This is an open access article distributed under the terms of the Creative Commons Attribution License (We all belong to multiple social groups. Some of the social groups we belong to afford us power and status (advantaged groups like men and White people), while others are routinely subject to prejudice and discrimination (disadvantaged groups like women and People of Colour). The social groups we belong to play an important role when seeking to improve the status of disadvantaged groups and achieve social equity through collective action (Subašić et al., 2008). Yet much of the social psychological literature is limited to examining collective action among advantaged and disadvantaged group members within one identity dimension (i.e., men and women who engage in collective action for gender equity) without considering how belonging to other social groups might affect this behaviour (i.e., whether they are also heterosexual or LGBTIQ+). This approach does not fully capture the complexity of social identities, which can be both advantaged and disadvantaged, and how they influence advocacy for social equity (Pham et al., 2023). Moreover, previous research has largely focused on disadvantaged group members who engage in collective action for their own group (van Zomeren et al., 2008) or advantaged group members who engage in allyship for a disadvantaged group (Radke et al., 2020) rather than solidarity between disadvantaged groups (cf. Cortland et al., 2017).
Therefore, in this paper, we investigate the role membership in multiple social groups – some of which are advantaged, and others disadvantaged – plays in understanding collective action to improve the status of another disadvantaged group. To do this, we examine participants’ willingness to engage in collective action against the discrimination experienced by two disadvantaged groups – Muslims and people with a disability1 – dependent upon the number of disadvantaged social groups they belong to across specific dimensions of gender, race, and sexual orientation. We also examine what role intersectional awareness as well as perceived similarity plays in understanding — intraminority solidarity.
Multiple Group Memberships
Traditional approaches in social psychology often examine the relationship between advantaged and disadvantaged groups without considering the influence of membership in multiple social groups. Intersectionality (Collins, 1990; Crenshaw, 1991), however, provides a critical framework for understanding the multifaceted and interconnected nature of a person’s social identity across a range of different identity dimensions (Cole, 2009; Overstreet et al., 2020). For example, White women in the US and the UK are both advantaged based on their racial identity and disadvantaged because of their gender identity. Moreover, in both of these contexts, a White woman’s experience of gender discrimination is qualitatively different to that which a Black woman experiences: Black women are subject to gender stereotypes (such as the jezebel, the angry black woman) that White women are not (Collins, 1990; Donovan, 2011). These stereotypes have negative effects on the treatment of Black women (Crenshaw, 1991). For instance, Black women are often unfairly discredited when reporting sexual violence and are less likely to see their perpetrators convicted compared to White women. Additionally, domestic violence in minority communities is sometimes concealed in an effort to prevent reinforcing racial stereotypes (Kelly et al., 2020). It is therefore necessary to consider intersectionality when investigating who, what and why people engage in collective action when fighting for the rights of disadvantaged groups. This analysis must also acknowledge and account for context, as experiences of advantage and disadvantage are specific to each society. Contextual factors such as intergroup conflict and social norms can influence the salience of social identities and impact attitudes and behaviours towards ingroup and outgroup members.
Collective Action
Collective action refers to behaviours such as protesting and signing petitions to improve the status of a disadvantaged group (Wright et al., 1990). Previous research has shown that disadvantaged group members are more likely to participate in collective action when they identify with other (politicised) disadvantaged group members (such as an activist), recognise that their group is unfairly treated by society and feel angry about it, believe that engaging in collective action will be effective, and hold moral beliefs about the importance of engaging in this behaviour (Agostini & van Zomeren, 2021; van Zomeren et al., 2008). While much of the literature has focused on disadvantaged group members engaging in collective action for their own group, more recent research has examined the role advantaged group members can play as allies in improving the status of disadvantaged groups (Radke et al., 2020). There is limited research, however, on how disadvantaged group members come to engage in collective action for one another.
Intraminority Solidarity
Research has begun to explore how disadvantaged group members come to engage in collective action on behalf of another disadvantaged group. Intraminority solidarity refers to two or more minority groups acting in unity by mutually supporting a common goal despite differences among these groups (Craig & Richeson, 2016). An example of intraminority solidarity could be a Black community organisation petitioning for police reform with support from the Latine community in the area after a Black community member was murdered by police based on racial profiling.
Yet contrary to popular opinion, disadvantaged group members do not always support one another when advocating for social equity (Craig & Richeson, 2012, 2014). For example, a Black community organisation might not assist with the clean-up of a vandalised LGBTQ+ centre because they are prioritizing their immediate community issues like systemic racism and economic inequality. Groups may act in solidarity because they perceive themselves as similar, both fighting for equity. But, when groups are impacted by more nuanced discrimination, the direct similarities are not immediately apparent (i.e., disproportioned violence against Black trans women for Black and LGBTIQ+ communities). Nevertheless, previous research has shown that solidarity between disadvantaged groups can be facilitated by encouraging their perceived similarity with one another (Cortland et al., 2017).
Initially, Craig and Richeson (2012) examined Black-Latine relations, groups who both face discrimination based on their race. The researchers found that participants, who were Black Americans, expressed closeness and common fate with Latine Americans after indicating the discrimination they (Black Americans) and their racial group face, and, as a result, reported increased positive attitudes towards Latine Americans (Craig & Richeson, 2012). Subsequent studies expanded on these findings, demonstrating that increased perceived similarity significantly explained the effect of perceived discrimination, leading to a reduction in anti-Black attitudes among another disadvantaged outgroup, Asian Americans.
