Do schools and teachers leave their mark? Studies on long-term effects of schools and teachers on student outcomes

In the field of educational effectiveness research, teacher and school effects are most often studied in the short term. Long-term school and teacher effects are less often studied and the findings are inconsistent. The objective of this dissertation is to study long-term school and teacher effects. The main question is: do schools or teachers still influence students’ outcomes at the moment these students already changed school or teacher? And in the case long-term effects were found, we were interested which school characteristics played a predictive role. Besides a focus on cognitive outcomes, long-term effects were also studied on non-cognitive outcomes, educational positions, and unemployment rates. The large-scale, longitudinal SiBO and LOSO data sets were analysed for this dissertation. In these projects, a large sample of students was followed during respectively primary and secondary education. Afterwards, we added follow-up data. The research questions were answered using cross-classified multilevel models. In the first three studies, long-term effects of primary schools were investigated on three outcomes: on educational positions of students during secondary education, on non-cognitive outcomes at age 17, and on mathematics achievement at age 17. The school network (i.e. private vs public schools), group composition, and effectiveness of the primary school were used to explain the long-term primary school effects. In the fourth study, long-term effects of first grade teachers on the achievement of students at the end of the second year were studied. Finally, in study 5, on the one hand, long-term school effects of primary and first cycle secondary schools were studied on the final position of students in secondary education. On the other hand, long-term effects of primary, first cycle secondary, and upper secondary schools on the unemployment rates after being, on average, 16 years on the labour market were investigated. The network, type, and region of the school were taken into account. The results indicate that primary schools have, after correction for student intake, long-term effects on students’ mathematics achievement at age 17, on a few non-cognitive outcomes, and on the final positions of students in secondary education. In other words, comparable students going to different primary schools can experience a difference in their mathematics achievement, non-cognitive outcomes, and final position in secondary education, due to the primary school. Different school characteristics were found to explain (part of) these effects. Students going to schools from the private network (i.e. mainly Catholic schools) attain a higher math achievement and a higher final position compared to students from the public network schools. The economic or language group composition of the primary school was found to make no difference in the long term. The schools’ effectiveness in mathematics has a significant effect on the educational positions of students, on some non-cognitive outcomes, and on their math achievement at age 17. First cycle secondary schools were found to have a long-term effect on students’ final position in secondary education and on their unemployment, when taking students’ background into account. Upper secondary schools were found to have long-term effects on the unemployment of students, after correction for student background variables. With regard to long-term effects of teachers, no significant effects were found. First grade teachers were found to have an important short-term effect, but once their students move on to the next teacher, no long-term effects were found. We can conclude that schools have small, but significant long-term effects on their students, while no long-term effects of teachers were found. Comparable students going to different schools can attain different achievement levels, final positions, or unemployment rates due to their school. This can consequently influence their choices in higher education and/or on the labour market, which then influences their further life. These small long-term differences between schools can thus leave lifelong marks and cannot be underestimated.status: publishe