Does Survey Respondents’ Immigrant Background Affect the Measurement and Prediction of Immigration Attitudes? An Illustration in Two Steps

Abstract

Over the past few decades, Western societies have become highly diverse, with an increasing share of the population having foreign roots. Just like the native population, defined here as national citizens with national ancestry and no foreign roots (up to the third generation), individuals with an immigrant background have opinions on newcomers to the country. Yet, most research on attitudes toward immigration using large-scale survey data routinely excludes their responses, without verifying whether their inclusion actually affects the findings. We argue here that it is crucial to examine whether methodological considerations actually justify exclusion. To illustrate how to do so, we define two necessary steps for evaluating the impact of respondents’ immigrant background and apply them to data from a Swiss survey. Most large-scale research on immigration attitudes relies on secondary data from international social surveys. In these surveys, respondents are generally invited to provide an evaluative judgment of immigrants or immigration in general (e.g., “The government spends too much money assisting immigrants”; International Social Survey Programme, 2003) or to immigrants from regions with different economic conditions (e.g., “people from the poorer countries in Europe”; European Social Survey, 2002). Because no specific group is mentioned, it is often argued that respondents with an immigrant background could have their own national group in mind when answering such questions, which would presumably lead them to adopt more positive attitudes (Hjerm, 2009). It has also been suggested that the reasons for adopting negative immigration attitudes differ as a function of immigrant background (Herda, 2010).