Can skill differences explain the gap in the track recommendation by socio-economic status?

Tracking early in the school career can influence a student's further educational path significantly. We study the track advice at the end of primary school in the Netherlands, where teachers give a track advice based on a student's previous performance and their impression of the student's ability. If the student outperforms the initial advice in the subsequent nationwide test, the school reevaluates the student and can – but does not have to – update the final advice. We use cognitive and non-cognitive skills measurements that are collected three years before the tracking decision is made, linked with the teachers initial and revised advice, as well as background information from register data. We find that with equal skills, students from lower socio-economic backgrounds receive on average lower advice, while students with a migration background receive on average higher advice. A decomposition of the total difference in initial advice between students from high versus low educated parents shows that around 55% of the difference in advice can be explained by differences in cognitive and non-cognitive skills. Adding additional information about the family, school and place of residence, we can explain about 71% of the difference between students with low and high educated parents. We do not find a significant change in the gap in advice between children from different socio-economic backgrounds after the nationwide test and reevaluation procedure.