Does witnessing gender discrimination predict women's collective action intentions for gender justice? Examining the moderating role of perceived female support

Previous research has indicated that witnessing gender discrimination may instigate women's participation in collective action for gender justice. However, relatively little is known about the role of perceived female support in motivating collective action among women who witness gender discrimination in public life. This study aims to analyse whether and when perceived support from feminist-minded women moderates the association between women's witnessing gender discrimination and their willingness to engage in collective action for gender justice. We argue that the association between witnessing gender discrimination and willingness to engage in collective action depends on the support women perceive from their female friends and family members. In studies of women in the U.S. (Study 1; N = 271) and Ukraine (Study 2; N = 256), witnessing gender discrimination predicted greater willingness to participate in collective action for gender justice, and this association was stronger when female support was perceived to be lower. Study 3 ( N = 1,304) replicated the findings of Studies 1 and 2 with self-identified feminist women in Turkey. Our research offers novel insights regarding why perceived lack of female support may encourage women to engage in collective action for gender justice.

whether and when perceived support from feminist-minded women moderates the association between women's witnessing gender discrimination and their willingness to engage in collective action for gender justice. We argue that the association between witnessing gender discrimination and willingness to engage in collective action depends on the support women perceive from their female friends and family members. In studies of women in the U.S. (Study 1; N = 271) and Ukraine (Study 2; N = 256), witnessing gender discrimination predicted greater willingness to participate in collective action for gender justice, and this association was stronger when female support was perceived to be lower. Study 3 (N = 1,304) replicated the findings of Studies 1 and 2 with self-identified feminist women in Turkey. Our research offers novel insights regarding why perceived lack of female support may encourage women to engage in collective action for gender justice.

