The Impact of Interviewer Effects on Skin Color Assessment in a Cross-National Context

Skin color of respondents is increasingly being collected by interviewers in face to face surveys and used in social science research. Nevertheless, skin color has the potential to be significantly influenced by interviewer effects. We use the AmericasBarometer (LAPOP) study to investigate interviewer variation in skin color ratings, the variation of this by country and interviewer characteristics as well as the impact of correcting for interviewer effects. We find that interviewers explain about 20% of the variation in skin color assessments and this varies greatly by country. We also find that the darker the skin color of the interviewer the darker their rating of the respondents and that correcting for interviewer effects leads to small differences in regression coefficients.