Abstract
Measuring the effects of HIV/AIDS prevention efforts through self-reported condom use data is problematic since such data are frequently subject to bias. The purpose of this study was to pretest a sexual behavior instrument to identify ways to improve comprehension and decrease recall and social desirability bias. The method used was cognitive interviewing, in which fifteen sex workers were asked to “think aloud” as they problem solved simulated tasks. Results indicated that (1) a lexicon of culturally appropriate terms for sex and partner types should be provided to interviewers, (2) recall of sexual behavior may be aided by using grids to record information, and (3) social desirability may be decreased if trust is established with the participants. Using the “think aloud” method to pretest instruments may help improve the validity of self-reported sex acts, but only more objective measures of condomuse may help to quantify such improvements.
Get full access to this article
View all access options for this article.
