Obtaining sensitive information from a wary population: a comparison of telephone and face-to-face surveys of welfare recipients in the United States.

Recent studies reveal the benefits of technological developments such as audio computer assisted self-interviewing (A-CASI) in interview methodology, especially for surveys of sensitive behavior and information. However, we believe that the selection of mode of administration depends not only on the technology available and the behavior of interest, but also on the specific population under study. We therefore assess survey mode effects on reported rates of alcohol and drug use among welfare recipients, an especially important group for scholars and public health agencies. The sample consisted of adult recipients of Temporary Assistance to Needy Families (TANF) in Oklahoma, January 2001. Adjusting for demographic characteristics, employment, and education, we employ odds ratios to compare 30-day, 1 year, and lifetime prevalence estimates from telephone and face-to-face surveys. Telephone methodology yields similar or higher estimates for lifetime prevalence of alcohol, marijuana, and hard drug use and abuse, though lower estimates of recent use. We discuss our findings in relation to underfunded public health agencies that must efficiently assess and respond to local levels of drug abuse and we conclude that mode selection may depend upon the population under study.