Using Behavior Coding to Examine the Role of Interviewer and Respondent Race/Ethnicity When Enumerating Hard-to-Reach Racial Minorities
*Rodney Terry, U.S. Census Bureau
Keywords: Behavior Coding, Race, Ethnicity, Coverage, Validation Study, Face-to-face Interview, 2010 Census
The purpose of this study is to examine the associations between interviewer and respondent race/ethnicity with respondents’ answers to coverage and race/ethnicity questions. We use data from an ethnographic evaluation of the 2010 Census Nonresponse Followup Operation (NRFU) to analyze portions of the NRFU interview in which interviewers asked respondents questions on coverage and race/ethnicity. The 2010 Census NRFU evaluation was conducted at eight sites targeted to different race/ethnic groups in the United States. With respondent permission, 193 of the NRFU interviews were tape-recorded and later behavior coded. We then examine these interviews to identify the associations between interviewer and respondent race/ethnicity with multiple interview interaction outcomes, including whether: (a) interviewers correctly read coverage and race/ethnicity questions, (b) respondents gave appropriate answers that fit into the response categories of coverage and race/ethnicity questions, and (c) respondents gave answers to initial coverage questions that are inconsistent with later coverage validation questions when either types of questions are incorrectly read by interviewers.