Abstract
For LGBTQ+ community members, one way to cope with the discrimination they experience is through a stronger ingroup identity. However, not all types of ingroup identity may be equally beneficial to LGBTQ+ individuals. A longitudinal (N = 1,044) and a cross-sectional (N = 8,464) study among LGBTQ+ people in Poland demonstrated that collective narcissism was a positive predictor of group-based anger (Study 2) and had a positive reciprocal relationship with group relative deprivation (GRD; Study 1), however, it was negatively related to life satisfaction and exhibited a stronger positive link with nonnormative than normative collective action. Secure LGBTQ+ identification was not longitudinally predicted by GRD (Study 1) and showed a weaker positive association with group-based anger (Study 2). It had a reciprocal positive relationship with life satisfaction and was a stronger predictor of normative than nonnormative collective action. These results show that whereas secure ingroup identity is a clearly positive coping mechanism, the effects of collective narcissism are mixed.
Introduction
Luckily, the red plague [i.e., communism] no longer walks our land, but it does not mean that there is no new plague that wants to take over our souls, hearts, and minds. Not a Marxist one, but born of the same spirit, neo-Marxist. Not red, but a rainbow plague. (Chadwick, 2019)
These words were a part of a homily that archbishop Marek Jędraszewski delivered during a mass commemorating the 75th anniversary of the Warsaw Uprising in the summer of 2019. And it is just one of many examples of discrimination and dehumanization that sexual and gender minorities face in today’s Poland.
One factor that may facilitate coping with belonging to a disadvantaged social group is a strong ingroup identity (e.g., Branscombe et al., 1999). Research shows that ingroup identity may act as a protective factor (i.e., as a social cure, see Haslam et al., 2018) because it is associated with social support and a sense of belonging (e.g., Meyer, 2015). As postulated in the minority stress model (Meyer, 2003), ingroup identity may ameliorate the negative impact of minority stressors by providing affiliation opportunities and access to community resources. Highly identified lesbian, gay, bisexual, trans* and/or otherwise nonheterosexual (LGBTQ+) individuals are also more likely to engage in collective action (i.e., activities undertaken by group members to achieve their group’s political goals; van Zomeren, 2016, p. 89) on behalf of their ingroup (e.g., Górska et al., 2017), which may contribute to improving the situation of the group as a whole.
Despite these positive outcomes, ingroup identity may also have a negative impact on LGBTQ+ individuals. Discrimination and rejection experienced by the representatives of this group lead to negative mental and physical health outcomes (Meyer, 2015; Meyer & Frost, 2013). These outcomes may be amplified among LGBTQ+ individuals who are more strongly committed to their group (Meyer, 2003). For instance, a meta-analysis of LGBTQ+ health outcomes found that stronger LGBTQ+ identity is positively associated with substance use (Lawley, 2019).
What most extant research on the effects of ingroup identity for members of the LGBTQ+ community has overlooked is the fact that ingroup identity is not a unitary construct (for an exception, see Reimer et al., 2017). Recent evidence and theorizing show that there are at least two qualitatively different ways of being identified with one’s social group: collective narcissism and secure (non-narcissistic) ingroup identity (e.g., Cichocka & Cisłak, 2020; Golec de Zavala & Lantos, 2020). Collective narcissism entails grandiose, but insecure attachment to one’s social group that requires constant external validation and has been linked to maladaptive psychological traits and states (e.g., lower personal control; Cichocka et al., 2018; Marchlewska et al., 2019). By contrast, secure ingroup identity is a confident, positive evaluation of one’s group, independent of the opinions of others, and has adaptive psychological concomitants (e.g., higher self-esteem; Golec de Zavala et al., 2020). Collective narcissism and secure ingroup identity differ in terms of their antecedents and consequences (Cichocka, 2016). However, it is important to note that the extant research was conducted almost exclusively with majority groups members. We argue that this differentiation should be taken into account when analyzing the role of identity in shaping intragroup processes among disadvantaged groups as well.
In this contribution, we examine individual- and group-level correlates of collective narcissism and secure ingroup identity among sexual and gender minorities. We propose that while secure ingroup identity benefits both individual well-being (operationalized as life satisfaction) and group interests (via willingness to engage in collective action on behalf of one’s group), the effects of collective narcissism are less homogenous. Specifically, we expect that collective narcissism would prompt LGBTQ+ individuals to take actions to improve the conditions of their entire group (via both normative and nonnormative collective action; see Tausch et al., 2011) and, simultaneously, relate to lower individual life satisfaction (see Golec de Zavala, 2019).
Ingroup Identity and Psychological Well-Being
People have a strong need to belong (Maslow, 1943), which finds its expression in the fact that social identity, that is, seeing oneself as a member of a social group(s), forms a significant part of the self (Tajfel & Turner, 1986). Belonging to social groups is associated with a multitude of positive outcomes. It contributes to people’s sense of who they are (Tajfel & Turner, 1986), satisfies the need for meaningful social connections (Baumeister & Leary, 1995), and is a source of meaning and purpose in life (Vignoles et al., 2006). In addition, ingroup identity has recently been recognized as a key factor contributing to psychological well-being and physical health (Haslam et al., 2009; Jetten et al., 2012). This social cure approach recognizes that since an important part of peoples’ self-concept is derived from their group memberships, these memberships are consequential for mental and physical health (Jetten et al., 2014). 1 For example, identifying with social groups buffers the negative effects of stress (Haslam et al., 2018), and is positively related to recovery from addiction (Best et al., 2016).
Even identities that are devalued by the majority of society (e.g., sexual and gender minority identities) may provide tangible benefits to group members (e.g., Kertzner et al., 2009). In fact, the Rejection Identification Model posits that stronger group identity not only makes discrimination faced by one’s disadvantaged group more self-relevant but might also constitute an adaptive strategy to cope with such discrimination (Branscombe et al., 1999; Doyle & Molix, 2014; Stronge et al., 2016). These assumptions were corroborated among members of sexual and gender minorities, whose perceptions of discrimination were associated with stronger group identification and, in turn, higher self-esteem (Bourguignon et al., 2020; Doyle & Molix, 2014). In addition, positive LGB identity, which captures positive experiences related to identifying as an LGB person (e.g., authenticity), was shown to correlate positively with well-being and the sense of social safeness (Petrocchi et al., 2020).