| INTRODUCTION
Efforts to promote gender justice persist in every corner of the globe. Yet, in many parts of the world, women still face systemic disadvantages in their societies that are veiled in impunity and deeply embedded discriminatory norms (United Nations, 2020). One possible means of addressing the persistent discrimination women face is promoting grassroots collective action to advance women's rights (Bhattacharjya, Birchall, Caro, Kelleher, & Sahasranaman, 2013;Radke, Hornsey, & Barlow, 2018).
We focus on the mobilizing potential of these two variables because structural and incidental forms of discrimination target women worldwide (Bergen, Edleson, & Renzetti, 2005). While news media typically cover high-profile cases of gender discrimination (e.g., Topping, 2021), there are also countless subtle and blatant incidents of hostility and discrimination against women as a group in many spheres of public life, such as in employment, education, politics, and public service (European Institute for Gender Equality, 2019). Moreover, gender discrimination and systemic violence against women have recently caused a wave of social protests across the world, including both in the Global North (e.g., Kende et al., 2020) and the Global South (e.g., Saldarriaga et al., 2020). The worldwide scale of mobilization aimed to promote and protect women's rights indicates that, under certain social-psychological circumstances, women may be motivated to take action rather than be passive observers when they witness acts of discrimination against other women.
The starting point of our analysis is the notion that witnessing can act as a catalyst from which mobilization for collective action and prosocial behaviour can begin (Staub, 1993; see also Curtin & McGarty, 2016;Mizock & Page, 2016;Techakesari et al., 2017;Tropp & Ulu g, 2019;Ulu g & Tropp, 2021). For example, in the context of Black Lives Matter protests, witnessing racial discrimination has been shown to foster White people's willingness to engage in collective action for racial justice to the extent that they had gained awareness of racial privilege (Ulu g & Tropp, 2021) and felt close to people targeted by prejudice (Tropp & Ulu g, 2019). One can argue that witnessing discrimination and a sense of perceived injustice may affect people's collective action tendencies (e.g., van Zomeren, Postmes, & Spears, 2008). We argue that witnessing discrimination and a sense of perceived injustice are theoretically similar yet distinct concepts. People may witness these discriminatory incidents without necessarily acknowledging that those incidents are discriminatory. We aim to show that mere exposure to discriminatory incidents may predict people's willingness to engage in collective action for justice.
These prior studies mentioned above examined what motivates support for solidarity-based collective action among members of historically advantaged groups (i.e., White Americans in the U.S. context). But, it is also pivotal to understand under which conditions women, whose group systematically faces structural disadvantage and discrimination, would be inclined to recognize the scale of gender injustice as an issue and take collective action to remedy it. Gender injustice may take different forms from more severe and pervasive forms such as sexual harassment to milder forms of discrimination such as widespread differences in societal expectations. While top-down attempts to resolve women's historical disadvantage do take place (Kelan, 2020;Menasce Horowitz & Igielnik, 2020), many women do not necessarily recognize gender discrimination as a persistent societal issue due to unconscious bias, beliefs about meritocracy and/or lack of transparency in institutions and thus may not see the need to take action to promote gender equality and justice (Maxmen, 2018). Building on prior research, we therefore argue that witnessing gender discrimination can encourage women to become more willing to take action to promote greater gender justice.
1.1 | The moderating role of perceived female support Furthermore, we seek to test whether the link between witnessing gender discrimination and women's willingness to engage in collective action for gender justice depends on the perceived availability of support from other women in their lives who stand for equal gender rights. We conceptualize perceived female support as support women receive from their social networks to take action to promote gender justice, and we argue that, like other forms of social behaviour, collective action aimed at addressing gender discrimination is likely to be shaped by descriptive group norms and contingent upon social validation from significant others (Prentice & Paluck, 2020). Therefore, the present research focuses on perceived female support as a variable that has promise in explaining the social psychology of collective behaviour among the disadvantaged (i.e., women). More precisely, we are concerned with the role that perceived support of feminist-minded female relatives and friends plays in this relationship.
Although mainstream models of collective action have recognized the importance of perceived social support in general, including that from ingroup members, for the emergence of social movement behaviour (e.g., see van Zomeren, Spears, Fischer, & Leach, 2004), they have not included such processes as central and enduring phenomena to be used in predicting one's willingness to engage in collective action for social change. The perceived availability (or thereof lack) of such support is an important individual-(e.g., significant others) and contextual-factor (e.g., communities and society at large) to examine, especially in the light of the recent call of placing the theory into context and moving beyond the role of perceived grievances and group identities as central explanatory constructs in collective action research (see Agostini & van Zomeren, 2021). Against this backdrop, in the current paper, we seek to analyse the role that the perceived availability of female support of (pro-)feminist significant others (e.g., family members and friends) plays in moderating the relationships between women's witnessing gender discrimination and their willingness to engage in feminist collective action.
Previous empirical studies have emphasized the role of support networks (e.g., parents, extended family, and/or close friends) in predicting people's participation in collective action (Gil de Zúñiga & Valenzuela, 2011), especially with respect to normative social influence and social validation by significant others (e.g., González et al., 2021).
However, these studies have ignored, for the most part, how the variability of perceived support within the larger society can determine women's willingness to become politically and civically engaged when they witness incidents of gender discrimination. At a theoretical level, this individual-level variable can have different impacts in motivating feminist collective action among women. On the one hand, according to the buffering hypothesis (Cohen & Wills, 1985), perceived peer support is a powerful resource that is capable of protecting disadvantaged group members, including women, against the deleterious effects of adverse social conditions. Perceived social support has been shown to reduce the psychological impact of one's exposure to stressful or otherwise problematic events at both the interpersonal-and the societal-levels (e.g., Gee et al., 2006;Mossakowski & Zhang, 2014;Noh & Kaspar, 2003). In particular, the perception of readily available support from other females was shown to provide women with one of the most significant environmental resources enabling them to better cope with the consequences of social injustice through active problem solving (e.g., Dennehy & Dasgupta, 2017;Holahan, Valentiner, & Moos, 1995). Therefore, one can expect that the perceived availability of female support can moderate the relationship between witnessing gender discrimination and willingness to engage in collective action for gender justice such that this association will be stronger for women who feel supported by their female network.
On the other hand, according to the resource model of political participation (Brady, Verba, & Schlozman, 1995;Smets & van Ham, 2013) as well as resource mobilization theory (McCarthy & Zald, 2001), it has been suggested that the perceived lack of social support in one's immediate social networks can mobilize individuals to engage in civic and political activities to foster a desired social change. When people witness discrimination and perceive their immediate networks (e.g., female friends and relatives) as inadequate or incapable of providing desirable support for expressing their political voice, they may choose to get involved in a larger social movement demanding social change along with ideologically similar social groups (e.g., Gil de Zúñiga & Valenzuela, 2011;Son & Lin, 2008). Lack of perceived female support in women's lives, or in other words, getting less support from other women who supposedly stand for equal gender rights, can signal women that: they need to rely on themselves in redressing gender inequality after they witness gender discrimination, especially in contexts where gender inequality is a more severe problem. Therefore, one can also argue that perceived lack of support (even) from feminist-minded family members and friends may mobilize women into collective action, thereby turning witnesses into actors. This perceived lack of support may actually motivate women to break a vicious cycle of systemic violation of women's rights on their own, without necessarily waiting for others to intervene.
Together, these studies constitute a sound theoretical basis for the notion that perceived female support is a crucial mechanism that can potentially explain the variation in women's willingness to act collectively for gender justice when they witness incidents of gender discrimination. Even though the link between witnessing discrimination and collective action has been established in previous studies among the advantaged (e.g., Ulu g & Tropp, 2021), the primary purpose of the current empirical research paper is to amend social psychological approaches to collective action research by systematically examining the extent to which perceived female support moderates the association between witnessing incidents of gender discrimination and willingness to engage in collective action for gender justice among the disadvantaged across different cultural contexts.

