteristics, (3) teaching quality, and (4) school climate. Together, these studies aim to identify which factors are associated with differences in learning progress in mathematics over one year. While each study addresses a specific substantive domain, all analyses are conducted within a shared and harmonized methodological framework to ensure consistency and comparability across papers. Analytically, the project focuses on modeling within-student growth in achievement, using both change-score approaches and linear multilevel regression models that control for baseline achievement. The hierarchical structure of the data (students nested within classes and schools) is explicitly taken into account, and all analyses respect the complex sampling design of TIMSS-L, including the use of appropriate sampling weights and plausible values for student achievement. Analyses are conducted separately by country and grade.