At the same time, belonging to a disadvantaged social group (e.g., identifying as an LGBTQ+ person) may relate to decreased well-being (e.g., Stronge et al., 2016). This is because for disadvantaged group members the benefits of group belonging may be curbed by the costs associated with being discriminated against by the majority (Branscombe et al., 1999; Meyer, 2015; Stronge et al., 2016). And since sexual and gender minorities face discrimination in all areas of life (e.g., Rosenbaum et al., 2021; Yılmaz & Göçmen, 2016), these pervasive negative experiences are consequential for their well-being, especially among strongly identified individuals. Research shows that strong disadvantaged identity relates to a host of negative outcomes for health and well-being (Jetten et al., 2017). For instance, sexual minorities have significantly higher risk of suicide and depression than the general population (Plöderl et al., 2014; see also minority stress model, e.g., Meyer, 2003; Meyer et al., 2021). Moreover, negative LGB identity is associated with higher levels of stress and anxiety (Flanders, 2015).
Ingroup Identity, Relative Deprivation, and Collective Action
Besides being consequential for psychological and physical well-being, ingroup identity plays a fundamental role in shaping people’s willingness to engage in activities aimed at improving the situation of their ingroup (i.e., collective action; van Zomeren, 2016). Such activities may be either normative (i.e., conform to the norms of the social system, e.g., peaceful demonstration, signing petitions) or nonnormative (i.e., violate norms, e.g., violent protests, sabotage; see Becker & Tausch, 2015). In this literature, ingroup identity is considered the central prerequisite of engagement among advantaged and disadvantaged groups (van Zomeren et al., 2008; Jost et al., 2017)).
According to the Social Identity Model of Collective Action (SIMCA; van Zomeren, 2016), which received strong meta-analytical support (van Zomeren et al., 2008; see also Agostini & van Zomeren, 2021), ingroup identity promotes collective action not only directly, but also indirectly by increasing group efficacy and perceptions of injustice (for the longitudinal test of SIMCA, see Thomas et al., 2020). 2 Specifically, people are more likely to engage in normative and nonnormative collective action when they recognize their ingroup’s disadvantaged status as illegitimate and experience the related feelings of group (as opposed to individual; see Runciman, 1966; Smith et al., 2012) relative deprivation (GRD; Agostini & van Zomeren, 2021; Jost et al., 2017; Zubielevitch et al., 2020) or anger (van Zomeren et al., 2008). Although often studied separately, some authors argue that GRD consists of both: perceptions of injustice and the associated feelings of anger and resentments (Grant et al. 2015). In the current work we measured both of these aspects: GRD in Study 1 and injustice-related anger in Study 2. Importantly, existing research shows that GRD and/or perceptions of injustice that befall the ingroup are predicted by ingroup identity (e.g., Abrams & Grant, 2012) but also predict such identity (e.g., Thomas et al., 2012). Emerging longitudinal evidence suggests that the relation between identification and GRD may in fact be reciprocal (i.e., both strengthen the other longitudinally), but the effect of identity on GRD is stronger than vice versa (Zubielevitch et al., 2020).
The positive association between ingroup identity and collective action has been found in studies with sexual and gender minorities (e.g., Górska & Bilewicz, 2015; Górska et al., 2017; Reimer et al., 2017, Study 1a). Similar effects were also registered for politicized LGBTQ+ identity, defined as identification with the LGBTQ+ rights movement (see Simon & Klandermans, 2001). Gay men and lesbian women politicized identity related positively to diverse collective action behaviors both in cross-sectional (Swank & Fahs, 2013, 2016) and longitudinal (Stürmer & Simon, 2004) research.
Collective Narcissism and Secure Ingroup Identity
Notwithstanding the established links between strength of identity, collective action engagement, and well-being among disadvantaged social groups, the extant research did not take into consideration (for an exception, see Górska et al., 2020) that positive attitudes toward the ingroup may take various shades. Specifically, people may relate to their groups in at least two qualitatively different ways: through collective narcissism and secure ingroup identification (Cichocka & Cisłak, 2020).
The essence of collective narcissism lies in a belief that one’s own group is exceptional but not sufficiently recognized by others (Golec de Zavala & Lantos, 2020). As such, narcissism—both at the individual and group level—is a way for people to compensate for frustrated needs (Adorno, 1963/1998; Cichocka et al., 2018; Fromm, 1973; Jasko et al., 2020). Consequently, research conducted to date shows that the primary sources of collective narcissism lie in low self-esteem (Golec de Zavala et al., 2020), lack of control (Cichocka et al., 2018), and GRD (Golec de Zavala et al., 2021; Marchlewska et al., 2018).
Regarding individual well-being, collective narcissism emerged as significantly and uniquely related to negative emotionality, low social connectedness, low gratitude, self-criticism, and sensory processing sensitivity—a genetically determined exaggerated response to negative stimuli (Golec de Zavala, 2019). By contrast, secure ingroup identity showed consistent positive associations with positive emotionality, social connectedness, and self-compassion (Bagci et al., 2021; Golec de Zavala, 2019).
In the realm of intergroup relations, collective narcissism is related to constant vigilance for any threats to the ingroup (e.g., Golec de Zavala & Cichocka, 2012). Consequently, collective narcissism relates to negative attitudes towards outgroups (e.g., Górska et al., 2022), and perceiving them as threatening and conspiring against the ingroup (Cichocka & Cisłak, 2020; Marchlewska et al., 2019). Crucially, it also predicts one’s willingness to engage in violent actions against other groups (Górska et al., 2022; Jasko et al., 2020).
By contrast, secure ingroup identity involves a positive evaluation of the ingroup without the need for external validation (Cichocka & Cisłak, 2020; Golec de Zavala & Lantos, 2020). It is based on a strong individual self, and satisfaction of collective and individual needs (Cichocka et al., 2018; Marchlewska et al., 2018). Due to the low sense of external threat, secure ingroup identity relates to positive intergroup attitudes (Marchlewska et al., 2020) and support for the ingroup (Cichocka, 2016). Because no separate scale tapping into secure ingroup identification (but not collective narcissism) has been created, the effects of secure ingroup identification are obtained by factoring out variance shared with collective narcissism from the general measures of ingroup identification (e.g., Golec de Zavala et al., 2013). As such, the unique effect of ingroup identification when collective narcissism is accounted for is considered an indicator of secure ingroup identity. 3 So far, the sources and consequences of narcissistic and secure ingroup identity have been investigated mostly in the context of advantaged groups, such as host societies (Dyduch-Hazar et al., 2019), religious majority members (Marchlewska et al., 2019), men (Marchlewska et al., 2022), or heterosexuals (Górska et al., 2020). However, it is members of the disadvantaged social groups who seem especially likely to develop narcissistic identification with their ingroups. Specifically, in comparison to the advantaged groups’ members, the disadvantaged experience higher frustration of psychological needs (e.g., van der Leij, 2020) and GRD (e.g., Moore, 2003), both of which were demonstrated to positively predict collective narcissism (e.g., Cichocka et al., 2018; Marchlewska et al., 2018). Moreover, external recognition and respect—two elements collective narcissists strive for the most—are of greater importance to the members of low- than high-status groups (Shnabel et al., 2008). Lastly, as collective narcissism constitutes a blend of high ingroup regard and a negative meta-perception—the belief that outgroup members hold an unfavorable image of one’s ingroup (e.g., Lees & Cikara, 2020)—it may be particularly pronounced among those who exhibit strong identification with their disadvantaged ingroup and are simultaneously aware of prejudice toward its members.
