When to Protect? Using the Crosswise Model to Integrate Protected and Direct Responses in Surveys of Sensitive Behavior

The applied social sciences have witnessed a growing use of sensitive survey techniques (SSTs) to study the relationship between facets of an individuals background and his propensity to engage in sensitive behavior. The rationale undergirding the use of these techniques is the assumption that the rate of misrepresentation and/or nonresponse under direct questioning among individuals bearing the sensitive trait would be so high as to make the use of direct questioning infeasible. But is this indeed the case? Presently available methodological tools do not provide an answer. The current paper presents a survey questioning strategy and corresponding statistical framework that simultaneously addresses the question of whether or not the use of a SST is required to study a given sensitive behavior, provides an estimate of the prevalence of the sensitive behavior in the population of interest, and, in its extended form, describes how individual characteristics relate to the likelihood of engaging in the behavior.