ITALIAN JOURNAL OF SOCIOLOGY OF EDUCATION

This article provides empirical evidence for the ethnic concentration effect on natives’ and non-natives’ performance in mathematics and reading in upper secondary schools in the North and Center of Italy, where the large majority of immigrants is concentrated. Once schools and compositional controls are included in the models, results show that the ethnic concentration effect on students’ achievement is non-linear and negative effects might occur only once a rather high and empirically still rare concentration of immigrants is reached. The misspecification as linear effect will lead to wrong conclusions and consequently, to wrong policy implications. Further, it is shown how immigration affects differently natives and non-natives, only in vocational schools, where social problems and exclusion tend to concentrate. Last but not the least, results show the overrepresentation of non-native students in vocational schools, with a low quality teaching, and with a large concentration of students from the most disadvantaged social strata.