The Current Research
Based on the above, we formulated six hypotheses regarding the associations between GRD (or group-based anger), different types of ingroup identity, well-being, and collective action support among the LGBTQ+ community members. Specifically, we hypothesized that narcissistic and secure ingroup identity should predict the well-being of the low-status groups’ members in opposite ways. As collective narcissism, but not secure ingroup identity, is associated with hypersensitivity to negative stimuli and strong negative emotionality (Golec de Zavala, 2019), it should exert a negative effect on well-being among members of low-status groups. On the other hand, secure ingroup identity, which is known to foster positive emotionality (Bagci et al., 2021; Golec de Zavala, 2019), should affect the disadvantaged groups’ well-being positively (which is in line with the social cure approach; Haslam et al., 2018). As such, distinguishing between the two types of ingroup identity may reconcile the conflicting findings on the association between identification with the LGBTQ+ community (and other low-status groups) and physical as well as psychological health (Bagci et al., 2020; Bourguignon et al., 2020; Lawley, 2019).
Both types of ingroup identity entail positive evaluations of the ingroup and should positively predict engagement in collective action that benefits the ingroup (van Zomeren et al., 2008). However, there are also reasons to expect that collective narcissism may be a stronger predictor than secure ingroup identity, particularly when it comes to nonnormative collective action (Becker & Tausch, 2015; Górska et al., 2022; Jasko et al., 2020). Specifically, collective action has the potential to change power relations and elevate an ingroup’s status in society (Jost et al., 2017). Thus, successful collective action may confirm the ingroup’s unique value that collective narcissists believe in (Golec de Zavala et al., 2009).
Across two studies (one longitudinal and one cross-sectional), we examined three key hypotheses 4 :
H1. Life satisfaction will be predicted a) negatively by collective narcissism and b) positively by secure ingroup identity.
H2. Collective action (both normative and nonnormative) will be predicted positively by a) collective narcissism and b) secure ingroup identity.
H3. The effects of collective narcissism on life satisfaction and collective action will be partially mediated via increased GRD (or group-based anger).
In addition to these key hypotheses, we were also interested in testing the possible reciprocal relations between perceptions of group-based injustice (operationalized as GRD in Study 1 and group-based anger in Study 2) and collective narcissism (see Zubielevitch et al., 2020). We expected that:
H4. Collective narcissism would be a stronger positive predictor of group relative deprivation (or group-based anger) than secure ingroup identity.
H5. Group relative deprivation (or group-based anger) would be a stronger positive predictor of collective narcissism than secure ingroup identity.
H6. GRD (or group-based anger) will be a negative predictor of life satisfaction and a positive predictor of collective action, and these effects will be partially mediated via increased collective narcissism.
Sociopolitical Context
We conducted our research in Poland, a country no longer classified as a full democracy (Wójcik & Wiatrowski, 2020). The process of Poland’s sliding away from liberal democracy began in 2015 when a conservative Law and Justice party (Prawo I Sprawiedliwość; PiS) came to power. The ideological program presented by this political grouping involves, among others, the fight against the party-identified “enemies of the Polish nation” (Mole et al., 2021). This was first evidenced by a hate campaign against Muslim refugees in 2015 and followed up with a campaign against the LGBTQ+ community in Poland (Żuk & Żuk, 2020). The examples of inciting hatred against LGBTQ+ people include archbishop Jędraszewski’s homily cited above (Chadwick, 2019) but also Poland’s president Andrzej Duda’s statement that LGBT people are “not people, but ideology” as well as the fact that by the end of 2020 a number of municipalities, boroughs, and regions declared themselves as “LGBT-free zones” (ILGA Europe, 2021; Żuk & Żuk, 2020). As shown recently (Górska & Tausch, 2022), the political discourse centered around depicting LGBTQ+ individuals as a threat to the national and Catholic moral code (Mole et al., 2021) is associated with the increase in prejudice toward gay people.
While Study 1 was conducted at the turn of 2016 and 2017, data collection in Study 2 began in late 2020 and ended in early 2021. Even though Study 1 had been carried out before the beginning of the full-fledged anti-LGBTQ+ campaign (Górska & Tausch, 2022; Żuk & Żuk, 2020), the situation of Polish sexual and gender minorities was grim even then. According to the annual report by ILGA Europe, Poland is the most homophobic country in the European Union (ILGA Europe, 2021). The legal system not only denies LGBTQ+ people rights to same-sex marriage and civil unions but also does not provide this group with hate crime protection.
Study 1
Method
Participants and procedure
Study 1 was a part of a larger national survey of LGBTQ+ individuals (Świder & Winiewski, 2017) conducted online by the Campaign Against Homophobia, Lambda Warsaw Association, and Trans-Fuzja Foundation—the largest LGBTQ+ organizations in Poland. The link to the first part of the study was distributed through listservs, LGBTQ+-related websites, and dating apps. The informed consent form presented prospective participants with information about the goal of the study, the confidentiality of the provided data, data management, and the possibility of not answering specific questions or withdrawing from the study at any point. The final screen provided contact to a mental health helpline for those who experienced distress due to study participation.
Time 1 data was collected between November 2016 and February 2017. Individuals who participated in the first measurement were asked to provide an email address so that they could be later invited to fill out the second measurement questionnaire. Time 2 measurement was carried out in June 2017.
Out of 7,672 adult LGBTQ+ individuals 5 who accessed the questionnaire at Time 1, 3,143 (40.97%) provided their email addresses to be contacted about the second measurement. The final sample consisted of N = 1,044 (33.22% of those who left their email addresses) individuals who completed the questionnaire at both measurement times (for a comparison of participants who did vs. did not enter the second measurement, see the Online Supplemental Material). Based on their responses to sexual orientation and gender identity questions participants were categorized to gay men (51.8%), bisexual women (16.8%), lesbians (16.7%), trans* persons (7.8%), bisexual men (4.8%), and queer persons (2.2%). There were no asexual individuals in the group of adults who took part in both measurements. Participants’ age ranged from 18 to 96.