On the other hand, when disadvantaged group members differ on identity dimensions (i.e., White women and racial minorities), this can lead to increased intergroup bias (Craig et al., 2012; Craig & Richeson, 2014). Previous research has shown that White women who read an article outlining the social and economic consequences of sexism in the United States reported more pro-White racial bias compared to those who read an unrelated article about left-handedness and brain function (Craig et al., 2012). This research has implications for intraminority solidarity as it shows the potential disconnect between disadvantaged groups who face discrimination on differing dimensions of identity, hindering solidarity efforts. However, more recent research has built upon this literature and demonstrated perceived similarity as an effective mechanism to promote intraminority solidarity regardless of identity dimension (Cortland et al., 2017).
Perceived Similarity
Previous research has found that perceived similarity between disadvantaged groups can predict positive attitudes towards other disadvantaged groups as well as foster intraminority solidarity (Cortland et al., 2017; Craig & Richeson, 2012; Craig et al., 2012). Cortland and colleagues (2017) expand upon previous research to encompass how perceived similarity can be used as a mechanism through which positive relations between groups that do not share a common identity dimension, such as their race or sexual identity (i.e., Black Americans and sexual minorities; Asian Americans and sexual minorities) can be facilitated.
The researchers conducted a series of studies to explore how perceived similarity between minority groups can enhance positive intergroup relations and attitudes (Cortland et al., 2017). Asian Americans, who after reading about the similarities in the discrimination faced by their group and sexual minorities, reported more positive attitudes towards sexual minorities and increased support for gay civil rights. Increased perceived similarity was found to explain the relationship between shared discrimination experience salience and positive attitudes toward gay men and lesbians as well as more support for gay and lesbian civil rights. They then found that when a similarity-seeking mindset was induced in White women which was unrelated to discrimination, this led to a reduction in racial bias after they read an article about sexism, compared to participants who read an unrelated article about a lawsuit against McDonald’s (Cortland et al., 2017). These findings indicate that perceived similarity can foster positive attitudes between different disadvantaged groups.
In a recent series of studies, Pham and colleagues (2023) examined the relationship between belonging to multiple disadvantaged groups and support for intraminority solidarity and perceived similar experiences across disadvantaged groups. The number of disadvantaged identities for each participant was based on race/ethnicity (White, participants of colour), gender (women, transgender, or gender-diverse), and sexual orientation (LGBTQI+). Each identity was counted, resulting in a total ranging from 0 to 3. Pham and colleagues (2023) found that belonging to more disadvantaged groups was associated with increased support for intraminority solidarity and policy support for a disadvantaged outgroup, low-SES individuals.
These findings support the idea that belonging to multiple disadvantaged groups can foster intraminority solidarity. They also highlight the role of perceived similarity to increase collective intentions and efforts for another disadvantaged group. However, we believe that intraminority solidarity can be facilitated through additional – and potentially more effective – mechanisms among multiple disadvantaged groups. In this paper, we propose that intersectional awareness is a key mechanism behind intraminority solidarity that has been overlooked in the previous literature.
Intersectional Awareness
Intersectional awareness is a recognition of systemic inequities and the unique experiences they create across intersecting identities (Curtin et al., 2015). People who are aware of intersectionality understand how various forms of discrimination originate from multiple interlocking systems of oppression. This is in opposition to the idea that discrimination emanates from a single form, such as the patriarchy in the case of gender-based discrimination (Greenwood, 2008). For example, a person with high intersectional awareness would be more likely to be critical of racism, while also recognising how other identities can shape one’s experience of racism, without being a racial minority.
Previous research has found that higher intersectional awareness predicts increased promotion of diversity, identification as an activist, and rights-based activism for a disadvantaged outgroup (Curtin et al., 2015). More specifically, in a sample of non-activist male and female students, a higher intersectional awareness predicted increased rights-based activism across three minority-rights domains (women’s rights activism, racial equality activism, and LGBTIQ+ rights activism; Curtin et al., 2015). Increased intersectional awareness also predicted participants' attitudes towards general social change, particularly in terms of taking actions to promote diversity and actively interrupt prejudice (Curtin et al., 2015).
These findings have implications for understanding intraminority solidarity, as individuals with higher intersectional awareness are more likely to recognize the interconnected nature of various forms of oppression. This recognition increases their willingness to engage in collective action for a more diverse range of disadvantaged social groups. In this paper, we propose that intersectional awareness can foster solidarity among disadvantaged groups by highlighting common struggles and shared goals, strengthening collective efforts toward social justice and equity.
