Contextual effects in the formation of educational expectations among the children of immigrants in Madrid

This paper explains the differences existing among the children of immigrant and native-born families in their adjusted educational expectations (expectations controlling for prior performance) in Spain, an under-represented case in the international literature. We use individual-level predictors to account for the gap between these two groups and explore the explanatory potential of broader contextual factors, namely the school and classes within schools. We use the best empirical tools available for the study of contextual effects in the formation of educational expectations: (a) the recently available survey 2011-Chances, a large-scale dataset of children before the end of compulsory schooling in Madrid; and (b) the 2010 Secondary Schooling National Evaluation Survey, for the first time exploited along this line of research. Most of the unconditional disadvantage experienced by immigrants in Spain is accounted for by individual-level variables. Once the family background and the basic socio-demographic characteristics are considered, the migrant background appears to have a positive effect on expectations, confirming the what has recently been seen in other European countries. Surprisingly, schools appear to be relevant in unconditional models, but the conditional impact of school effects is at best modest.