Actions Speak Louder Than Values: Testing the Robustness of the Inaction-Driven Prejudice Increase Effect

Discrimination and prejudice are serious problems in our society. The desire for cognitive consistency is eminently influential on how such implicit and explicit attitudes translate into behavior and how attitudes change over different contexts. While it has been found that experiencing cognitive dissonance between discriminating behaviors and one’s values can lead to bias reduction, a recent study by Szekeres et al. (2022) also found the opposite: Inaction, or a failure to confront, in the face of witnessing prejudice, subsequently correlates with more discriminating attitudes. This effect was reported to be mediated by how much a person values confronting; it was specifically observed for respondents that valued confronting highly. The present paper aimed to test the robustness of this interaction of confronting importance with confronting behavior on attitude change across other subsamples. For this, three different resampling approaches instrumentalizing bootstrapping were...Show moreDiscrimination and prejudice are serious problems in our society. The desire for cognitive consistency is eminently influential on how such implicit and explicit attitudes translate into behavior and how attitudes change over different contexts. While it has been found that experiencing cognitive dissonance between discriminating behaviors and one’s values can lead to bias reduction, a recent study by Szekeres et al. (2022) also found the opposite: Inaction, or a failure to confront, in the face of witnessing prejudice, subsequently correlates with more discriminating attitudes. This effect was reported to be mediated by how much a person values confronting; it was specifically observed for respondents that valued confronting highly. The present paper aimed to test the robustness of this interaction of confronting importance with confronting behavior on attitude change across other subsamples. For this, three different resampling approaches instrumentalizing bootstrapping were employed that excluded different percentages of random respondents. The exclusions were either completely random or random within either extreme or central scoring participants. The comparison of the mean interaction coefficients and variances of the different resampling conditions showed that the mean interaction remained quite stable after resampling. Only excluding large percentages of extreme values led to somewhat of a drift to a larger interaction effect.Show less

Leiden University Student Repository

Documents

In Collections

Actions Speak Louder Than Values: Testing the Robustness of the Inaction-Driven Prejudice Increase Effect