Current Study
This current study investigates the importance of belonging to multiple disadvantaged groups in shaping intraminority solidarity efforts as well as the role perceived similarity and intersectional awareness might play in understanding this relationship. Across two studies, participants were asked to complete an online survey. Based on a combination of reported demographics and screening done through Prolific Academic, participants either belonged to zero, one, two, or three (0-3) disadvantage groups based on their gender (i.e., Cis-male, Cis-female2), race (i.e., Black, White), and sexual identity (i.e., Heterosexual, LGBTQI+). For example, a Heterosexual White Man would be categorised as belonging to zero disadvantaged groups, while a Black LGBTIQ+ Woman would belong to three. Participants indicated their intentions to engage in collective action for two disadvantaged outgroups (i.e., Muslims, people with a disability), perceptions of similarity between their collective identity and a disadvantaged outgroup, and awareness of intersectionality.
In line with Pham and colleagues (2023), we predict that participants belonging to more disadvantaged groups will report greater intentions to engage in collective action for two disadvantaged outgroups (i.e., Muslims, people with a disability). We then examine whether perceived similarity or intersectional awareness serves as a stronger positive mediator of collective action intentions among individuals belonging to more disadvantaged group memberships.
Participants for this study were recruited from the United States and the United Kingdom. These nations provide an interesting case study to examine intraminority solidarity – both countries have equal opportunity and anti-discrimination legislation, but this is increasingly under attack from far-right movements and populism suggesting there is still a need for intraminority solidarity. Examining both these contexts also allows us to consider how intraminority solidarity plays out across similar disadvantaged groups which differ in terms of their history and current political situation (e.g., access to abortion is currently not under threat in the UK but has been banned in several states across the US; trans-exclusionary radical feminists are particularly vocal in the UK while police brutality towards Black Americans has been a salient political issue in recent years). Previous intraminority solidarity research has also been conducted in these contexts, so it is necessary to recruit a similar sample to establish the role intersectional awareness plays when considering intraminority solidarity in comparison to the already identified role of perceived similarity (Cortland et al., 2017; Curtin et al., 2015; Pham et al., 2023). Nevertheless, we do acknowledge that these samples are not representative of intraminority solidarity in other nations.
Our research extends previous work by examining an additional potential explanation for intraminority solidarity: Intersectional awareness. We also examine collective action intentions for two specific disadvantaged outgroups (i.e., Muslims and people with a disability). Moreover, our measure of similarity pertains to the similarity between the individual person and their collective experience of discrimination related to the two disadvantaged outgroups, rather than the similarity of discrimination experienced by disadvantaged groups in general. The extent to which individuals' awareness of intersectionality is influenced by belonging to multiple advantaged and/or disadvantaged identities and the implications this has for advancing social change remains underexplored in the psychological literature.
Study 1
Method
In Study 1, we examined the effect of the number of disadvantaged groups participants belong to on collective action intentions for another disadvantaged outgroup (i.e., Muslims, people with a disability).
We also examined perceptions of similarity with the disadvantaged outgroup and intersectional awareness as potential mediators of this relationship. We predict that an increase in the number of disadvantaged group memberships will be associated with greater intentions to engage in collective action for both disadvantaged outgroups, and that the relationship will be explained through higher levels of perceived similarity and intersectional awareness.
Participants
For the ANOVA analyses, power analysis using G*Power (Faul et al., 2007) indicated that a total sample of 280 participants (70 per condition) would be needed to detect a small to medium effect (f = .20) with 80% power and an alpha of .05. To estimate the statistical power for detecting indirect effects in our proposed mediation model3, a Monte Carlo power analysis was conducted. The Monte Carlo simulation indicated that the sample size needed to detect the indirect effects for the hypothesized model with a power of 0.80 was 116. The results suggest that the proposed study design is adequately powered to detect indirect effects with medium effect sizes (β = 0.30) for all paths in the model.
The final sample consisted of 278 participants who were recruited online using Prolific (www.prolific.co)4. Participants were paid £1 for taking part in the study. There are strengths and weaknesses associated with collecting data via Prolific. For instance, collecting data via Prolific allowed us to screen for and obtain a well-powered sample of participants who belonged to the different disadvantaged groups which is required to explore our research question. We also took steps to ensure the quality of our data by checking that participants belonged to the identified groups at the end of the questionnaire and including an attention check. However, we do acknowledge the limitations of using crowdsourcing sampling methods such as self-selection and non-naivete of participants which we discuss further in the general discussion. Muslim participants (N = 5) were excluded from the ANOVA and mediation analyses pertaining to collective action intentions for Muslims, and disabled participants (N = 88) were excluded from the ANOVA and mediation analyses pertaining to collective action intentions for people with a disability.
The participants' ages ranged between 18 and 77 years old, with a mean of 35.57 years (SD = 13.53). Participants were located in the UK (N = 159) and the US (N = 119). The sample was comprised of 139 (50%) men and 139 (50%) women, 144 (50.3%) White and 138 (49.7%) Black participants, and 155 (55%) heterosexual and 123 (44%) LGBTIQ+ participants. Demographic information can be found in Table 1. Participants completed the study in an average of 9.3 minutes (with a median time of 7.18). Participants completed the study in January 2021. The study was preregistered on AsPredicted (Falco & Radke, 2022S), and the data is available on the OSF (Falco & Radke, 2025S-a).
