Race of Interviewer Effects Under Explicit and Implicit Race-Matching in a Telephone Survey
LeBaron, P., Boland-Perez, C., & Cobb, P. (2007, May). Race of Interviewer Effects Under Explicit and Implicit Race-Matching in a Telephone Survey. Presented at American Association for Public Opinion Research Conference, Anaheim, CA.
Race of interviewer effects on data quality in both face-to-face and telephone surveys have been investigated for some time. These effects are particularly strong when the topic is racially sensitive (Hatchett & Schuman 1975-1976; Cotter, Cohen & Coulter1982.) The impact of race of interviewer effects depends upon whether the respondent is able to determine the race of the interviewer (Cotter, Cohen & Coulter 1982). In order to further examine these effects in a telephone interview, the Annual Racial Attitudes Survey of Pulaski County (AR) matches the race of interviewer to that of the respondent in every interview. Given demographics of Pulaski County, of which Little Rock is the county seat, only respondents who reported their race as "White" or "Black or African American" were race-matched for the interview. Data collection for the 2006 cycle of the Annual Racial Attitudes Survey of Pulaski County included a quasi-experiment. The purpose of this was to assess potential effects of explicit versus implicit race-matching. The protocol included random assignment to two primary groups.Respondents in Group 1 received explicit notification at the beginning of the survey that the interviewer was of the same race and received a brief overview of the rationale for the race match. Respondents in Group 2 were not notified of the race match, but each respondent was asked to identify the interviewer's race at the conclusion of the survey. Accordingly, this group can be divided into two subgroups - those respondents who realized they were race matched even without explicit early notification and a much smaller proportion who did not realize they were of the same race as the interviewer. The goal of this analysis is to assess the potential differences, if any, among these groups by comparing responses to the most racially sensitive items on the questionnaire. Although public opinion scholars recognize the potential benefits of race-matching on telephone surveys, few surveys actually employ race matching protocols. Thus, this analysis contributes to knowledge of the potential implications for the variety ways of implementing race-matching.