Estimating and correcting for response bias in self‐reported HIV risk behavior

Publisher Website
Google Scholar
Abstract
One of the most difficult problems of behavioral research on human sexuality and drug use is its reliance on people's self‐reports about their behavior. Given the essentially private nature of sexual and drug‐taking practices, many aspects of these practices are difficult, even impossible, to validate. The lack of a gold standard for self‐reports is particularly troubling with regard to HIV‐related behavior. To the extent that self‐reports are affected by response bias, the prevalence of high risk practices will be underestimated, hindering efforts to slow the spread of HIV. In this paper the extent of response bias in seven measures of high‐risk injection‐related and sexual behaviors is estimated. Self‐reported high‐risk injection practices may be significantly to substantially underreported due to response bias, whereas self‐reports of some high‐risk sexual practices do not appear to be contaminated by response bias. In regression analyses, the response bias in self‐reports of needle sharing and other high‐risk injection practices was modeled, and estimates of the probable actual prevalence of these practices derived. In situations in which behavioral self‐reports cannot be validated, such derivations may help in arriving at more accurate estimates of behavior.