Table 1
Demographic Characteristics of Participants for Study 1
| Number of disadvantaged group memberships | Zero | One | Two | Three | Full Sample | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | n | % | n | % | |
| Country | ||||||||||
| United States | 68 | 97 | 56 | 75 | 30 | 40 | 5 | 8 | 159 | 57 |
| United Kingdom | 2 | 3 | 19 | 25 | 44 | 60 | 54 | 92 | 119 | 43 |
| Race | ||||||||||
| White | 70 | 100 | 50 | 67 | 20 | 27 | 0 | 0 | 140 | 50 |
| Black | 0 | 0 | 25 | 33 | 54 | 73 | 59 | 100 | 138 | 50 |
| Gender | ||||||||||
| Male | 70 | 100 | 47 | 63 | 22 | 30 | 0 | 0 | 139 | 50 |
| Female | 0 | 0 | 28 | 37 | 52 | 70 | 59 | 100 | 139 | 50 |
| Sexual orientation | ||||||||||
| Heterosexual | 70 | 100 | 53 | 71 | 32 | 43 | 0 | 0 | 155 | 56 |
| LGBTIQ+ | 0 | 0 | 23 | 29 | 42 | 57 | 59 | 100 | 123 | 44 |
Note. N = 278 (Zero, n = 70; One, n = 75; Two, n = 74; Three, n = 59).
Design and Procedure
The study employed a one-way between-subjects design with four levels (number of disadvantaged group memberships: 0, 1, 2, and 3). The participants varied in their combination of identities regarding race, gender, and sexual identity (i.e., Black LGBTIQ+ men belonged to two disadvantaged groups while White LGBTIQ+ men belonged to one disadvantaged group; see Supplementary Materials [Falco & Radke, 2025S-b] for a table of the recruitment strategy).
The sample comprised of participants who did not belong to any disadvantaged group (i.e., White heterosexual men; n = 74), participants who belonged to one disadvantaged group (i.e., Black heterosexual men, White LGBTIQ+ men, White heterosexual women; n = 77), two disadvantaged groups (i.e., Black LGBTIQ+ men, White LGBTIQ+ women, Black heterosexual women; n = 74, and three disadvantaged groups (i.e., Black LGBTIQ+ women; n = 65). Participants were asked to complete a questionnaire that measured the proposed mediators (i.e., perceived similarity, intersectional awareness), dependent variables (collective action intentions for Muslims, collective action intentions for people with a disability), an attention check, and demographic information. We screened participants through Prolific to obtain our eight specific groups based on their combinations of identity and matched this information with their reported demographics in the survey. Participants were additionally asked to indicate their religion and disability status.
Measures
The following variables were measured on a 1 (strongly disagree) to 7 (strongly agree) Likert-type scale. The full list of items can be found in Falco and Radke, 2025S-b.
Collective action intentions for Muslims were measured using 10 items adapted from van Zomeren et al. (2008) and Radke et al. (2018; “I would be willing to protest against the discrimination that Muslims experience”; α = .95). Prior to completing the measure, participants were provided with a short paragraph discussing the discrimination that Muslims experience (see Falco & Radke, 2025S-b).
Collective action intentions for people with a disability were measured using 10 items adapted from van Zomeren et al. (2008) and Radke et al. (2018; “I would be willing to protest against the discrimination that people with disabilities experience”; α = .92). Prior to the measure, participants were provided with a short paragraph introduction to discrimination faced by people with a disability (see Falco & Radke, 2025S-b).
Perceived similarity was measured using 2 items adapted from Craig and Richeson (2012) and Cortland et al. (2017). Participants were asked how similar they perceived themselves to be towards two disadvantaged outgroups: Muslims (i.e., “I think I’m very similar to Muslims” and “I have a lot in common with the average Muslim”; r = .79, p < .001), and people with a disability (“I think I’m very similar to people with disabilities” and “I have a lot in common with the average person with a disability”; r = .82, p < .001).
Intersectional awareness was measured using 5 items (i.e., “Understanding the experiences of women from different ethnic groups and with different sexual orientations is important”; α = .80) adapted from Curtin et al. (2015) and Greenwood (2008). The specific intersections included in the measure were directly aligned with the identities we explored in the study: race, sexual identity and gender.
Attention check5: Within the measure of collective action intentions for Muslims, we asked participants to respond to the following statement: “I would be willing to boycott companies that discriminate against Muslims. Please ignore this statement and select ‘Agree’”.
Results and Discussion
Means, standard deviations, and correlations between the variables are summarised in Table 2. To test our initial hypothesis, we conducted two univariate ANOVAs for each of the dependent variables. To test our second hypothesis6, we ran a parallel mediation to examine whether either of the two would be a stronger mediator.
Table 2
Means, Standard Deviations, and Pearson’s Correlations for Study 1
| Variable | M (SD) | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|---|
| 1. Intersectional Awareness | 5.64 (1.05) | – | |||||
| 2. Perceived Similarity, Muslims | 3.39 (1.56) | .24*** | – | ||||
| 3. Perceived Similarity, Disability | 4.09 (1.71) | .09 | .36*** | – | |||
| 4. Collective Action Intentions, Muslims | 4.66 (1.49) | .54*** | .47*** | .10 | – | ||
| 5. Collective Action Intentions, Disability | 5.16 (1.21) | .52*** | .28*** | .19*** | .74*** | – | |
| 6. Number of Disadvantaged Group Memberships | 1.43 (1.08) | .37*** | .12* | .05 | .32*** | .33*** | – |
*p < .05. **p < .01. ***p < .001.