Measurement
The measures presented below were embedded in a longer questionnaire that included scales assessing diverse psychological constructs (e.g., internalized homophobia; Herek et al., 1997). Unless otherwise noted, all measures used a response scale ranging from 1 (strongly disagree) to 7 (strongly agree).
Ingroup identification
Three items from Cameron’s (2004) scale were used to measure ingroup identification: “I have a lot in common with other LGBTA 6 people,” “In general, being an LGBTA person is an important part of my self-image,” and “In general, I’m glad to be an LGBTA person,” αT1 = .79, αT2 = .80.
Collective narcissism
Collective narcissism with LGBTA people as a reference category was gauged with the 5-item version of the Collective Narcissism Scale (Golec de Zavala, Cichocka, & Bilewicz, 2013; sample item: “It really makes me angry when others criticize LGBTA people”), αT1 = .79, αT2 = .79.
Group relative deprivation
To measure GRD, two questions were used (see Górska & Bilewicz, 2015): “In comparison to the heterosexual and cisgender majority, the situation of LGBT people is worse,” and “I am dissatisfied with the fact that the situation of LGBT people is worse in comparison to the heterosexual and cisgender majority,” rT1 = .82, p < .001, rT2 = .83, p < .001.
Life satisfaction
Life satisfaction was assessed using the 5-item Satisfaction With Life Scale (SWLS; Diener et al., 1985; sample item: “If I could live my life over, I would change almost nothing”). The measure employed a response scale ranging from 1 (strongly disagree) to 5 (strongly agree), αT1 = .87, αT2 = .88.
Normative collective action
To measure normative collective action intentions, we asked respondents how likely they were (1 = very unlikely, 7 = very likely) to engage in three activities (i.e., petition signing, attending a demonstration, and distributing informational materials) to improve the situation of LGBT people in Poland, αT1 = .69, αT2 = .71.
Covariates
Covariates involved the LGBTQ subgroup (1 = gay man, 2 = lesbian, 3 = bisexual woman, 4 = bisexual man, 5 = transgender person, 6 = queer person), age, years of full-time education, settlement size (1 = rural area, 2 = town up to 20,000 residents, 3 = town between 20,001 and 50,000 residents, 4 = town between 50,001 and 100,000 residents, 5 = town between 100,001 and 500,000 residents and 6 = city above 500,000 residents), subjective economic situation in comparison to other members of society (1 = the lowest income, 10 = the highest income) and political orientation (1 = left, 10 = right).
Results
Analytic strategy
To verify our theorizing, we tested a series of saturated autoregressive cross-lagged panel models (Selig & Little, 2012) using Maximum Likelihood (ML) estimation. As our data violated the multivariate normality assumption, all 95% CIs were obtained with bootstrapping (10,000 re-samples). Model 1 (Figure 1) was intended to test H1 and H2 and involved four variables (i.e., secure ingroup identity, collective narcissism, life satisfaction, and normative collective action intentions) assessed across two measurements. Each variable measured at Time 2 was regressed on all variables assessed at Time 1. The aim of Model 2 (Figure 2) was to test the remaining hypotheses. The only difference in comparison to Model 1 relied on adding GRD as an exogenous variable at Time 1 and an endogenous variable at Time 2. To verify hypotheses about reciprocal relations between collective narcissism and GRD, we estimated four half-longitudinal indirect effects (Cole & Maxwell, 2003), each of which was operationalized as a product of two parallel paths. For example, to verify H3, we multiplied the path from Time 1 collective narcissism to Time 2 GRD by the path from Time 1 GRD to Time 2 normative collective action.
Over-time effects of secure ingroup identity, collective narcissism, life satisfaction, and normative collective action (Study 1, Model 1).
Over-time effects of group relative deprivation, secure ingroup identity, collective narcissism, life satisfaction, and normative collective action (Study 1, Model 2).
To check the robustness of our results, we added covariates to Models 1 and 2 (the LGBTQ subgroup was coded as five dummy variables with gay men as a reference category). Missing data (3.75%) was handled with Full Information Maximum Likelihood (FIML) estimation. Across all models presented in this contribution, specific paths were compared with the Wald test. All analyses reported below were carried out with R lavaan (Rosseel, 2012). Data and code necessary to replicate present results may be downloaded from the Open Science Framework repository: https://osf.io/mnwds/?view_only=d0fe4714ee544cfab323aa5f8fd26f46.
Hypotheses testing
Means, standard deviations, and intercorrelations among the variables of interest are presented in Table 1. 7 As shown by the results of Model 1 (Figure 1), Time 2 life satisfaction was predicted positively by Time 1 secure ingroup identity, B = 0.04, 95% CI [0.01, 0.07], and negatively by Time 1 collective narcissism, B = -0.05, 95% CI [-0.09, -0.01]. As the difference between these effects was significant, χ2(1) = 7.68, p = .006, H1 received strong support. Time 1 collective narcissism, B = 0.09, 95% CI [0.02, 0.17], but not Time 1 secure ingroup identity, B = 0.05, 95% CI [-0.02, 0.12], was a positive predictor of Time 2 normative collective action. As such, H2 received only partial support. The effects of secure ingroup identity and collective narcissism did not differ significantly, χ2(1) = 0.50, p = .482 and the results changed slightly when we accounted for the covariates. 8
Means, standard deviations and correlation coefficients (Study 1).
Note. SES = Subjective economic status. Ns ranged from 893 to 1,044.
p < .05. **p < .01. ***p < .001.
To test for the presence of indirect effects we multiplied the path from Time 1 collective narcissism to Time 2 GRD by the path from Time 1 GRD to Time 2 life satisfaction. This effect did not reach significance, IE = -0.001, SE = 0.01, 95% CI [-0.01, 0.01]. By contrast, the indirect effect constructed as a product of the path from Time 1 collective narcissism to Time 2 GRD and the path from Time 1 GRD to Time 2 normative collective action was positive and significant, IE = 0.01, SE = 0.01, 95% CI [0.01, 0.02]. Thus, H3 was partially supported by collective narcissism having a positive indirect effect on normative collective action via increased GRD.
Testing our additional hypotheses, Model 2 (Figure 2) revealed that Time 1 collective narcissism, B = 0.10, 95% CI [0.02, 0.18], but not Time 1 secure ingroup identity, B = -0.01, 95% CI [-0.08, 0.06], served as a positive longitudinal predictor of Time 2 GRD, which was consistent with H4. Nevertheless, the difference between these effects did not reach significance, χ2(1) = 3.07, p = .080. Partially supporting our predictions (H5), Time 1 GRD was a positive predictor of Time 2 collective narcissism, B = 0.06, 95% CI [0.01, 0.11], but not Time 2 secure ingroup identity, B = 0.03, 95% CI [-0.03, 0.09]. However, the difference between these effects was not significant, χ2(1) = 0.81, p = .368. At the same time, the path from Time 1 GRD to Time 2 collective narcissism did not differ from the path from Time 1 collective narcissism to Time 2 GRD, χ2(1) = 0.68, p = .408. Thus, the association between GRD and collective narcissism was reciprocal.
