Abstract
In this study, we investigated perceptions of hypocrisy among Christian and non-Christian individuals. Specifically, participants (N = 121) were asked to evaluate people of different religious orientations in terms of perceptions of both claimed morality, actual behavior, and general hypocrisy. Across all designations, perceptions of moral hypocrisy were determined to be greater to the extent that moral claims exceeded actual behavior. Among non-Christian participants, perceptions of hypocrisy were highest with respect to “devout” and “evangelical” Christian targets. In contrast, among Christian participants, the highest ratings of hypocrisy were reserved for “casual” Christians (i.e., Christians who do not incorporate their faith into their daily life). Our findings have both theoretical implications regarding the nature of religious hypocrisy perceptions, as well as real-world implications for understanding inter-faith perceptions and interactions.
Perceptions of Religious Hypocrisy: When Moral Claims Exceed Moral Action
Not many of you should become teachers, my fellow believers, because you know that we who teach will be judged more strictly. (NIV Bible: James 3:1
Prior research has demonstrated that individuals who are religious, as opposed to non-religious, benefit from a so-called “halo effect” (Bailey and Doriot 1985) in that they are perceived more positively on a wide variety of social dimensions (e.g., Swan et al. 2012; Galen et al. 2011; Tan and Vogel 2008). However, Swan et al. (2014) demonstrated that this perceptual halo effect for religious targets is more pronounced among perceivers who themselves identify as religious. In contrast, non-religious individuals are less likely to have such positive views of those who are religious. Additionally, Swan et al.'s findings suggest that individuals that profess dogmatic faiths are perceived more negatively than religious others who are more “flexible” in their beliefs (for similar findings, see Bobkowski and Kalyanaraman 2010; Galen et al. 2011; Tan and Vogel 2008).
The present study examined perceptions of hypocrisy among various Christian and non-Christian designations. Vermurlen (2014) demonstrated that Christians tend to be rated as more hypocritical than non-Christians; however, this study left important questions unanswered. First, it was not clear why Christians were perceived to be more hypocritical. Alicke et al. (2013) have demonstrated that that the essence of hypocrisy is a discrepancy between professed standards and actual behavior. The applicability of this equation to perceptions of religious individuals who profess lofty moral standards is straightforward. In addition, Alicke et al. (2013) demonstrated that public pronouncements of morality were associated with perceptions of hypocrisy. As it relates to the topic of religion, one could argue that those who are religious hold themselves to generally higher moral standards or are at least perceived to do so and also commonly make corresponding public pronouncements—particularly those belonging to denominations associated with proselytization (e.g., evangelical Christians).
The main focus of the present study was to assess the relationship between moral claims and moral behavior, on the one hand, and perceptions of hypocrisy, on the other, in the context of Christianity. Prior research in this domain has shown that type of religious designation can influence both perceptions and evaluations of perceivers. In the case of Isaac et al. (1995), regular churchgoers were rated more positively than occasional churchgoers. However, in other work, greater levels of dogmatism or religiosity predicted more negative evaluations (e.g., Galen et al. 2011; Swan et al. 2014). Based upon these findings, we assessed perceptions of casual Christians and two groups that would presumably be perceived as being more highly bound to their standards: devout Christians and evangelical Christians.
Study Overview and Predictions
Participants in the study were presented with a range of religious and non-religious designations and were asked to rate each group on two key dimensions: perceptions of making moral claims and perceptions of actual moral behavior. Based upon the work of Alicke et al. (2013), the discrepancy between these two measures was interpreted as an indirect measure of hypocrisy. We also assessed perceptions of hypocrisy using a direct measure.
To isolate the impact of spiritual designations on perceptions in the absence of other characteristics, we provided only minimalistic descriptions for each religious or non-religious designation. The use of such minimalistic descriptions allowed individual participants to factor in their own idiosyncratic category associations and perceptions. Generally, we were less concerned about individual differences in specific religious or non-religious designation construals, and more with general patterns that might emerge across all participants.
In general, we predicted that perceptions of hypocrisy for any given religious designation would be a function of the difference between perceptions of moral claims and moral behavior (with higher deviations corresponding with increased perceptions of hypocrisy). Regarding perceptions of the different religious designations, our predictions differed for Christian and non-Christian perceivers. For non-Christian perceivers, we expected perceptions of hypocrisy to be highest for evangelical Christians due to the association between this designation and moral claiming. For Christian perceivers, our predictions were less clear. It was possible that they would perceive increased hypocrisy amongst the non-religious due to perceptions of generally increased immorality. However, it was also possible that perceptions of hypocrisy would be higher for casual Christians, who might be perceived as hypocritical by definition by virtue of being less committed to their beliefs and ideals.