Collective Action Intentions
The one-way ANOVA revealed a significant effect of multiple disadvantaged group memberships on collective action for Muslims, F(3,269) = 10.18, p < .001, 95% CI [.03, .16], = .10. The second one-way ANOVA revealed a significant effect of multiple disadvantaged group memberships on collective action for people with a disability, F(3,186) = 7.97, p < .001, 95% CI [.03, .19], = .12). See Figure 1 and Falco and Radke, 2025S-b, for differences between the number of disadvantaged group memberships.
Figure 1
Differences of Disadvantaged Group Memberships on the Dependent Variables
Note. The legend corresponds to the number of disadvantaged groups.
*p < .05. **p < .01. ***p < .001.
Parallel Mediation: Perceived Similarity and Intersectional Awareness
Mediation analysis using bootstrapping procedures (R PROCESS macro; Hayes, 2022) was used to examine the indirect effect of the number of disadvantaged groups (coded as 0 = no groups, 1 = one group, 2 = two groups, 3 = three groups) on collective action intentions for Muslims (Model 1) and people with a disability (Model 2) via perceived similarity and intersectional awareness. We believe that it is justified to treat the number of disadvantaged group memberships as a continuous variable for the mediation analysis given the linearity of this variable observed when testing Hypothesis 1. We used 5000 bootstrap samples to estimate 95% confidence intervals for the indirect effect.
As can be seen in Figure 2, the mediation analysis revealed a significant indirect effect of multiple disadvantaged group memberships on collective action intentions for Muslims through intersectional awareness (B = .11, BSE = .03, 95% CI [.06, .18]) but not perceived similarity (B = -.01, BSE = .02, 95% CI [-.06, .02]).
Figure 2
Parallel Mediation Analysis for the Effect of the Number of Disadvantaged Group Memberships on the Outcome Variables Via Intersectional Awareness and Perceived Similarity
Note. Group memberships: None = 0, one group = 1, two groups = 2, three groups = 3.
We also found a significant indirect of multiple disadvantaged group memberships on collective action intentions for people with a disability through intersectional awareness (B = .11, BSE = .03, 95% CI [.04, .19]) but not through perceived similarity (B = -.01, BSE = .01, 95% CI [-.04, .01]).
As predicted, we found that belonging to a greater number of disadvantaged groups would predict increased intentions to engage in collective action for two disadvantaged outgroups (i.e., Muslims and people with a disability). This finding aligns with prior research showing that having multiple disadvantaged identities is positively associated with intraminority solidarity (Pham et al., 2023).
We also expected that intersectional awareness and perceived similarity would explain this relationship. However, we did not find that increased perceived similarity explained the relationship between belonging to more disadvantaged groups and increased collective action for Muslims or people with a disability. This finding contrasts with research which has found that perceived similarity positively mediated the relationship between having multiple disadvantaged identities and increased support for another disadvantaged group (Cortland et al., 2017; Pham et al., 2023).
We did, however, find that increased intersectional awareness explained for the relationship between belonging to more disadvantaged groups and collective action intentions for both disadvantaged outgroups (i.e., Muslims and people with a disability). This finding is supported by prior research showing that a higher level of intersectional awareness is positively associated with intentions to engage in rights-based activism (Curtin et al., 2015). This study makes a unique contribution to the literature by identifying a new explanation for the relationship between belonging to multiple disadvantaged groups and intentions to engage in collective action for a disadvantaged outgroup, extending and building upon previous research. Our findings suggest that intersectional awareness is critical to fostering collective action among disadvantaged group members.
Study 2
Method
While existing literature has found that perceived similarity plays a pivotal role in facilitating intraminority solidarity (see, i.e., Cortland et al., 2017; Craig & Richeson 2012, 2014; Pham et al., 2023), our initial study did not provide evidence to support this hypothesis. Instead, we found that belonging to more disadvantaged groups predicted higher levels of intersectional awareness. Given these contradictory finding, we sought to replicate our results in Study 2.
Participants
For the ANOVA analyses, power analysis using G*Power (Faul et al., 2007) indicated that a total sample of 280 participants (70 per condition) would be needed to detect a small to medium effect (f = .20) with 80% power and an alpha of .05. To estimate the statistical power for detecting indirect effects in our proposed mediation model, a Monte Carlo power analysis was conducted. The Monte Carlo simulation indicated that the sample size needed to detect the indirect effects for the hypothesized model with a power of 0.80 was 116. The results suggest that the proposed study design is adequately powered to detect indirect effects with medium effect sizes (β = 0.30) for all paths in the model.
The final sample consisted of 272 participants living in the United Kingdom recruited online using Prolific (www.prolific.co)7. We encountered the same strengths and weaknesses in collecting data via Prolific as with Study 1. Participants were paid £1 for taking part in the study. Muslim participants (N = 11) were excluded from the ANOVA and mediation analyses pertaining to collective action intentions for Muslims, and disabled participants (N = 67) were excluded from the ANOVA and mediation analyses pertaining to collective action intentions for people with a disability.
