Student characteristics and mathematics achievement growth: Evidence from the TIMSS longitudinal study

Understanding how student characteristics are associated with learning outcomes is central to evaluating educational systems. However, international comparative evidence has relied largely on cross-sectional studies that do not account for prior achievement, thereby limiting insight into the roles of prior achievement, gender, and socioeconomic status. Using panel data from the TIMSS Longitudinal Study, this paper examines mathematics achievement development across multiple education systems. Two cohorts were analyzed: students followed from Grade 4 (2023) to Grade 5 (2024) in nine countries, and from Grade 8 (2023) to Grade 9 (2024) in three countries. Country-specific regression models were estimated to assess whether achievement growth is associated with prior achievement (Matthew effect), gender, or socioeconomic status. Results indicate consistent compensatory patterns, with prior achievement negatively associated with subsequent gains. When prior achievement is controlled, gender and socioeconomic differences are generally small and inconsistent. In contrast, models excluding prior achievement show larger disparities. These findings underscore the importance of accounting for baseline achievement when interpreting educational inequalities.