Response Decoupling and Partisans' Evaluations of Politicians' Transgressions

There is growing interest in the study of expressive responding in public opinion surveys, with scholars seeking to determine whether partisan differences in response to survey items attest to genuine differences in perceptions or to insincere responses meant to signal in-party approval or out-party disapproval. This study focuses on partisan gaps in evaluations of the inappropriateness of politicians' transgressions and tests the effectiveness of a technique designed to reduce expressive responding. This "response decoupling" technique gives respondents the opportunity to separate their evaluations of a politician's performance from their evaluations of a transgression committed by the politician, thereby allowing partisans to both disapprove of the transgression and signal support for the politician. The technique was experimentally tested in a study in Israel (N = 906) and in two pre-registered studies in the US (total N = 3,172), as these studies presented respondents with a real-life transgression of an actual politician. Overall, the technique had a weak effect, as only in the Israeli study was the effect statistically significant. On the whole, these results suggest that while partisan gaps in evaluations of politicians' transgressions may reflect genuine perceptual differences, it may also be the case that allowing respondents to decouple their responses is not a sufficiently powerful method to reduce expressive responding. The paper concludes by discussing the implications and limitations of these findings.

The online version contains supplementary material available at 10.1007/s11109-022-09796-0.