The participants' ages ranged between 18 and 77 years old, with a mean of 35.19 years (SD = 12.66). The sample was comprised of 136 (50%) men and 136 (50%) women, 144 (53%) White and 128 (47%) Black participants, 165 (60%) heterosexual and 107 (40%) LGBTIQ+ participants. Demographic information can be found in Table 3. Participants completed the study in an average of 13.4 minutes (with a median time of 10.21). Participants completed the study in November 2022. The study was preregistered (Falco & Radke, 2022S), and the data is available on the OSF (Falco & Radke, 2025S-a).
Table 3
Demographic Characteristics of Participants for Study 2
| Number of disadvantaged group memberships | Zero | One | Two | Three | Full Sample | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | n | % | n | % | |
| Race | ||||||||||
| White | 70 | 100 | 52 | 65 | 22 | 29 | 0 | 0 | 144 | 53 |
| Black | 0 | 0 | 28 | 35 | 53 | 71 | 47 | 100 | 128 | 47 |
| Gender | ||||||||||
| Male | 70 | 100 | 50 | 62 | 16 | 21 | 0 | 0 | 136 | 50 |
| Female | 0 | 0 | 30 | 38 | 59 | 79 | 47 | 100 | 136 | 50 |
| Sexual orientation | ||||||||||
| Heterosexual | 70 | 100 | 58 | 72 | 37 | 50 | 0 | 0 | 165 | 60 |
| LGBTIQ+ | 0 | 0 | 22 | 28 | 38 | 50 | 47 | 100 | 107 | 40 |
Note. N = 272 (Zero, n = 70; One, n = 80; Two, n = 75; Three, n = 47).
Design and Procedure
The same design used in Study 1 was employed in Study 2. The sample comprised of participants who did not belong to any disadvantaged group (i.e., White heterosexual men; n = 71), participants who belonged to one disadvantaged group (i.e., Black heterosexual men, White LGBTIQ+ men, White heterosexual women; n = 79), two disadvantaged groups (i.e., Black LGBTIQ+ men, White LGBTIQ+ women, Black heterosexual women; n = 76, and three disadvantaged groups (i.e., Black LGBTIQ+ women; n = 46). We recruited participants who belonged to these groups using screeners available on Prolific, and these were checked against the demographics reported at the end of the study.
Participants were asked to complete a questionnaire that included the same measures as in Study 1 (i.e., perceived similarity, intersectional awareness), dependent variables (collective action intentions for Muslims, collective action intentions for people with a disability), an attention check8, and demographic details.
Measures
Perceived similarity with Muslims (r = .86, p < .001), perceived similarity with people with a disability (r = .83, p < .001), intersectional awareness (α = .80), collective action intentions for Muslims (α = .95), and collective action intentions for people with a disability (α = .94), were measured the same as in Study 1.
Results and Discussion
Means, standard deviations, and correlations between the variables are summarised in Table 4. We ran the same series of tests as we had done in Study 1.
Table 4
Means, Standard Deviations, and Pearson’s Correlations for Study 2
| Variable | M (SD) | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|---|
| 1. Intersectional Awareness | 5.60 (1.03) | – | |||||
| 2. Perceived Similarity, Muslims | 3.61 (1.79) | .30*** | – | ||||
| 3. Perceived Similarity, Disability | 4.04 (1.83) | .24*** | .35*** | – | |||
| 4. Collective Action Intentions, Muslims | 4.55 (1.51) | .55*** | .42*** | .10 | – | ||
| 5. Collective Action Intentions, Disability | 5.04 (1.34) | .44*** | .30*** | .18** | .74*** | – | |
| 6. Number of Disadvantaged Group Memberships | 1.36 (1.04) | .36*** | .21*** | .08 | .41*** | .35*** | – |
*p < .05. **p < .01. ***p < .001.
Intentions to Engage in Collective Action
The one-way ANOVA revealed a significant effect of multiple disadvantaged group memberships on collective action intentions for Muslims, F(3,257) = 17.54, p < .001), 95% CI [.08, .24], = .17) (see Figure 3). The second one-way ANOVA revealed a significant effect of multiple disadvantaged group memberships on collective action intentions for people with a disability F(3,201) = 9.79, p < .001), 95% CI [.04, .21], = .13). See Figure 3 and Falco and Radke, 2025S-b for differences between the number of disadvantaged group memberships.
Figure 3
Differences of Disadvantaged Group Memberships on the Dependent Variables
Note. The legend corresponds to the number of disadvantaged groups.
*p < .05. **p < .01. ***p < .001.
Parallel Mediation: Perceived Similarity and Intersectional Awareness
Mediation analysis using bootstrapping procedures (R PROCESS macro; Hayes, 2022) was used to examine the indirect effect of the number of disadvantaged groups (coded as 0 = no groups, 1 = one group, 2 = two groups, 3 = three groups) on collective action intentions for Muslims (Model 1) and people with a disability (Model 2) via perceived similarity and intersectional awareness. We used 5000 bootstrap samples to estimate 95% confidence intervals for the indirect effect. Again, perceived similarity and intersectional awareness were examined as parallel mediators in the models.
As can be seen in Figure 4, the mediation analysis revealed a significant indirect effect of multiple disadvantaged group memberships on collective action intentions for Muslims through intersectional awareness (B = .07, BSE = .02, 95% CI [.02, .12]) but not perceived similarity (B = .01, BSE = .02, 95% CI [-.02, .04]).