| Overview of studies
With samples of women from the U.S. (Study 1), Ukraine (Study 2), and Turkey (Study 3), we systematically examined the extent to which perceived female support moderates the relation between women's witnessing acts of gender discrimination and their willingness to engage in collective action for gender justice. Even though these countries differ in terms of gender equality and democracy levels, gender inequality is still a global problem in more democratic (e.g., the U.S.) and less democratic (e.g., Ukraine & Turkey) contexts. Therefore, we argue that witnessing gender discrimination can be an important motivator for women to take action irrespective of the characteristics of the context.
Growing from previous research suggesting that perceived social support can either help people cope with the negative effects of discrimination (e.g., Dennehy & Dasgupta, 2017;Gee et al., 2006;Holahan et al., 1995;Mossakowski & Zhang, 2014;Noh & Kaspar, 2003) or encourage action to fight against it (e.g., Gil de Zúñiga & Valenzuela, 2011;Son & Lin, 2008), we sought to test whether perceived support might strengthen or weaken the link between witnessing gender discrimination and willingness to engage in collective action for gender justice with general populations of women in the first two studies (U.S. and Ukraine). We also controlled for women's own reported discrimination experiences as prior research shows that being a target of discrimination often drives collective action participation (Foster & Matheson, 1998;Friedman & Leaper, 2010;Tropp & Brown, 2004).
Furthermore, because we focused on the moderating role of perceived support that women receive from their family members and friends who stand for equal gender rights, in Study 3, we aimed to replicate and extend beyond these first two studies with a feminist sample in Turkey. We chose a feminist sample to examine the associations among women who may have more female support in their lives and, thus, be more willing to engage in activism and collective action for gender rights compared to other subpopulations of women (Sweetman, 2013 2 | STUDY 1: UNITED STATES Study 1 tested our hypotheses in a sample of adult women in the U.S., which ranks 30th among 156 countries in global gender gaps (Global Gender Gap Report, 2021). The U.S. has a long-standing history of gender movements in the struggle for women's rights (England, Levine, & Mishel, 2020;Kelly, 2015), yet its social institutions still reveal gender disparities and discrimination in many public domains (e.g., Stoet & Geary, 2020;Vescio & Schermerhorn, 2021). It therefore presents an interesting case for us to examine the role of witnessing gender discrimination, perceived female support, and their interaction in women's willingness to engage in collective action for gender justice.