To examine whether participants spontaneously associate hypocrisy with moral claims and behaviors, we counterbalanced the order in which the direct hypocrisy measures appeared in the survey. If participants do not explicitly construe hypocrisy as the difference between moral claims and behaviors, we might expect stronger correlations between these measures and a direct assessment of perceptions of hypocrisy when the hypocrisy measures are presented last. In contrast, if participants construe hypocrisy as the difference between moral claims and moral behavior, the order of measures should not make a difference.
Method
Participants
Participants were recruited using the Mechanical Turk's (MT) online participant pool, in exchange for a nominal monetary compensation of fifteen cents. Buhrmester et al. (2011) examined the reliability of data collected using MT as compared to typical college student studies and comparable internet samples. They concluded that MT participants were slightly more representative of the general population than internet samples in general, and especially as compared to college-based samples, and the quality of data derived using MT samples was virtually identical to comparable internet or college student samples. More recently, Lewis et al. (2015), specifically explored characteristics related to religion in MT samples. In general, they found MT samples to be much less inclined to religion than the more representative samples captured by the General Social Survey. We hoped that this characteristic of MT samples would work in our favor, as we desired a relatively equal proportion of Christian and non-Christian participants. In addition, we were more interested in processes and associations relating to hypocrisy perceptions than constructing a demographically precise portrait of the population.
A total of 121 participants completed the survey (female = 52.9%). The mean participant age was 36.47 (SD = 12.14), and ages ranged from 18 to 67 years.
Materials and Procedures
In the context of an online survey on views of people who differ in terms of religious orientation, participants were prompted to consider “a person” belonging to six different religious designations with corresponding brief descriptions: (1) Atheist: a person who is a committed Atheist (does not believe in God); (2) No Interest: a person who has no interest in religious topics whatsoever; (3) Agnostic: a person who is Agnostic (believes you can neither confirm nor deny the existence of God); (4) Casual Christian: a person who is a “casual” Christian (believes in Christ, but it does not impact their daily life); (5) Devout Christian: a person who is a devout Christian (life centered around Christ); (6) Evangelical Christian: a person who is an evangelical Christian (committed to spreading the word of Christ).
Participants were asked to provide morality and hypocrisy perceptions regarding these six groups. General descriptions of these two dimensions of interest were provided to participants: Morality, “The degree to which a person upholds good practices/behaviors in terms of general society. For example, the degree to which a person lives by the law, makes “good” decisions, and treats others fairly,”; Hypocrisy, “A hypocritical person claims moral principles that they do not actually follow (i.e. says one thing and does another).” To assess perceptions of moral claims, participants were asked, “How moral does this person claim to be?” using a nine-point scale (1 = does not claim to be moral, 9 = claims to be exceptionally moral). Participants were then asked, “How morally does this person actually act in their everyday life?” (1 = does not act morally, 9 = acts exceptionally moral). To assess perceptions of hypocrisy explicitly, participants were asked, “How hypocritical would you consider this person to be?” (1 = not at all hypocritical, 9 = extremely hypocritical).
The order in which these three dimensions were presented in the survey was counterbalanced such that the hypocrisy portion was presented either first or last. The moral claims dimension was always presented prior to the actual behavior dimension. In addition, for each of the three dimensions, the six religious designation categories were presented in random order. At the conclusion of the survey, all participants were asked to briefly describe their religious orientation as well as their gender and age, and were automatically credited fifteen cents to their MT account.
Preliminary analyses indicated consistently high correlations among the three non-religious designations (ranging from .67 to 87) for all measures. Because we had no strong predictions regarding differences between these three designations, and also to minimize the number of statistical tests required for our analyses, we combined them all, by our three key measures, into a non-religious composite. The reliability coefficients for the three resulting composites are as follows: moral claims (α = .91); actual behavior (α = .93); hypocrisy (α = .89). In contrast, the correlations between the three religious designations were lower and less consistent (as expected), and so these were analyzed individually.
Results
Hypocrisy and Moral Perceptions
To examine the relationship between perceptions of hypocrisy and the remaining two moral perception dimensions, a series of difference scores were computed by subtracting perceptions of moral behavior from perceptions of moral claims for each of the four religious designations. Higher scores were indicative of greater gaps between moral claims and actual behavior. Hence, based on our predictions, positive correlations were expected between these difference scores and the respective hypocrisy scores.
