National and regional proportion of immigrants and perceived threat of immigration: A three-level analysis in Western Europe

Immigration is of growing significance to the demographic makeup of Western Europe. A long-standing and highly disputed question is whether a larger number of immigrants are associated with more negative attitudes toward immigration or whether the reverse is true. Previous studies yielded contradictory results on various levels of analysis (national, regional, local). These inconsistencies may partly be linked to what is known as the modifiable areal unit problem' in spatial analysis. This article seeks to address this issue by analyzing the relationship between the percentage foreign-born and perceived group threat in 15 Western European countries on the national as well as on three differing regional levels (N=70, 207, and 624 regions, respectively), together with survey data from the European Values Study's fourth wave. I expect threat effects to operate through national communication systems while contact and habituation to immigrants to work on the regional level. Consistent with theoretical expectations, the results show a positive correlation between the national proportion of immigrants and perceived threat, while the coefficients are negative on the regional level. More immigration might thus lead to a more negative evaluation of the presence of immigrants in European countries, but apparently not within the regions where most of the newcomers reside. Two recent examples illustrate this seemingly paradoxical relationship. As a methodological result, effect size and statistical significance vary with the delimitation of the regional units of analysis (Nomenclature des Unites Territoriales Statistiques (NUTS)-1, -2, or -3). This suggests that research in this field should pay more attention to how and why spatial units are defined.