Collective narcissism and explicit and implicit collective self-esteem revisited: A preregistered replication and extension
Introduction
Collective narcissism gained increased interest in the last decades (for a review, see Golec de Zavala & Lantos, 2020). Collective narcissism is defined as an excessive in-group love contingent on external validation (Golec de Zavala et al., 2009). Religion, political stance, nationality, or any other group may define the in-group (Golec de Zavala & Lantos, 2020). While the classic definition of narcissism focuses on the individual (i.e., individual narcissism; Morf and Rhodewalt, 2001, Raskin and Terry, 1988), collective narcissism extends this concept to similar exaggerated beliefs of a person’s in-group and suggests that a group can function as a narcissistic entity (Golec de Zavala et al., 2009). For example, individuals high in collective narcissism agree that the importance and value of their in-group is not sufficiently recognized by others and that their in-group deserves special treatment (e.g., Dyduch-Hazar et al., 2019, Golec de Zavala and Keenan, 2020).
One influential perspective that has been used to understand the concept of collective narcissism is the mask model of narcissism (Bosson et al., 2008, Kuchynka and Bosson, 2018, Sedikides, 2020). Originally introduced to explain the dynamics of individual narcissism, the mask model postulates that while narcissists hold external grandiose and inflated self-views, they actually suffer from deep-seated feelings of low self-worth (also see Kernberg, 1975). Most typically, such pattern emerges as the discrepancy between (positive) explicit self-esteem and (less positive) implicit self-esteem (i.e., discrepant self-esteem hypothesis; e.g., Gregg & Sedikides, 2010). Whereas explicit self-esteem pertains to deliberately processed self-evaluations that can be measured via self-report, implicit self-esteem pertains to automatically activated self-evaluations that can be measured via indirect measures, such as the Implicit Association Test (IAT; Greenwald et al., 1998, Greenwald and Farnham, 2000). Typically, explicit and implicit self-esteem are not or only weakly related (e.g., Klavina et al., 2012).
Following this conceptualization of individual narcissism, the seminal work of Golec de Zavala et al. (2009, Study 4) proposed, and showed, that the mask model of (individual) narcissism translates to the intergroup domain. Indeed, collective narcissism was shown to be highest among individuals who self-reported a positive regard of their in-group (i.e., high private collective self-esteem; Luhtanen & Crocker, 1992), but at the same time reported that others do not share this positive regard of their in-group (i.e., low public collective self-esteem; Luhtanen & Crocker, 1992). Moreover, collective narcissism was shown to be highest among individuals who self-reported a positive regard of their in-group (i.e., high private collective self-esteem), but at the same time held a less positive implicit evaluation of their in-group (i.e., low implicit collective self-esteem). Finally, explicit and implicit collective self-esteem were not significantly correlated. Hence, collective narcissism describes a combination of apparent group-based feelings of superiority that are masked by an underlying in-group fragility (Golec de Zavala et al., 2009).
Golec de Zavala et al.’s (2009) pioneering work on collective narcissism and explicit and implicit collective self-esteem represents the first empirical attempt to demonstrate a direct link between outward beliefs in one’s in-group greatness that are simultaneously accompanied by low collective self-esteem, as predicted by the mask model. To date, their influential paper has been cited more than 200 times (Google scholar search; September 14, 2021). However, the initial findings have not been re-assessed by an independent research group and are still considered “preliminary” (Golec de Zavala et al., 2009, p. 1087). Moreover, there are at least two concerns related to these findings both methodologically and conceptually: First, the domain of in-group enhancement and, second, the way in which implicit collective self-esteem was assessed.
