A multi-level analysis of the effect of interviewer characteristics on survey respondents' reports of sensitive topics

Self-report survey research has become the bedrock of theory construction and testing in the social sciences. Given the sensitive nature of topics examined in psychology, criminology, and elsewhere empirical understanding of factors that affect the validity and reliability of the data are crucial. As a result, a wide swath of research has evaluated the extent to which various methods of self-report techniques affect data quality. Left relatively under examined in this literature is the extent to which interviewer characteristics influence variation in self-reported data regarding sensitive behaviors. The current study addresses this gap by assessing the influence of a variety of interviewer characteristics on reports of antisocial behavior, victimization, and sexual behavior. Employing data from a nationally representative sample of American adults, multi-level analyses indicated that interviewer characteristics had virtually no effect on the variation in these behaviors. Discussion focuses on the validation of the data for analyses using these measures and study limitations.