Figure 4
Parallel Mediation Analysis for the Effect of the Number of Disadvantaged Group Memberships on the Outcome Variables Via Intersectional Awareness and Perceived Similarity
Note. Group memberships: None = 0, one group = 1, two groups = 2, three groups = 3.
We also found a significant indirect effect of multiple disadvantaged group memberships on collective action intentions for people with a disability through intersectional awareness (B = .08, BSE = .02, 95% CI [.04, .14]) but not perceived similarity (B = -.004, BSE = .01, 95% CI [-.02, .01]).
Providing support for our hypotheses and replicating the results from Study 1, we found that an increase in disadvantaged group memberships was associated with greater intentions to engage in collective action for another disadvantaged group (in this case, for Muslims and people with a disability). As in Study 1, we found partial support for our second hypothesis. We did not find that perceived similarity explained the relationship between the number of disadvantaged group memberships and intentions to engage in collective action for another outgroup. However, increased intersectional awareness again mediated this relationship. We successfully replicated our results from the first study. We discuss these findings and their implications in more detail below.
General Discussion
In this paper, we investigated the role belonging to multiple disadvantaged groups has on collective action intentions for another disadvantaged outgroup. Specifically, we examined whether collective action intentions for another disadvantaged outgroup (i.e., Muslims, people with a disability) would increase as the number of disadvantaged groups a person belongs to also increased. We further examined increased perceptions of similarity with the disadvantaged outgroups and intersectional awareness as potential mediators of this relationship. Across two studies, in support of our initial hypothesis, we found that belonging to more disadvantaged group memberships significantly predicted greater collective action intentions for both disadvantaged outgroups. We further found that this effect was explained by increased intersectional awareness but not perceived similarity with the disadvantaged group, partially supporting our second hypothesis.
We expected that in addition to the number of disadvantaged groups an individual belongs to, perceiving their collective group memberships as similar to the disadvantaged outgroup would predict an increase in collective action for both disadvantaged groups. Previous research has found increased perceived similarity as a mechanism through which positive relations as well as collective action and solidarity can be fostered between disadvantaged groups (Cortland et al., 2017; Craig & Richeson, 2012, 2014). Yet, we found that perceived similarity did not explain the relationship between increased disadvantaged group membership and collective action intentions for another disadvantaged group. Instead, we found that the more disadvantaged groups a person belongs to, the higher their intersectional awareness, which in turn was associated with greater collective action for Muslims and people with a disability.
We argued that belonging to more disadvantaged groups would expose individuals to a more nuanced and interconnected experience of discrimination, fostering higher levels of intersectional awareness. Previous research has found that intersectional awareness is a predictor of pro-social attitudes towards an unspecific disadvantaged group (Curtin et al., 2015). Intersectional awareness also predicted intentions to engage in rights-based activism across three minority-rights domains (women’s rights, racial equality, and LGBTIQ+ rights; Curtin et al., 2015).
Our finding indicates that individuals who belong to an increased number of disadvantaged social groups may have a better understanding of how discrimination is systemic, which impacts their inclination to advocate for other disadvantaged groups. Emphasizing the interconnected and overlapping discrimination experienced by different groups through intersectional awareness may be particularly impactful for individuals belonging to multiple marginalized communities and facilitate stronger solidarity among groups, as opposed to focusing on similarity. These findings suggest that for people with multiple disadvantaged group memberships, intersectional awareness may be closely tied to their lived experiences of discrimination. However, this work should be interpreted with caution and replicated with other multiple disadvantaged intersectional groups to ensure its broader applicability and validity. In addition, future studies could explore whether intersectional awareness captures a more generalized understanding of systemic inequalities or if it is shaped by personal experiences of discrimination within one's own intersecting identities.
The main contribution this paper makes to the psychological literature is identifying a novel and under researched explanation for the relationship between belonging to multiple disadvantaged groups and intentions to engage in collective action for a disadvantaged outgroup. Our work explores the complexities of intraminority solidarity, emphasising the critical role of intersectionality in fostering effective coalitions. It aligns with previous work but offers an alternative theoretical perspective on intraminority solidarity. This research has implications for the necessity to include intersectional awareness when seeking to promote intraminority solidarity. In the US and UK, activists and organisations practise intersectionality; emerging feminists of the fourth wave support and promote using intersectionality as a tool to be more inclusive in movements and when talking about current issues (i.e., reproductive rights, domestic violence). It is also important due to current social justice organisations’ preferences for an intersectional framework rather than a similarity-based framework regarding their structure, mission, and goals (Montoya, 2021). However, while it is being widely promoted and practised, there has not been sufficient research on its effectiveness in increasing solidarity and establishing coalitions in social psychological research.
Secondly, this paper contributes to the psychological literature by further examining social identity as multidimensional rather than a singular and isolated construct. Much literature has examined disadvantaged and advantaged groups based on a single identity (i.e., Women, Latino Americans) without considering how these identities are influenced by the other social identities a person possesses. As long as discrimination is seen as singular and viewed as isolated, it will remain in place. Only by recognising the connections between forms of discrimination as they emanate from an interconnected source, can we start to dismantle oppression.
