Skin Tones and Polarized Politics: How Skin Color Differences Between Interviewers and Respondents Influence Survey Answers in Bolivia

Abstract

Abstract American scholarship claims that the racial make-up of interviews influences the attitudes disclosed in public opinion surveys. It remains unclear whether such an effect travels to other cases where racial cleavages are less salient, and whether it affects all respondents. We address these gaps by using a flexible approach focusing on skin tone rather than race. Relying on survey data from Bolivia, where polarization maps onto ethnic lines, we investigate whether the skin color difference between an interviewer and a respondent influences the latter’s answers. Building on the race-of-interviewer effect and colorism literatures, this article investigates the effect of the skin color dynamic of interviews by leveraging the random interviewer-to-respondent assignment process of LAPOP surveys. The results suggest that nonresponses are more likely in cross-skin tone interviews and that respondents questioned by interviewers of lighter skin tone than them will express opinions that more closely align with the stereotypical opinions of the interviewer than if their interviewer shared their skin tone. This article contributes to the interviewer effect literature by testing the competing claims of the social distance and social acquiescence theories, along with providing an adaptation of the race-of-interviewer effect to cases that are not polarized along racial lines.