In the current study, we aimed to test the replicability of the findings in a larger sample (i.e., N = 481 vs. N = 262 of Golec de Zavala et al., 2009, Study 4; for a discussion of power, see Methods section below) and to extend them by adopting the Agency-Communion model of individual narcissism (Gebauer et al., 2012) to the collective domain (Żemojtel-Piotrowska et al., 2021a, Żemojtel-Piotrowska et al., 2021b). We aimed for a conceptual (vs. direct) replication because we sought to probe the validity of the theoretical rationale and hypothesis rather than the reproducibility of a specific methodology (Stroebe & Strack, 2014). To this end, we made two classes of changes to the original study, thereby providing not only a more stringent test of the mask model of collective narcissism but also a fundamental theoretical advance.
First, and most importantly, the aforementioned relation between collective narcissism and collective self-esteem was uniquely examined in the agentic (vs. communal) domain. However, because the collective self can be based both on agentic (e.g., intelligent, ambitious) and communal (e.g., tolerant, helpful) traits (Fiske et al., 2002), collective narcissists can similarly base their inflated in-group views on both agency and communion (Sedikides, 2020, Żemojtel-Piotrowska et al., 2020, Żemojtel-Piotrowska et al., 2021a, Żemojtel-Piotrowska et al., 2021b). Initial empirical evidence suggests that “classic” collective narcissism (as assessed by the Collective Narcissism Scale [CNS]; Golec de Zavala et al., 2009) is predominantly agentic, while communal collective narcissism (as assessed by the recently developed Communal Collective Narcissism Inventory [CCNI]; Żemojtel-Piotrowska et al., 2021a, Żemojtel-Piotrowska et al., 2021b) is a related, yet separate, form of collective narcissism in the communal domain (Żemojtel-Piotrowska et al., 2021a, Żemojtel-Piotrowska et al., 2021b).
Therefore, an important, yet open, question is whether the presumed discrepancy between implicit and explicit collective self-esteem extends to agentic and communal in-group enhancement. As agentic collective narcissists enhance their in-group in the agentic domain (Żemojtel-Piotrowska et al., 2021a, Żemojtel-Piotrowska et al., 2021b), the discrepancy between implicit and explicit collective self-esteem should be most evident on agency. Theoretically, the opposite pattern should emerge for communal collective narcissists. Hence, in the current study, we aimed to supplement the original study of Golec de Zavala et al. (2009) by measuring both agentic and communal collective narcissism (i.e., CNS and CCNI).
Second, and related to the previous point, although we kept the IAT as a measure of implicit collective self-esteem, we strived for a clear conceptual distinction between implicit agentic versus communal in-group views. We thereby answer recent calls for a more precise examination of specific (vs. global) esteem domains when examining the mask model of narcissism (Kuchynka & Bosson, 2018). To this end, we designed two different IATs, one of which assessed implicit collective self-esteem in the agentic domain (i.e., implicit agentic collective self-esteem) and the other in the communal domain (i.e., implicit communal collective self-esteem). As such, we were able to carefully match the domain of in-group enhancement between collective narcissism measures (agentic vs. communal) and measures of implicit collective self-esteem (agentic vs. communal). Furthermore, to obtain parallel measures of explicit collective self-esteem, we asked our participants for explicit ratings of the attribute stimuli used in the IATs. That is, we also assessed explicit agentic and communal collective self-esteem.