There are, however, limitations to the present research. For instance, other identities (such as class, education level, and age) might influence the results. While beyond the scope of the current work, we encourage future research to examine whether varying combinations of other identities (i.e., class, age, education) would result in similar findings. Second, we recruited our participants through Prolific. While we acknowledge the limitations of using crowdsourcing sampling methods (e.g., self-selection, non-naivete of participants), we considered this method of data collection necessary to ensure that we obtained a well-powered sample of participants who belonged to the different disadvantaged groups required to examine our research question (i.e., the role belonging to multiple disadvantaged groups plays in understanding intraminority solidarity). Moreover, we took steps to address issues of data quality such as screening and asking participants to report whether they belonged to the identified disadvantaged groups or not and including an attention check within the survey. Third, we examined intentions to engage in collective action as opposed to behaviours although we do note that intentions are a strong predictor of behaviour (Conner & Norman, 2022).
Our mediation models are grounded in previous research which has experimentally established shared experiences as an antecedent of intraminority solidarity (Cortland et al., 2017). However, we recognise that conducting a mediation analysis on cross-sectional data means that we are unable to establish the temporal order and causal relationship between the variables in these studies (Fiedler et al., 2018). We therefore encourage future experimental research to establish the temporal order and causal relationships between the variables examined in this paper.
Additionally, our measure of similarity measured an individual’s perceived similarity to another disadvantaged group, rather than prompting the participant to consider their multiple disadvantaged group identities when assessing perceived similarity with another disadvantaged group. Nevertheless, this measure was derived from and used in prior literature on intraminority relations (see i.e., Cortland et al., 2017; Craig & Richeson, 2012; Craig et al., 2012). Moreover, our similarity measure was a significant predictor of collective action intentions as found in previous research (Cortland et al., 2017; Pham et al., 2023). Due to practical constraints, we grouped participants who identified as LGBTIQ+ into a single category for analysis. However, we acknowledge that different subgroups within this identity dimension (e.g., gay men, lesbians) experience distinct forms of discrimination and therefore encourage future research to explore the unique experiences of these subgroups. We further acknowledge that our operational definition of multiple disadvantaged group membership is somewhat detached from the social context and how participants might view themselves. We encourage future research to undertake qualitative work to address this limitation. While beyond the scope of this paper there may be other plausible reasons for intraminority solidarity, such as anger about injustice or inclusive victimhood, and encourage future research to follow this up.
Finally, we acknowledge that our findings are restricted to understanding intraminority solidarity among multiple disadvantaged groups within the US and UK. We do, however, acknowledge that our operational definition of multiple disadvantaged group membership is somewhat detached from the social context and how participants might view themselves. We encourage future research to address this limitation by undertaking qualitative research to further understand how multiple disadvantaged group membership is tied to the social context and how participants view themselves. While we were limited to only examining advantaged and disadvantaged group membership across the identity dimensions of race, gender, and sexual orientation in this paper due to practical constraints, we encourage future research to investigate the role perceived similarity and intersectional awareness plays in predicting intraminority solidarity for other identity dimensions to obtain a comprehensive understanding of this relationship. In addition, context could shape the results due to the dynamic nature of identity including how it is perceived and treated within different social contexts. Attitudes towards the outgroups, people with disabilities and Muslims, can change over time based on real-world events. We encourage future research to explore how our findings may shift over time.
Our research provides an insight into how intersectional awareness might be more effective as a mechanism to increase intraminority solidarity for disadvantaged groups regardless of the dimension of identity due to the need to acknowledge and bridge connections between disadvantaged groups. For instance, perceived similarity fails to acknowledge the importance of group differences’ role in effectively fostering intraminority solidarity, in which the goal is social equity rather than equality. Rather than fostering intraminority solidarity, focusing on similarities can potentially undermine it (Ufkes et al., 2016). For instance, establishing a common ingroup identity between disadvantaged groups can foster intergroup harmony due to the positive contact it promotes. Ufkes and colleagues (2016) found that when a common American identity was highlighted, Black Americans and Latinx Americans reported lower intentions to engage in collective action compared to when their distinct ethnic or racial identities were made salient. This finding suggests that intersectional awareness may be more effective for fostering intraminority solidarity and advancing social equity. They further indicate that when positive contact through a common identity does not acknowledge differences amongst its group members, the concerns of those on the margins of the common identity are more likely to be overlooked.
In conclusion, the findings from both studies supported our initial hypothesis. Greater collective action intentions depended upon the number of disadvantaged groups participants belonged to, with those belonging to three disadvantaged groups reporting greater collective action intentions than those belonging to fewer or none. We found partial support for our second hypothesis. Perceived similarity did not mediate the relationship between belonging to multiple disadvantaged groups and collective action intentions; however, intersectional awareness fully and positively mediated the relationship. Our finding has significant implications for understanding and promoting collective action among disadvantaged group members who belong to multiple disadvantaged groups. They suggest that belonging to multiple disadvantaged groups can serve as a catalyst, increasing intentions to engage in collective action for another disadvantaged group via increased intersectional awareness. This insight could be leveraged to design more effective strategies for fostering solidarity and collective action among marginalised communities.