Search

Privacy and confidentiality are often of great concern to respondents answering sensitive questions posed by interviewers. Using the 1993–2010 General Social Survey, we examined trends in the provision of social security numbers (SSNs) and correlates of those responses. Results indicate that the rate of SSN provision has declined over the past three decades, that is, from about 60% in 1993–1994 to 20% after 2004. Although younger, unmarried, and less-educated respondents are more likely to provide SSNs, an indicator of trust toward others is not a significant predictor of SSN provision. Further, those who refuse to report their income and who receive an incentive are less likely to provide their SSN. Our findings have implications for studies involving sensitive questions as well as privacy and confidentiality issues.

This article examines the correlates of nonresponse for self-administered questionnaire (SAQ) sections by comparing two General Social Survey (GSS) years, 2000 and 2002, during which the GSS switched the paper-and-pencil personal interview (PAPI) and SAQ to the computer-assisted personal interview (CAPI) and computer-assisted self-administered interview (CASI) format. The authors find that CASI not only lowers the supplemental response rate but also changes the dynamics of who responds. Unlike PAPI and SAQ data collection in 2000, even those aged 45—64 were less likely to respond to the CASI supplemental survey compared to CAPI in 2002. Compared to Whites, Blacks and other ethnic groups were more likely to be nonrespondents with CASI. Researchers should be more attentive to the potential bias of supplemental nonresponse in CASI.