Assessing Sensitive Attributes Using the Randomized Response Technique: Evidence for the Importance of Response Symmetry

Abstract

Randomized response techniques (RRTs) aim to reduce social desirability bias in the assessment of sensitive attributes but differ regarding privacy protection. The less protection a design offers, the more likely respondents cheat by disobeying the instructions. In asymmetric RRT designs, respondents can play safe by giving a response that is never associated with the sensitive attribute. Symmetric RRT designs avoid the incentive to cheat by not allowing such responses. We tested whether a symmetric variant of a cheating detection model (CDM) increases compliance with the instructions in a survey of academic dishonesty among 2,254 Chinese students. As we observed more noncompliance in the asymmetric than symmetric variant, we recommend the use of symmetric CDMs, which can easily be tested within multinomial models.

Identifying indicators of illegal behaviour: carnivore killing in human-managed landscapes

“…through access to police records) suggest that RRT returns more accurate responses compared with conventional survey instruments [19]; and, studies comparing survey methods found that RRT returned significantly higher estimates of sensitive or illegal behaviours compared with conventional surveys, which has been interpreted as evidence of more honest reporting [16,17,21,60,61]. We also used what is known as a symmetrical RRT design (prescribing fixed responses as both yes (when dice sum two, three or four), and no (when dice sum 11 or 12), which has been shown to increase the extent to which respondent follow RRT instructions [62]. Compared with conventional methods, RRT has one principle disadvantage owing to the random noise (added by the forced responses), RRT requires large samples in order to get estimates with acceptable precision [19].…”

A stochastic lie detector

“…Finally, comparable to related randomized-response models (Clark & Desharnais, 1998;Cruyff et al, 2008;Ostapczuk et al, 2009a), the stochastic lie detector assumes that response behavior does not depend on features of the randomized-response design. Specifically, it is assumed that the probability to respond honestly (t) is independent of the randomization probabilities p 1 and p 2 .…”

“…Several extensions of the forced-response variant of the RRT have been proposed that allow for determining the proportion of respondents who fail to adhere to the instructions by always providing a self-protecting response (Böckenholt, Barlas, & van der Heijden, 2009;Böckenholt & van der Heijden, 2007;Clark & Desharnais, 1998;Cruyff, van den Hout, & van der Heijden, 2008;Ostapczuk, Moshagen, Zhao & Musch, 2009a;van den Hout, Böckenholt, & van der Heijden, 2010). However, all of these approaches share the limitation that the true status of nonadherent respondents cannot be determined.…”

On the comprehensibility and perceived privacy protection of indirect questioning techniques

“…Academic cheating served as the sensitive attribute, as used in several studies of indirect questioning techniques (e.g., Hejri et al, 2013;Lamb & Stem, 1978;Ostapczuk, Moshagen, Zhao, & Musch, 2009;Scheers & Dayton, 1987). The wording of the sensitive question was identical in all questioning technique conditions, reading "Have you ever cheated on an exam?"…”