Education and Cultural Integration among Ethnic Minorities and Natives in The Netherlands: A Test of the Integration Paradox

Abstract

This contribution sets out to identify the main determinants of two attitudinal indicators of cultural integration that figure in the construction of acculturation strategies: opposition toward ethnically mixed relationships and identification with the origin country. We derive our expectations from Social Identity Theory on the direct and indirect effects of education. To test our hypotheses we use data from the Survey on the Integration of Minorities 2006 (SIM 2006) and apply multiple mediator models. SIM 2006 covers random samples from the four largest ethnic-minority groups in the Netherlands—Turks, Moroccans, Surinamese and Antilleans—and a native Dutch comparison group. We do not consistently observe the presumed negative effect of education on opposition toward ethnically mixed relationships and identification with the origin country among ethnic minorities. On the contrary, the presumed beneficial impact of education on cultural integration is absent for second-generation migrants, which is considered to be evidence of an integration paradox.

Keywords:

• Integration Paradox
• Attitudes of Immigrants
• Education
• Media
• Mediator Models
• The Netherlands

Notes

1. A suppressor effect is present when the mediator is positively correlated to both education and the dependent variable. Consequently, the direct negative effect of education would become stronger after controlling for this mediator. Conversely, we speak of an interpretation effect when the direct negative effect of education becomes smaller after introduction of the mediator.

2. It is theoretically possible that someone who identifies more with the ethnicity of origin than with the Netherlands identifies more strongly with the Netherlands than someone who exclusively identifies with the Netherlands. A dichotomous variable with the categories: (1) feel completely the ethnicity of origin country versus (0) do not feel completely the ethnicity of origin country, does not suffer from this ambiguity. We did not observe differences in signs among the effects of education and the mediators on both these variables within, respectively, OLS and logistic regression models.