| Participants and procedure
After receiving IRB approval for this research, we recruited self-identified female participants in the U.S. They were asked to complete online surveys in English through Amazon's Mechanical Turk and received $1.25 USD as compensation for their participation. In total, 353 participants clicked on the survey link. We used a filter question (i.e., what is your gender) at the beginning of the survey and asked the same gender question as part of the demographic questions at the end of the survey. Therefore, participants who chose any choice other than "woman" at the beginning (n = 69) and the end (n = 13) were excluded, and we focused on 271 participants.
The order in which the measures were presented to participants was not randomized; however, the items within each scale were randomized. Participants first responded to questions related to support they receive from their friends and society, later to questions on how frequently they have witnessed or been a target of discrimination as well as questions about their collective action tendencies. In the end, they also answered some demographic questions.

| Witnessing gender discrimination
Eight items, including some adapted from Krieger, Smith, Naishadham, Hartman, and Barbeau (2005), assessed how often women reported having witnessed an incident of gender discrimination in their day-to-day lives, in which a woman was treated differently than other people would be treated (e.g., a woman has been treated with less courtesy than other people; a woman has been sexually abused or assaulted by a man). Responses to these items ranged from 1 (never) to 5 (very often; α = .87).

| Being a target of gender discrimination
We used a parallel set of eight items to assess how often women themselves reported being a target of gender discrimination in their day-to-day lives (e.g., I have been treated with less courtesy than other people; I have been sexually abused or assaulted by a man). Responses to these items ranged from 1 (never) to 5 (very often; α = .88).

| Perceived female support
We developed four items to assess perceived female support for gender justice. These items were: (a) I am surrounded by women who are active in protecting women's rights; (b) most women in my life support other women that stand for equal gender rights; (c) most women I am close to think it's important to address gender inequality and (d) most of my female friends support the idea of equal gender rights (1 = strongly disagree; 7 = strongly agree; α = .79).

| Willingness to engage in collective action for gender justice
Six items adapted from Oda g, Ulu g, and Solak (2016) assessed participants' willingness to promote gender justice by taking diverse forms of normative collective action: (a) protesting on the streets; (b) attending forums, meetings, or discussion groups related to gender equality; (c) signing a petition pertaining to women's issues; (d) changing their avatar picture to express their support to the cause for instance as #WomensMarch, #MeToo or #YesAllWomen; (e) protesting on social media such as Facebook and Twitter and (f) adding their name to email-signature campaigns. 1 Responses to these items ranged from 1 (never) to 5 (very often; α = .90).

| Preliminary analyses
Correlations among the variables and their means and standard deviations are presented in Table 1. Greater witnessing of gender discrimination was associated with greater perceived female support and greater willingness to engage in collective action for gender justice. In addition, higher levels of female support corresponded with higher levels of willingness to engage in collective action for gender justice.

| Moderation analysis
Data were analysed using SPSS Statistics 26.0 (IBM Corp., 2019) and PROCESS Macro v3.4 (Hayes, 2018). Moderation analysis (Model 1) was performed to examine whether perceived female support moderated the link between witnessing gender discrimination and willingness to engage in collective action for gender justice. We also included being a target of gender discrimination as a covariate in the model as a statistical control. 2 Below we reported the model that does not include being a target of gender discrimination as a covariate.
Simple slopes for the association between witnessing gender discrimination and willingness to engage in collective action for gender justice were tested at low (-1SD below the mean) and high (+1SD above the mean) levels of perceived female support. Witnessing gender discrimination was more strongly related to willingness to engage in collective action at low levels of perceived female support (B = .394, SE = .046, t = 8.65, p < .001, CI [.30, .48]) than higher levels of perceived female support (B = .228, SE = .035, t = 6.60, p < .001, CI [.16, .30]; see Figure 1).
The findings highlight that women who frequently witness gender discrimination and perceive less availability of female support are likely to show greater willingness to challenge gender inequality in the public sphere through collective action. These findings remained significant even after we controlled for participants' own discrimination experiences (see Foster & Matheson, 1998;Friedman & Leaper, 2010). Findings from Study 1, thus, suggest thatregardless of their own experiences with gender discrimination-the link between witnessing gender discrimination and willingness to take action against gender discrimination was stronger among women who perceived less support than those who perceived higher support.