In a similar vein, we addressed an important concern related to the outgroups used in original study (e.g., [South] Korea and Indonesia). An important criterion for IAT stimuli selection is that they are both unambiguously classifiable and sufficiently familiar (e.g., Greenwald et al., 1998). However, the original study contrasted “Poland” with a putatively “difficult to identify and unknown […]” (Golec de Zavala et al., 2009, p. 1085) second category. (South) Korea and Indonesia, for example, are very different from Poland on various (i.e., not only cultural; Hofstede et al., 2010) dimensions, thereby obscuring the relative preference for one’s in-group (i.e., Poland). Hence, in the current study, we used European countries (i.e., Estonia, Slovenia, and Latvia), as these countries resemble Poland in terms of cultural values (e.g., individualism-collectivism, see Hofstede et al., 2010), geographical location (Europe) and religion (Christianity). As such, we are able to separate the effects of individual differences in implicit in-group enhancement from other confounds, such as religion or cultural background.1
Section snippets
Methods
The present research (i.e., design, hypotheses, power analysis, and main analyses) was fully preregistered on the Open Science Framework (OSF; https://osf.io/pwjhx). We report how we determined our sample size, all data exclusions (if any), all manipulations, and all measures in the study (Simmons et al., 2012). We also make publicly available Supplemental Online Material, complete data (i.e. Open Data), code / syntax (i.e. Open Code), and materials (i.e., Open Materials) used for the analyses
Measures
Implicit Agentic and Communal Collective Self-Esteem. Implicit agentic and communal collective self-esteem was measured using an adapted version of the individual self-esteem IAT (Greenwald and Farnham, 2000, Greenwald et al., 1998). We created two different implicit collective self-esteem IATs, one that focused on implicit in-group views in the agentic domain (i.e., implicit agentic collective self-esteem) and one that focused on implicit in-group views in the communal domain (i.e., implicit
Descriptive statistics and zero-order correlations
Table 2 details descriptive statistics and zero-order correlations among study variables. Replicating the correlational analyses by Golec de Zavala et al. (2009), agentic collective narcissism was significantly positively related to explicit collective self-esteem (i.e., private, public, membership, and identity). Results concerning communal collective narcissism were highly similar. Importantly, both agentic and communal collective narcissism were significantly positively related to explicit
Discussion
In their original study, Golec de Zavala et al. (2009, Study 4) revealed that agentic collective narcissists seem to dislike their in-group “deep down inside”. Specifically, agentic collective narcissists were shown to have a positive regard of their in-group, but at the same believe that others do not share this positive regard of their in-group. In addition, agentic collective narcissists were shown to possess high explicit and low implicit in-group views. To our knowledge, the current study
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (41)
- et al.
Narcissism and trust: Differential impact of agentic, antagonistic, and communal narcissism
Personality and Individual Differences
(2019) - et al.
Towards an operationalization of the fundamental dimensions of agency and communion: Trait content ratings in five countries considering valence and frequency of word occurrence
European Journal of Social Psychology
(2008) Validation of the Polish version of the Collective Self-Esteem Scale
Current Issues in Personality Psychology
(2015)- et al.
Implicit attitudes towards homosexuality: Reliability, validity, and controllability of the IAT
Zeitschrift für experimentelle Psychologie
(2001) - Bosson, J. K., Lakey, C. E., Campbell, W. K., Zeigler‐Hill, V., Jordan, C. H., & Kernis, M. H. (2008). Untangling the...
- et al.
Implicit self-esteem: Nature, measurement, and a new way forward
Journal of Personality and Social Psychology
(2011) - et al.
Accuracy of effect size estimates from published psychological experiments involving multiple trials
The Journal of General Psychology
(2011) - Campbell, W. Keith, Bosson, Jennifer K., Goheen, Thomas W., Lakey, Chad E., Ker- nis, Michael H., 2007. Do narcissists...
- Dyduch-Hazar, K., Mrozinski, B., & Golec de Zavala, A. (2019). Collective narcissism and in-group satisfaction predict...
- et al.
Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses
Behavior Research Methods
(2009)
A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition
Journal of Personality and Social Psychology
Communal narcissism
Journal of Personality and Social Psychology
Agency-communion and self-esteem relations are moderated by culture, religiosity, age, and sex: Evidence for the “self-centrality breeds self-enhancement” principle
Journal of Personality
Collective narcissism and its social consequences
Journal of Personality and Social Psychology
Collective narcissism as a framework for understanding populism
Journal of Theoretical Social Psychology
Using the implicit association test to measure self-esteem and self-concept
Journal of Personality and Social Psychology
Narcissistic fragility: Rethinking its links to explicit and implicit self-esteem
Self and Identity
Cited by (1)
Effects of brand community social responsibility: roles of collective self-esteem and altruism
2023, Journal of Brand Management