Analyses also showed that an indirect effect built by multiplying the path from Time 1 GRD to Time 2 collective narcissism and the path from Time 1 collective narcissism to Time 2 life satisfaction was negative and significant, IE = -0.003, SE = 0.002, 95% CI [-0.01, -0.0003]. In accordance with our expectations (H6), GRD exerted a negative indirect effect on life satisfaction by increasing collective narcissism. Contrary to expectations (H6), the indirect effect created as the product of the path from Time 1 GRD to Time 2 collective narcissism and the path from Time 1 collective narcissism to Time 2 normative collective action was not significant, IE = 0.003, SE = 0.01, 95% CI [-0.01, 0.01]. As such, GRD did not translate into normative collective action by increasing collective narcissism.
Model 2 revealed several associations that were not a part of our theorizing. First, life satisfaction was a positive over-time predictor of secure ingroup identity, B = 0.11, 95% [0.03, 0.18], but not of collective narcissism, B = 0.04, 95% [-0.03, 0.10]. At the same time, the path from Time 1 life satisfaction to Time 2 secure ingroup identity did not differ in its direction or magnitude from the path from Time 1 secure ingroup identity to Time 2 life satisfaction, χ2(1) = 2.34, p = .126. This pattern of results suggests that the association between secure ingroup identity and life satisfaction is, in fact, reciprocal. It seems that secure ingroup identity not only fosters psychological well-being (see Golec de Zavala, 2019) but also derives from it.
Second, although life satisfaction at Time 1 had a positive longitudinal effect on normative collective action at Time 2, B = 0.14, 95% CI [0.05, 0.24], the reverse path from Time 1 normative collective action to Time 2 life satisfaction did not reach significance, B = 0.003, 95% CI [-0.03, 0.03]. As the difference between these two effects was significant, χ2(1) = 8.60, p = .003, the current results suggest that life satisfaction precedes normative collective action in the causal order. Interestingly, the indirect effect built by multiplying a path from Time 1 collective narcissism to Time 2 life satisfaction and the path from Time 1 life satisfaction to Time 2 normative collective action was negative and significant, IE = -0.01, SE = 0.004, 95% CI [-0.02, -0.001]. As such, there was a negative indirect effect of collective narcissism on normative collective action by decreased life satisfaction. At the same time, the indirect effect constructed as the product of the path from Time 1 secure ingroup identity to Time 2 life satisfaction and the path from Time 1 life satisfaction to Time 2 normative collective action was positive and significant, IE = 0.01, SE = 0.003, 95% CI [0.001, 0.01]. Thus, by promoting life satisfaction, secure ingroup identity exerted a positive indirect effect on normative collective action.
Finally, normative collective action measured at Time 1 had a positive effect on GRD at Time 2, B = 0.09, 95% CI [0.04, 0.15], and the opposite path from Time 1 GRD to Time 2 normative collective action was also positive and significant, B = 0.08, 95% CI [0.02, 0.15]. As the two effects did not differ significantly, χ2(1) = 0.03, p = .865, this result is consistent with the dynamic models of collective action, according to which protest engagement not only stems from but also promotes the sense of injustice (Becker & Tausch, 2015; van Zomeren et al., 2012; for empirical evidence see Becker et al., 2011).
Accounting for the covariates slightly changed the results. While H1, H2, and -H5, and parts of H6 remained supported by the data, the partially supported H3 did not hold anymore. In particular, the indirect effect of collective narcissism on collective action via increased GRD ceased being significant, IE = 0.01, SE = 0.01, 95% CI [-0.01, 0.02].
Discussion
Study 1 demonstrated that the two types of ingroup identity among LGBTQ+ individuals differ regarding antecedents and outcomes. GRD had a positive over-time effect on collective narcissism, but not on secure ingroup identity. At the same time, collective narcissism served as a negative longitudinal predictor of life satisfaction and a positive longitudinal predictor of GRD. In contrast, secure ingroup identity exerted a positive longitudinal effect on life satisfaction and did not predict GRD.
Analysis of indirect effects revealed that collective narcissism mediated the negative effect of GRD on life satisfaction. On the other hand, the positive indirect effect of collective narcissism on normative collective action was mediated by increased GRD. As such, the two indirect effects differed in the relative position of GRD and collective narcissism. This pattern of results suggests that, even when the same variables (i.e., collective narcissism and GRD) are involved as a focal predictor and a mediator, the processes underlying life satisfaction and normative collective action among LGBTQ+ individuals are distinct.
One limitation of Study 1 is that it relied on measuring solely normative forms of collective action. Meanwhile, recent events such as destroying a van with homophobic propaganda by a group of queer activists (Metcalfe, 2020) suggest that the Polish LGBTQ+ movement has developed a more radical flank (Tompkins, 2015). Although the instances of nonnormative collective action were practically absent just a few years ago, they have become a part of the protest repertoire under the Law and Justice government. 9
Given that nonnormative collective action originates from extremely negative emotions (e.g., contempt) toward an outgroup (Shuman et al., 2016; Tausch et al., 2011) and that collective narcissism demonstrates close association with outgroup-directed prejudice (e.g., Golec de Zavala, Cichocka, & Bilewicz, 2013), it is reasonable to expect that, in comparison to its normative counterpart, nonnormative collective action would be better predicted by collective narcissism (see Górska et al., 2022). At the same time, secure ingroup identity was shown to decrease rather than promote outgroup hostility (e.g., Cichocka et al., 2018). Thus, besides hoping to replicate results of Study 1, in Study 2 we expected that nonnormative collective action will be better predicted by collective narcissism than secure identity, whereas the opposite will be true for its normative counterpart (H7).
Study 2
Method
Participants and procedure
Study 2 used data collected as part of the 2019–2020 edition of the survey introduced in Study 1. Data was collected between November 2020 and February 2021. Similar to Study 1, convenience sampling was employed: the link to the questionnaire was distributed through Polish LGBTQ+ organizations, LGBTQ+-related websites, forums, groups, and dating apps. The informed consent form was similar to that used in Study 1.
The final sample comprised N = 8,464 10 adults who completed the survey. Based on their responses to sexual orientation and gender identity questions, participants were divided into seven categories: gay men (30%), bisexual women (25.7%), lesbians (17.6%), trans* individuals (15.3%), asexual individuals (4.5%), bisexual men (4.4%), and queer individuals (2.5%). Respondents’ age ranged from 18 to 100 (M = 26.07, SD = 7.61).
