What Makes Hypocrisy? Folk Definitions, Attitude/Behavior Combinations, Attitude Strength, and Private/Public Distinctions

Abstract

Past research has rarely examined what makes behaviors appear more or less hypocritical. This work expands our understanding, identifying and exploring factors contributing to perception of hypocrisy. An initial study surveyed participants’ definitions of the concept. Experiments 2a/2b then demonstrate that attitude–behavior inconsistency is viewed as most hypocritical, followed by attitude-attitude and behavior-behavior inconsistency. Experiments 3 and 4 examined how perception of hypocrisy depends on attitude strength, communication method, and whether attitudes/behaviors are privately or publicly held/enacted. We conclude that hypocrisy is perceived as strongest when attitudes are publicly imposed on others in an attempt to appear morally superior.

Notes

Acknowledgments

We thank Sara Hodges for her very helpful comments on an earlier version of this manuscript.

Notes

1 We note that perceived hypocrisy was measured in three studies and reported separately in an online supplement to this article.

2 The full coding instructions are provided in the Online Supplementary Materials (OSM), available at https://osf.io/vzqad/.

3 Between these anchors, under the numbers 4 and 5, the words “moderately hypocritical” were used.

4 Some minor variability in hypocrisy ratings as a function of the interaction between stimulus set and pairing condition was found in both experiments. This suggested the presence of method variance. However, within each stimulus set, pairing-condition means were ordered in the same way as the aggregate analyses, with very similar effect sizes for comparisons between pairings.

5 A replication of this experiment, using a within-participants design (Experiment 3S), is reported in the OSM. Results were entirely consistent with those presented here, although some additional findings emerged.

6 Analyses that included these participants did not substantively differ from those that excluded them.

7 Vignette-level analyses are presented in the OSM. Results of those analyses are fully consistent with those reported here.

8 In the replication of this experiment that used a comparative (within-participants) design (Experiment 3S), the effect size for this comparison (d = 0.15) was similar to what was reported here.

9 A replication of this experiment, using a comparative (within-participants) design (Experiment 4S), is reported in the OSM. Although some variations emerged (see footnote 13), results were consistent with those presented here.

10 Given that average reading times were close to 1 min, we reasoned that anyone finishing in 10 s or less probably hadn’t read the vignettes closely. Analyses that retained these participants did not substantively differ from those presented here.

11 Vignette-level analyses are presented in the OSM. Results of these analyses were fully consistent with those reported here.

12 Using a within-participants design in Experiment 4S (reported in the OSM), tests of H1a and H1b both tended to show the opposite effect (e.g., greater hypocrisy when behaviors were public rather than private), suggesting that more research is needed before firm conclusions can be drawn.

13 In the replication of this study, the effect size was small (d = .07) and in the opposite direction

14 We thank an anonymous reviewer for offering this last suggestion.

15 Of course, it is important to note that when attitudes are truly private and remain so, no one will ever know they exist, making any attribution of hypocrisy on their basis impossible.