The longitudinal impact of student characteristics, school composition and track placement on mathematics performance: Inter- and cross level intersectionality

In recent decades, much sociological inquiry has focused on whether and to what extent education systems are capable of compensating for and equalizing social inequalities. While earlier studies have mainly focused on educational inequalities and their relationship with individual students’ characteristics (e.g., Boudon, 1974; Bourdieu, 1984); or school-level factors, particularly school composition and segregation (e.g., Jencks, 1974), few studies have been concerned with the intersectionality of individual and school-level factors and their impact on the performance of students within and across levels of education (Gross et al., 2016).
The present study aims to investigate the intersectional impact of students´ academic and socio-demographic characteristics, school composition and school tracks on students’ mathematics performance in Luxembourg. It draws on data collected as part of the Luxembourg school monitoring programme “Épreuves Standardisées” (ÉpStan; Fischbach et al., 2014) and included all students enrolled in public education Grade 3 (November 2013) matched with data from the same students in Grade 9 (November 2017-2021) including those repeating once or twice (N≈3600).

Results of multilevel mixed effects regression analysis show that math achievement in Grade 9 is affected by student gender, SES, migration background and prior achievement, as well as by the school track and school composition (i.e., percentage of Low SES families in school). In addition, a robust cross-level gender x school track interaction effect was found. Results show that after controlling for prior performance and other individual characteristics, the institutional placement of students into different school tracks and school composition in Grade 3 influence test results in Grade 9. The cross-level interaction effect indicates that the boy-girl achievement gap is even more pronounced in the higher (academic) than in the middle (more technical) track. Results support earlier findings that both individual and school composition variables, and especially their intersectionality, contribute to differences in educational outcomes.