Measuring Immigrant Populations: Subjective versus Objective Assessments

Abstract

Innumeracy amongst survey respondents in estimating a country’s immigrant population is a well-known problem for the social sciences (Herda 2010; Nadeau et al 1993). In general, individuals, especially if they are prejudiced against or perceive immigrants as a threat, are predisposed to overestimate the immigrant population at the country level. If this tendency can be generalized to other contexts such as neighborhoods, research that use subjective assessments of immigrant populations at lower levels of inquiry might be biased. Using survey data from Sweden, this paper compares individuals’ subjective assessments of the immigrant population in their neighborhoods with actual register/census data. The results show that subjective assessments correlate at least moderately with objective assessments. In contrast to previous research, we find that, under control for the objective number of immigrants in their neighborhood, prejudiced people bias their response by giving lower estimates of the number of immigrants in their neighborhoods. Finally, we show that the subjective measurement tend to yield analogous effects to the objective measurement in regression models when predicting generalized trust. However, when predicting attitudes towards immigrants, the objective and subjective measurements render somewhat disparate results. Overall, our results point to that the subjective assessment of immigrant populations should be used with some caution, especially if the interest is attitudes and prejudice towards immigrant.

Keywords: Objective assessments, Subjective assessments, Generalized trust, Attitudes towards Immigrants