| STUDY 2: UKRAINE
In an effort to replicate findings from Study 1 in a different cultural context, we conducted Study 2 in Ukraine, a country ranked 74th of 156 countries in global gender gaps (Global Gender Gap Report, 2021). The 1991 fall of the Soviet Union marked a new period in Ukraine, characterized by an ideological turn towards liberalization and democratization of Ukrainian society and the development of Western-style feminism (Plakhotnik, 2017 F I G U R E 1 The interaction of witnessing gender discrimination and female support on willingness to engage in collective action for gender justice in Study 1 international reports indicate that discriminatory gender practices remain high in many public domains, which reinforce existing barriers to women's advancement (Advocates for Human Rights, 2020).
6 | METHOD 6.1 | Participants and procedure In Study 2, self-identified female participants in Ukraine were recruited to complete online surveys in Ukrainian through convenience and snowball sampling. The questionnaire was originally constructed in English and then translated into Ukrainian by the study's bilingual collaborator in Ukraine. Similar procedures were followed in terms of randomization, but witnessing gender discrimination items were presented before female support and collective action items in this study.
In total, 379 participants clicked on the survey link. As in Study 1, we used a filter question (i.e., what is your gender) at the beginning of the survey and asked the same gender question as part of the demographic questions at the end of the survey. Participants who did not answer the gender question (n = 19), did choose any gender other than "woman" at the beginning (n = 17) as well as participants who did not fill out any of the study variables (n = 87) were excluded, and we focused on 256 participants.
Participants' ages ranged from 17 to 70 years (M = 33.12, SD = 12.28). Thirty-nine participants (18%) had a high school education (including vocational school) or less, 132 (62%) had completed higher education (undergraduate and graduate degree), 35 (16%) had a PhD degree or its equivalent, and eight (4%) chose "other." Forty-two participants did not respond to the question about their educational level. They also indicated a wide variety of responses across the political spectrum (M = 3.96, SD = 2.44), indicating a more liberal-leaning sample (1 = liberal; 11 = conservative).

| Preliminary analyses
Correlations among the variables and their means and standard deviations are presented in Table 2. Greater witnessing of gender discrimination was associated with greater willingness to engage in collective action for gender justice. In addition, greater perceptions of female support correlated positively with greater willingness to engage in collective action for gender justice.

| Moderation analysis
We used the same approach to statistical analysis as described in Study 1 and reported the model that does not include being a target of gender discrimination as a covariate. 3 Results indicated that the more women witnessed gender discrimination (B = .583, SE = .072, t = 8.13, p < .  Figure 2). A sensitivity power analysis also showed that with the available sample size (N = 256), we had 95% power to detect an effect size of Cohen's d = .04.
Taken together, Studies 1 and 2 show similar patterns of findings in the U.S. and Ukraine, two very distinct national contexts. The more women reported witnessing gender discrimination in their lives, the more willing they were to take action for gender justice. Intriguingly, across both studies, this link was consistently found to be stronger among women who reported lower (compared to higher) perceived female support.
Rather than being fueled by perceived support more generally, it is possible that women's willingness to engage in collective action could be bolstered by a sense of belonging to and support from a community of women who adhere to values congruent with feminist ideology (i.e., gender-based egalitarianism). Therefore, we conducted a third study with a sample of self-identified feminist women in Turkey to examine how perceived female support might function among women who tend to have more female support in their lives and who may be more inclined to engage in activism and collective action for gender rights (Sweetman, 2013;Uysal et al., 2022). This research extension is particularly relevant for social movements that aim to promote gender justice as it will help to understand how these processes function among women who are more likely to be engaged in collective action for gender justice than the more general female populations surveyed in Studies 1 and 2. F I G U R E 2 The interaction of witnessing gender discrimination and female support on willingness to engage in collective action for gender justice in Study 2