To examine our prediction that hypocrisy judgments are based upon the perception that moral behavior falls short of moral claims, correlational analyses between the direct hypocrisy measure and the difference score described above were computed for each designation. The results were as follows: non-religious—r(119) = .30, p < .001; casual Christian—r(121) = .41, p < .001; devout Christian—r(120) = .35, p < .001; evangelical Christian—r(119) = .45, p < .001. These findings strongly confirmed our predictions regarding the nature of hypocrisy.
To determine whether the order in which the hypocrisy measures occurred in the survey may have influenced responses, we examined the effect of the order manipulation on the pattern of correlations between the difference scores (moral claims – actual behavior) and the hypocrisy measures. As described above, if people normally construe hypocrisy as a deviation between moral claims and moral behavior, the order manipulation should not have any effect. In contrast, if people do not naturally associate hypocrisy with moral deviations, having them consider moral deviations prior to making hypocrisy judgments would have been expected to strengthen the correlations between the difference scores and hypocrisy ratings.
The patterns of correlations by condition were as follows: non-religious—rHypocrisy first = .33, p = .012, rHypocrisy last = .27, p = .032; casual Christians—rHypocrisy first = .17, p = .203, rHypocrisy last = .56, p < .001; devout Christians—rHypocrisy first = .40, p = .002, rHypocrisy last = .33, p = .008; evangelical Christians—rHypocrisy first = .52, p < .001, rHypocrisy last = .47, p < .001. A series of Fisher's r–z transformations revealed that the only significant difference based on the order manipulation was for casual Christians (p = .005), for whom presenting the hypocrisy measures first decreased the association between difference scores and hypocrisy ratings. In general, the overall pattern of primarily null effects in this section is most consistent with the notion that participants naturally associated hypocrisy with a deviation between moral claims and moral behavior.
Perceptions of Hypocrisy and Religious Status
Our next aim was to examine differences in actual perceptions of hypocrisy, moral claims, and moral behavior across the four religious designations. Previous work has demonstrated consistent differences in religious perceptions based on the religious-designation of the perceiver (Swan et al. 2014). In our sample, participants were asked to report their religious orientation, and their open-ended responses were coded as either Christian (n = 55) or non-Christian (n = 62; four participants who provided unclear responses were excluded from the analyses that followed).
Perceptions of Religious Designations by Dimension and Participant Religious Orientation
For moral claims, moral behavior, and hypocrisy, higher means correspond with higher levels of these variables. The difference composite is moral claims minus moral behavior. Total degrees of freedom for the between participants analyses range from 114 to 116 for these analyses. Degrees of freedom for the within participants analyses range from 54 to 55 for Christian participants, and between 59 and 61 for non-Christians. Means in a given row that do not share a common subscript differ at p < .05
Beginning with the between-participant analyses, non-Christians perceived evangelical Christians to make significantly higher moral claims. Comparisons for all other conditions were not significant. For all remaining dimensions, a pattern of ingroup positivity bias was evident. For perceptions of actual moral behavior, non-Christians rated the non-religious as more moral than Christians did. In contrast, Christians rated devout and evangelical Christians as significantly more moral. Interestingly, Christians also rated casual Christians as marginally less moral—a pattern that suggests that Christians differentiate to some extent between serious (devout and evangelical) and non-serious (casual) Christians. This same pattern emerged for the hypocrisy measure, with non-Christians perceiving significantly more hypocrisy among devout and evangelical Christians, and Christians perceiving significantly more hypocrisy among casual Christians and the non-religious. Finally, regarding the difference scores, non-Christians exhibited larger difference scores for devout and evangelical Christians.
Focusing next on the within-participants analyses, the main point of interest is that, although both Christians and non-Christians appear to believe that devout and evangelical Christians profess more stringent moral standards, it is the non-Christians in the sample who believe that these two designations are falling short of meeting these standards. This pattern, then, would appear to explain the differences in perceptions of hypocrisy for the devout and evangelical Christian designations across Christian and non-Christian participants. Also of interest is that, among Christians, the highest hypocrisy ratings are assigned to casual Christians, but only for the direct hypocrisy measure.
In summary, the religious status of participants had a consistent and pervasive relationship with moral perceptions of various religious designations in a way that was largely consistent with our hypotheses.
