Through the Eyes of Narcissus: Competitive Social Worldviews Mediate the Associations that Narcissism has with Ideological Attitudes

ABSTRACT

The present research examined whether the associations that the assertive/extraverted, antagonistic/disagreeable, and vulnerable/neurotic aspects of narcissism had with ideological attitudes were mediated by social worldviews . The assertive/extraverted and antagonistic/disagreeable aspects of narcissism had similar associations with the competitive social worldview which, in turn, mediated their associations with certain ideological attitudes. In contrast, social worldviews did not mediate the tendency for vulnerable/neurotic narcissism to be characterized by the rejection of right-wing ideological attitudes and the endorsement of left-wing ideological attitudes. These results suggest that these aspects of narcissism diverged in their associations with ideological attitudes and that the competitive social worldview played an important role in the associations that the assertive/extraverted and antagonistic/disagreeable aspects of narcissism had with ideological attitudes.

KEYWORDS:

• Narcissism
• competition
• worldviews
• ideological Attitudes
• right-Wing Authoritarianism
• social Dominance Orientation

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. We chose to frame our analyses for each of the present studies as concerning mediation but they could have been framed as confounding analyses. Mediational analyses and confounding analyses are statistically identical but they have conceptual differences (see MacKinnon et al., 2000, for a review). The most important difference is that confounding analyses do not imply a causal relationship among the variables, whereas mediational analyses do imply causal relationships. We relied on cross-sectional data for each of the present studies which prevented us from determining whether causal relationships existed between the variables so our approach was consistent with confounding analyses. However, we decided to frame our analyses as concerning mediation because the underlying process model that guided these studies was mediational in nature (i.e., narcissistic personality features were believed to contribute to the development of particular social worldviews which, in turn, would play a role in the development of ideological attitudes).

2. Gender differences have consistently been found for narcissism (e.g., Grijalva et al., 2015) and ideological attitudes (e.g., Pratto et al., 1997). Gender has also been shown to moderate the associations that narcissism has with certain outcomes (e.g., Sauls et al., 2019) which led us to conduct preliminary analyses for all three studies that included gender as a potential moderator of the mediational associations that narcissistic personality features had with ideological attitudes through the dangerous and competitive worldviews. However, gender did not moderate the results reported for any of the studies (i.e., there was no evidence of moderated mediation) nor were the reported results significantly altered by the inclusion of gender in those analyses. As a result, we did not include gender in the final analyses nor do we discuss gender differences in the interest of parsimony.

3. The underlying process model that we adopted was that social worldviews mediated the associations that narcissistic personality features had with ideological attitudes. However, we relied exclusively on cross-sectional data so it is certainly possible that other causal patterns exist between these variables. To examine one of these alternative possibilities, we supplemented our parallel multiple mediation analyses with a series of alternative models that switched the positions of the predictors and the potential mediators. That is, narcissistic personality features served as the mediators in these alternative analyses with the dangerous and competitive social worldviews serving as the predictors. The results of these alternative analyses found that narcissistic personality features did not mediate the associations that the dangerous social worldview had with RWA or SDO. In contrast, the competitive social worldview had indirect associations with RWA through narcissistic admiration and narcissistic vulnerability as well as an indirect association with SDO through narcissistic rivalry. Although the results of these analyses provided some support for mediation, the indirect associations that emerged from the original analyses were somewhat stronger in magnitude than the indirect associations that were observed in these alternative analyses. This pattern suggests somewhat stronger evidence for the original model than the alternative model. These alternative analyses are informative but it is important to acknowledge that the utility of this approach has been challenged in recent years (e.g., Thoemmes, 2015).

4. As in Study 1, we supplemented our analyses with a series of alternative models that switched the positions of the predictors and the potential mediators such that narcissistic personality features served as the mediators with the dangerous and competitive social worldviews serving as the predictors. The results of these alternative analyses showed that narcissistic personality features did not mediate the associations that the dangerous social worldview had with RWA or SDO. Further, these analyses revealed that narcissistic personality features did not mediate the association that the competitive social worldview had with RWA but antagonistic narcissism did mediate the association that it had with SDO. Although the results of these analyses provided some limited support for mediation, the indirect associations that emerged from the original analyses were stronger in magnitude than the indirect associations observed in these alternative analyses.

5. As in Studies 1 and 2, we supplemented our analyses with a series of alternative models that switched the positions of the predictors and the potential mediators such that narcissistic personality features served as the mediators with the dangerous and competitive social worldviews serving as the predictors. The results of these alternative analyses showed that narcissistic personality features did not mediate the associations that the dangerous social worldview had with RWA, SDO, LWA, or equalitarianism. In contrast, the competitive social worldview had an indirect association with RWA through extraverted narcissism as well as indirect associations with SDO, LWA, and equalitarianism through antagonistic narcissism. The results of these analyses provided some support for mediation but the indirect associations that emerged from the original analyses were somewhat stronger in magnitude than the indirect associations observed in these alternative analyses.