Measurement
Similar to Study 1, Study 2 was a part of a larger questionnaire that included the scales of diverse psychological constructs (e.g., system justification; Kay & Jost, 2003). Ingroup identification (α = .77), collective narcissism (α = .75), and life satisfaction (α = .86) were assessed exactly as in Study 1.
Group-based anger. 11
Group-based anger was measured with a single question: “Discrimination of LGBTQ people in Poland makes me angry” (1 = Strongly disagree, 7 = Strongly agree).
Support for nonnormative collective action
As people may be reluctant to declare their intentions to engage in nonnormative and often illegal forms of collective action, we asked participants to report their endorsement of different forms of engagement. To assess nonnormative collective action, respondents were asked to indicate the extent to which they support three activities: displaying LGBTQ+ symbols on the objects of religious worship, organizing demonstrations that allow the use of violence (e.g., attacking counterdemonstrators or the police), and damaging vans carrying homophobic slogans (1 = I do not support it all, 7 = I fully support it), α = .75.
Support for normative collective action
To assess the endorsement for normative forms of collective action participants were asked to what extent they support the organization of peaceful demonstrations and parades (e.g., the Equality Parade), displaying LGBTQ+ symbols in the public space, and distributing information leaflets about LGBTQ+ people (1 = I do not support it all, 7 = I fully support it), α = .78. 12
As shown by the principal axis EFA, normative and nonnormative collective action items created two separate factors that accounted for 42.01% and 13.08% of variance, respectively (see the Online Supplemental Material).
Covariates
Covariates involved LGBTQ+ subgroup, age, education, settlement size, subjective economic status, and political conservatism. The only difference in measurement concerned the LGBTQ+ subgroup, which had seven instead of six categories (1 = lesbian, 2 = gay man, 3 = bisexual woman, 4 = bisexual man, 5 = asexual person, 6 = trans* person, 7 = queer person). Before accounting for the LGBTQ+ subgroup in our analyses, we recoded it into six dummy variables with gay men as a reference category.
Results
Analytic strategy
Four saturated path models were evaluated using ML estimation. Again, to account for the violation of multivariate normality, 95% CIs were obtained with bootstrapping (10,000 re-samples). In Model 1 (Figure 3, Panel A), we regressed life satisfaction on group-based anger. In Model 2 (Figure 3, Panel B), collective narcissism and secure ingroup identity were added as parallel mediators. Model 3 (Figure 4, Panel A) included support for normative and nonnormative collective action as the dependent variables and the two types of ingroup identity as the predictors. In Model 4 (Figure 3, Panel B), group-based anger was specified as the intervening variable. Importantly, the causal chains implied by Models 2 and 4 reflected the order of variables suggested by the results of Study 1. To check the robustness of our results, we estimated four corresponding models in which the covariates were accounted for. Other features of our analyses were the same as in Study 1. Missing data accounted for 5.58% of the analyzed dataset.
Total, indirect, and direct effects of group-based anger on life satisfaction (Study 2, Models 1 and 2).
Total, indirect, and direct effects of secure ingroup identity and collective narcissism on collective action (Study 2, Models 3 and 4).
Hypotheses testing
Table 2 provides means, standard deviations, and correlations among the variables of interest in Study 2. Model 1 (Figure 3, Panel A) revealed a negative effect of group-based anger on life satisfaction, B = -0.08, 95% CI [-0.11, -0.05]. In Model 2 (Figure 3, Panel B), secure ingroup identity and collective narcissism were added as parallel mediators. In line with H5, group-based anger was a better positive predictor of collective narcissism, B = 0.60, 95% CI [0.57, 0.63], than secure ingroup identity, B = 0.46, 95% CI [0.42, 0.51], χ2(1) = 54.55, p < .001. At the same time, life satisfaction was predicted negatively by collective narcissism, B = -0.10, 95% CI [-0.12, -0.07], and positively by secure ingroup identification, B = 0.08, 95% CI [0.07, 0.10], which supported H1. The direct effect of group-based anger was negative and significant, B = -0.06, 95% CI [-0.09, -0.03]. In accordance with H6, there was a negative indirect effect of group-based anger on life satisfaction via increased collective narcissism, IE = -0.06, SE = 0.01, 95% CI [-0.07, -0.04]. Additionally, group-based anger exerted a positive indirect effect on life satisfaction via increased secure ingroup identity, IE = 0.04, SE = 0.01, 95% CI [0.03, 0.05]. As shown by additional analyses (see the Online Supplemental Material), adding secure ingroup identity as a single mediator to Model 1 made the negative direct effect of group-based anger stronger. Moreover, the sum of the negative direct effect of group-based anger and the negative indirect effect of this variable through collective narcissism in Model 2 was greater in absolute value than the effect of group-based anger found in Model 1. This pattern of results suggests that secure ingroup identity acted as a suppressor that concealed the negative effect of group-based anger on life satisfaction (see Rucker et al., 2011).
Means, standard deviations, and correlation coefficients (Study 2).
Note. SES = Subjective economic status. Ns ranged from 5,155 to 7,992.
p < .05. **p < .01. ***p < .001.
In Model 3 (Figure 4, Panel A), support for normative and nonnormative collective action were regressed on collective narcissism and secure ingroup identity. Support for the normative forms of collective action was predicted positively both by collective narcissism, B = 0.25, 95% CI [0.23, 0.28], and secure ingroup identity, B = 0.19, 95% CI [0.17, 0.21]. In a similar vein, both collective narcissism, B = 0.51, 95% CI [0.48, 0.55], as well as secure ingroup identity, B = 0.12, 95% CI [0.09, 0.16], emerged as positive predictors of support for nonnormative collective action. In accordance with H7, collective narcissism was a stronger positive predictor of support for nonnormative than normative collective action, χ2(1) = 198.87, p < .001 whereas secure ingroup identity exerted a stronger positive effect on support for normative than nonnormative collective action, χ2(1) = 15.88, p < .001.
