Surveying Citizens with a Migration Background - A Quantitative Study of Identification versus Categorization

Survey research on racial/ethnic minority citizens of Germany and the Netherlands tends to use the category migration background, instead of measuring whether that category has any meaning for respondents by asking how they identify. Through surveying 1864 respondents in Germany and the Netherlands, including 401 respondents with a background in Türkiye, this paper shows that those who identify as Turkish hold different attitudes than those who have a background in Türkiye but who identify as German or Dutch. I provide proof of this with attitudes towards topics often associated with citizens with a migration background: belonging, Islam and sexuality. Respondents with a background in Türkiye, who do not identify as Turkish, hold the same attitudes as those without a migration background, and significantly different attitudes than when a researcher would categorize them with those having a migration background in Türkiye. This demonstrates that using identification rather than categorization yields more accurate and meaningful insights into respondents’ attitudes. Beyond this empirical advantage to identification over categorization, this paper outlines additional conceptual, methodological, and normative advantages to measuring identification in surveys.