Discussion
Although previous studies have examined general perceptions of those who are religious (e.g., Bailey and Doriot 1985; Isaac et al. 1995; Swan et al. 2014), the present study is the first to our knowledge to specifically assess and examine perceptions of hypocrisy among various religious and non-religious designations, while also accounting for the religious-orientation of the perceivers. Consistent with other perceptual dimensions (e.g., Bobkowski and Kalyanaraman 2010; Galen et al. 2011; Swan et al. 2014; Tan and Vogel 2008), both the religious designation of the target as well as the religious-orientation of the perceiver matter. Whereas non-Christians perceived higher levels of hypocrisy among devout and evangelical Christians, self-identified Christians reserved the highest hypocrisy ratings for casual Christians.
In addition, the results of our study provided strong confirmation that perceptions of religious hypocrisy were based upon a perceived deviation between moral claims and moral behavior, but only in certain cases. This finding is consistent with general models of hypocrisy (Alicke et al. 2013), yet our work is the first application of this model within the domain of religious perceptions. Our findings may also help to explain the commonality of perceptions of religious (specifically Christian) hypocrisy in a sample of non-regular church attenders in a recent study by Vermurlen (2014).
Our reported findings would appear to reflect more than a simple ingroup bias. First, although Christian participants rated casual Christians as higher on direct hypocrisy ratings compared to all other designations, the indirect measure of moral hypocrisy suggests that even Christian participants perceive significantly more moral hypocrisy among Christian designations compared to non-religious designations. Tellingly, these difference scores are due to higher ratings of moral claims compared to moral actions, suggesting that the perceptions of hypocrisy among Christians may be caused by perceptions of excessive moral claims. For this reason, non-Christian groups may view devout and evangelical Christians as increasingly more hypocritical due to the prevalence of moral claims-making.
Limitations and Future Directions
Several limitations of the present study warrant consideration. First, our study used a non-representative internet sample. Although this tactic may limit the generalizability of our findings, it should be noted that the patterns of effects were consistent with previous research, where available, and also with our a priori predictions. Furthermore, although our sample was not truly representative, it was more homogenous than traditional college-student samples tend to be and, for these reasons, has been utilized by other researchers within this domain (Swan et al. 2014). Finally, a primary contribution of this study was to demonstrate the process by which hypocrisy judgments are made. It is not immediately clear why this basic process should vary much from sample to sample.
Regarding this latter point, care should also be taken in the interpretation of our findings given the correlational methodology we employed. We interpreted the association between the moral claims/actual behavior difference scores and perceptions of hypocrisy as being due to a shared underlying dimension. It is possible, however, that the relationship is due to some additional unmeasured variable. Future studies might manipulate the religious orientation of an actor within a given vignette to more fully address theoretical mechanisms and mediators.
Although not assessed in the current work, it is likely that perceptions of hypocrisy would correspond with more negative evaluations and treatment of Christians. It would be particularly interesting to examine such responses in the realm of moral transgressions. As one example, in the domain of criminal justice, (Laurent et al. (2014) demonstrated that perceptions of defendant hypocrisy predicted more negative emotions and punishments on the part of mock jurors. To the extent that the more dogmatic varieties of Christianity, or any religion, are perceived to be hypocritical, and to the extent that moral transgression would increase perceptions of hypocriticality, similarly harsh reactions would be predicted.
It should also be noted that our use of general, abstractly defined religious-designations was not merely a methodological expedient, but was strategic given our aims. We recognize that the idiosyncratic nature of a given individual's religious orientation are likely to affect perceptions of hypocrisy; however, we were more interested in the trait associations that corresponded with abstract social categories.
Finally, it is unclear whether our findings would generalize to other religious affiliations, such as Judaism, Islam, Mormonism, or Buddhism. As with Christianity, each of these faiths entails lofty moral standards; however, other characteristics unique to these religions may influence perceptions of hypocrisy. Hence, future work should seek to include additional varieties of religious faith.
Conclusion
Our findings add to a small but growing literature on perceptions of Christians in terms of morality and hypocrisy. The findings from this study provide insight into why certain Christian groups may be perceived as more hypocritical and demonstrates the gap in perceptions between Christians and non-Christians. Such stark differences in perceptions between Christians and non-Christians may help to explain the corresponding cultural divide that might only be expected to intensify as the proportion of Christians in the US continues to decline (Religion n.d.). Given the prevalence of us versus them framing between the religious and non-religious in the media, additional research into the nature moral perceptions of the religious is clearly merited.