Next, we tested Model 4 (Figure 4, Panel B), which included group-based anger as a mediator. Although both secure ingroup identity, B = 0.07, 95% CI [0.06, 0.09], and collective narcissism, B = 0.27, 95% CI [0.25, 0.30], served as the positive predictors of group-based anger, the effect of the latter was significantly stronger, χ2(1) = 157.92, p < .001. This was consistent with H4 and the results of Study 1, where GRD was longitudinally predicted by collective narcissism but not by secure ingroup identity. Support for normative collective action was predicted positively by group-based anger, B = 0.39, 95% CI [0.34, 0.44], secure ingroup identity, B = 0.16, 95% CI [0.14, 0.18], and collective narcissism, B = 0.15, 95% CI [0.12, 0.17]. Interestingly, the effect of group-based anger was significantly stronger than the effects of secure ingroup identity, χ2(1) = 70.04, p < .001, and collective narcissism, χ2(1) = 63.77, p < .001. Similarly, support for nonnormative collective action was predicted positively by collective narcissism, B = 0.46, 95% CI [0.43, 0.50], group-based anger, B = 0.18, 95% CI [0.13, 0.23], and secure ingroup identity, B = 0.11, 95% CI [0.08, 0.14]. The effect of collective narcissism was significantly stronger than the effects of group-based anger, χ2(1) = 60.13, p < .001, and secure ingroup identity, χ2(1)= 144.33, p < .001.
In line with H3, group-based anger mediated the positive effects of collective narcissism on normative, IE = 0.11, SE = 0.01, 95% CI [0.09, 0.13], and nonnormative collective action, IE = 0.05, SE = 0.01, 95% CI [0.04, 0.06]. Similarly, the positive indirect effects of secure ingroup identity on support for normative, IE = 0.03, SE = 0.01, 95% CI [0.02, 0.04], and nonnormative collective action, IE = 0.01, SE = 0.002, 95% CI [0.01, 0.02], were also mediated by group-based anger. Accounting for the covariates did not change the results of Models 1–4 in a theoretically meaningful way. 13
Discussion
Study 2 corroborated our hypothesis that secure identity would be more predictive of normative than nonnormative collective action and that the opposite would be true for collective narcissism (H7). These findings are consistent with the research demonstrating a positive link between collective narcissism and destructive forms of political engagement (Górska et al., 2022; Jasko et al., 2020).
Regarding life satisfaction, Study 2 replicated the findings of Study 1. Whereas group-based anger and collective narcissism served as the negative predictors of life satisfaction, the effect of secure ingroup identity was positive. Interestingly, by promoting different types of ingroup identity simultaneously, group-based anger exerted opposite indirect effects on life satisfaction. In particular, the indirect effect through increased collective narcissism was negative (providing support for H6), whereas the indirect effect via secure ingroup identity was positive. As such, secure ingroup identity served as a suppressor (Rucker et al., 2011), which, when omitted, undermined the negative effect of group-based anger on life satisfaction (for similar findings, see Stronge et al., 2016). This pattern of results is consistent with the Rejection Identification Model (Branscombe et al., 1999), which predicts that recognizing one’s disadvantage lowers well-being directly and enhances it indirectly by strengthening ingroup identity. At the same time, Studies 1 and 2 suggest that not all types of ingroup identity neutralize the unfavorable outcomes of injustice perception (or GRD). Instead of counteracting the negative effect that group-based anger exerts on life satisfaction, collective narcissism seems to explain why this effect occurs in the first place.
General Discussion
The main goal of this research was to explore whether different types of ingroup identity (i.e., collective narcissism vs. secure ingroup identity) serve as equally adaptive strategies to deal with discrimination among LGBTQ+ people—a disadvantaged group that still has to fight for their rights. Using longitudinal and cross-sectional data collected in two surveys of the Polish LGBTQ+ community, we demonstrated that collective narcissism and secure ingroup identity differ regarding their antecedents and outcomes. Collective narcissism showed a stronger positive association with group-based anger and had a positive reciprocal relationship with GRD. At the same time, it served as a negative predictor of life satisfaction and exhibited a stronger positive link to nonnormative than normative collective action. In addition, collective narcissism mediated the negative effect of GRD (and group-based anger) on life satisfaction. We also found that GRD (and group-based anger) mediated the positive effect of collective narcissism on different types of collective action.
By contrast, secure ingroup identity was not longitudinally predicted by GRD and showed a weaker positive association with group-based anger. At the same time, it had a reciprocal positive relationship with life satisfaction and was a better positive predictor of support for normative than nonnormative collective action.
Theoretical Contribution
There are several ways in which our results contribute to the literature. To begin with, they add to the LGBTQ+ research by showing that treating ingroup identity as a homogenous construct (as was the case in most extant literature) may mask the processes responsible for life satisfaction and protest engagement among sexual and gender minorities. Specifically, we demonstrate that while secure ingroup identity indeed seems to operate as a “social cure” (i.e., it positively predicts life satisfaction and mitigates the negative effects of discrimination; see Haslam et al., 2018), its narcissistic counterpart acts instead as a “social curse” (i.e., it exerts a negative effect on life satisfaction; see Kellezi & Reicher, 2012; Kellezi et al., 2019).
Next, different patterns of associations with normative and nonnormative collective action warrant a reflection on whether the two types of LGBTQ+ individuals’ ingroup identity are equally beneficial for social change. The facilitating effect of secure ingroup identity seems rather clear (and in line with the SIMCA model, van Zomeren et al., 2008). This type of ingroup identity is primarily associated with normative collective action, which, as shown by the analyses of multiple social movements, is twice as effective in attaining political goals as its nonnormative counterpart (Chenoweth & Stephan, 2011). Moreover, as demonstrated in Study 1, secure ingroup identity increases life satisfaction over time, and life satisfaction is a positive longitudinal predictor of protest engagement among LGBTQ+ individuals.
The picture becomes more complex when collective narcissism is considered. On the one hand, it propels normative collective action, and, as such, may be considered functional for social change. On the other hand, by impairing the well-being of LGBTQ+ individuals, collective narcissism is likely to decrease their engagement and, therefore, impede change toward equality. Another reason to view collective narcissism as an obstacle on the way to equality is its strong positive link to nonnormative collective action. Extreme protest behaviors are known to reduce the support of the general public, which is considered a necessary condition for status quo to transform (Subašić et al., 2008). Perceiving nonnormative forms of collective action as immoral, the observers tend to distance themselves from the activists who engage in extreme protest behavior (Feinberg et al., 2020). Another route through which nonnormative protest behavior erodes public support for social movements, is via injustice perception and group efficacy beliefs—two proximal antecedents of collective action (van Zomeren et al., 2008, 2012). In comparison to its normative counterpart, nonnormative collective action is less successful in questioning the legitimacy of societal status quo and conveying faith that social change is possible (Thomas & Louis, 2014). Lastly, nonnormative engagement may thwart social change by provoking indiscriminate repressions from the authorities (Tompkins, 2015).
