The list of controls includes a wide range of individual characteristics of native and immigrant students, captured by Xil in equation (1). These include their age, sex, religion, natural or adopted child status, the number of children in their family, having a single parent as well as parents’ education, government employee status, occupation and age. Measures of family income, home ownership and mobility history are also included as controls. In addition, dummies for natives of Asian and other origin are specifically included in the native models. Immigrant-specific individual controls include dummies for European, Asian and other immigrants, for immigrants from developed countries and for those who speak English at home.
The variable of central interest is the immigrants’ share of student population in a locality in the native models and the like-type immigrants’ share of student population in a locality in the immigrant models.10Cl in equation (1) further captures a set of additional locality-level measures. These include the logarithm of mean income per capita, the Gini coefficient, the share of unemployed in the local labour force, the logarithm of the number of students and the private school student/teacher ratios.11 In the native models, the minority share (the share of native-born of non-European ancestry) of student population in a locality is included to control for the possibility of a flight from native-born minorities. Finally, state-fixed effects, denoted as Sl in equation (1), are included as controls. Table 1 provides the full list of variables and describes how they were defined.
Bookmarks