Current results also provide a longitudinal test of causal relations proposed in the ingroup identity literature. For instance, by demonstrating that life satisfaction is predicted negatively by the preceding level of collective narcissism and positively by secure ingroup identity, we offer longitudinal evidence in favor of previous findings obtained with cross-sectional data (Bagci et al., 2021; Golec de Zavala, 2019). At the same time, our results reveal effects that were not considered explicitly in past theorizing. For example, we show that secure ingroup identity not only promotes life satisfaction over time but is itself positively predicted by life satisfaction. This effect corresponds with the claim that satisfied individual-level needs (Cichocka & Cisłak, 2020; Cichocka et al., 2018) and high self-esteem (Golec de Zavala et al., 2020) facilitate the development of secure ingroup identification. In a similar vein, present results suggest that the positive association between GRD and collective narcissism (see Marchlewska et al., 2018) is, in fact, bidirectional.
In addition, we show that the distinction between collective narcissism and secure ingroup identification maintains its validity when applied to a disadvantaged group membership. Although highly informative, past research examined the correlates of the secure and narcissistic ingroup identity predominantly among advantaged groups (e.g., Górska et al., 2020). The only exceptions were two cross-sectional studies reported by Bagci et al. (2021), where the samples consisted of ethnic minorities. By demonstrating that similar associations may be observed longitudinally for a non-ethnic disadvantaged group, we add to this nascent literature.
Finally, our findings extend the collective action literature. By employing longitudinal panel data, we respond to the recent call for the examination of temporal aspects of protest engagement (Thomas et al., 2020) and by conducting our studies among sexual and gender minorities, we address one of the key knowledge gaps recently identified in the LGBTQ+ literature (Salvati & Koc, 2022, p. 229). We demonstrate that collective action depends positively on the preceding degree of life satisfaction rather than increases life satisfaction over time. This result is at odds with previous research suggesting the beneficial influence of activism on psychological well-being (e.g., Klar & Kasser, 2009). One of the potential explanations for such divergence may be found in the unique circumstances of the Polish LGBTQ+ community. Despite continued efforts of the local LGBTQ+ movement, the legal status of sexual and gender minorities in Poland has not improved since joining the European Union in 2004. Quite the contrary, the situation of LGBTQ+ Poles worsened significantly after the 2015 parliamentary elections, when the Law and Justice party—an openly homophobic right-wing party—rose to power (Gwiazda, 2020; ILGA Europe, 2021). As the post-engagement increase of positive emotions (i.e., pride) depends on the success of a given protest (Tausch & Becker, 2013), the absence of the positive longitudinal effect of collective action on life satisfaction in our research is not especially surprising. At the same time, an active response to collective disadvantage may require some degree of psychological resources. Specifically, emotions related to low life satisfaction such as sadness or hopelessness translate to withdrawal and apathy rather than increased collective action intentions (e.g., Cohen-Chen & van Zomeren, 2018; Greenaway et al., 2016; Smith et al., 2008).
Limitations and Future Directions
Our research is not without limitations. Most importantly, although the current results correspond to those obtained in the context of the advantaged group’s membership, the concept of collective narcissism seems to have different connotations when applied to advantaged and disadvantaged groups. One potential difference is the degree to which the sense of frustration inherent to collective narcissism is rooted in the objective level of inequality. Whereas narcissistic members of advantaged groups seem to experience relative deprivation despite their high standing in the social hierarchy (see Leach et al., 2007), narcissistic ingroup identity among the disadvantaged groups’ representatives may develop because of the hardships that their groups face. In the case of the Polish LGBTQ+ community, collective narcissism seems to be directly proportional to the degree of discrimination that it experiences. Specifically, the escalation of state-sponsored homo- and transphobia between 2015 and 2020 (e.g., Górska & Tausch, 2022; Mole et al., 2021; Żuk & Żuk, 2020) was accompanied by the increase in collective narcissism among LGBTQ+ individuals (see the Online Supplemental Material). Furthermore, collective narcissism of low- and high-status groups may be differentiated by the reasons for seeking external recognition. It may be that the disadvantaged yearn to be respected to the same degree as other groups, while the advantaged strive to gain greater respect than other groups. Thus, future research should systematically explore differences in collective narcissism of high- and low-status groups’ members.
Second, future studies would benefit from employing more comprehensive measures of group-based anger and secure ingroup identity. It is possible that the absence of some effects (e.g., the longitudinal effect of T1 secure ingroup identity on T2 collective action in Study 1) originates from the suboptimal measurement of these two variables.
Due to the limited number of measurements in Study 1, it was not possible to test for full longitudinal mediations. The half-longitudinal indirect effects (i.e., the products of parallel cross-lagged paths) that we estimated assume stationarity (i.e., the unchanging causal structure among variables; see Cole & Maxwell, 2003). As this assumption may not hold, there is a risk that the estimates of longitudinal indirect effects revealed in Study 1 are biased. Future studies should involve at least three measurements to obtain more reliable estimates of these effects.
Finally, secure ingroup identity was conceptualized as group identification without the narcissistic component. Thus, it cannot be fully ruled out that elements of insecurity unrelated to narcissism might still be present in such a measure. Having said that, we want to emphasize that to date no alternatives to measuring secure identity have been developed and our strategy was successfully employed in several other studies (e.g., Cichocka & Cisłak, 2020; Golec de Zavala et al., 2013; Golec de Zavala & Lantos, 2020).
Conclusion
The extant research suggested that strong ingroup identity among LGBTQ+ individuals (and other low-status groups’ representatives) may be considered a double-edged sword—a factor that facilitates as well as hinders dealing with the disadvantage. The current research offers some insights into exactly when the positive and negative consequences of ingroup identity are most likely to occur. Developing secure ingroup identity seems to help LGBTQ+ individuals in coping with their disadvantaged position both individually (through increased life satisfaction) and collectively (through collective action engagement). By contrast, the consequences of collective narcissism are clearly negative at the individual level of analysis (decreased levels of life satisfaction) and mixed when collective action outcomes are concerned.
Supplemental Material
sj-doc-1-gpi-10.1177_13684302221147125 – Supplemental material for The different effects of collective narcissism and secure ingroup identity on collective action and life satisfaction among LGBTQ+ individuals
Supplemental material, sj-doc-1-gpi-10.1177_13684302221147125 for The different effects of collective narcissism and secure ingroup identity on collective action and life satisfaction among LGBTQ+ individuals by Paulina Górska, Anna Stefaniak, Joanna Matera and Marta Marchlewska in Group Processes & Intergroup Relations
Footnotes
Acknowledgements
The authors wish to thank the Campaign Against Homophobia for sharing data employed in the present research.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: the preparation of this manuscript was funded from the National Science Centre Poland 2017/26/M/HS6/00689 and 2017/25/B/HS6/01116 grants conferred to the first author.
Supplemental material
Supplemental material for this article is available online.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