| STUDY 3: TURKEY
In Study 3, we tested our research hypotheses with a self-identified feminist sample of women in Turkey. The reason behind this choice was that compared to general female populations, feminists may (a) have witnessed more gender discrimination and be more aware of gender inequality (see, for example, Downing & Roush, 1985); (b) have been more active in taking action for gender equality (Sweetman, 2013) and (c) have formed (political) alliances with other women (see, for example, Downing & Roush, 1985;Hooks, 1986) and engaged in feminist solidarity (Wickström et al., 2021). This gave us an opportunity to test whether the trends we observed with the general female population with respect to perceived female support work in the same way among feminists as well as the results we found in the general female population are generalizable to a more politicized sample like feminists.
As of 2021, Turkey was ranked as having the 133rd largest gender gap among 156 countries (Global Gender Gap Report, 2021). While international observers have documented some progress with respect to gender equality in Turkey in recent decades, the country's conservative religious groups and right-wing populist parties are committed to restoring the gendered status quo by strengthening patriarchal attitudes and traditional gender roles (see Chayinska, Ulu g, Solak, Kanık, & Çuvaş, 2021;Kabasakal-Arat, 2020). 9 | METHOD 9.1 | Participants and procedure As in Study 2: (a) the survey was translated by the study's bilingual collaborator in Turkey and conducted in Turkish; (b) similar procedures were followed in terms of randomization and (c) witnessing gender discrimination items was presented before female support and collective action items in this study. In total, 1,668 participants, who were recruited to complete online surveys through both convenience and snowball sampling via Facebook, Twitter, and WhatsApp, clicked on the survey link. As in Studies 1 and 2, we used a filter question (i.e., what is your gender) at the beginning of the survey. In addition, participants were also asked if they identified as a feminist, and they were only allowed to complete the survey if they responded affirmatively. Participants who did not accept to participate in the survey (n = 11), did not answer the gender question (n = 20), did choose any gender other than "woman" at the beginning (n = 79), did not self-identify as feminist (n = 62) as well as participants who did not fill out any of the study variables (n = 192) were excluded, and we focused on 1,304 self-identified female and feminist participants.
Participants' age ranged from 17 to 63 years (M = 28.19, SD = 7.77). Two hundred eighty-three participants (27%) had a high school education or less, 686 (66%) had completed higher education (undergraduate and graduate degree) and 71 (7%) had a PhD degree or its equivalent. Two hundred sixty-four participants did not respond to the question about educational level. When describing their socio-economic status, 39 (4%) identified as lower class, 226 (22%) as lower-middle class, 583 (56%) as middle class, 164 (16%) as upper-middle class, and 27 (2%) as rich or upper class and 265 participants did not respond to this question. They also indicated a wide variety of responses across the political spectrum (M = 1.67, SD = .72), indicating a more leftist sample (1 = left; 5 = right).
10 | RESULTS AND DISCUSSION

| Preliminary analyses
Correlations among the variables and their means and standard deviations are presented in Table 3. Greater witnessing of gender discrimination was associated with greater perceived female support and greater willingness to engage in collective action for gender justice. In addition, greater perceived female support corresponded with greater willingness to engage in collective action for gender justice.

| Moderation analysis
We used the same approach to statistical analysis as in Studies 1 and 2 and reported the model that does not include Overall, the findings observed in Study 3 with a sample of self-identified feminist women in Turkey replicated the patterns of results observed in Study 1 (U.S.) and Study 2 (Ukraine). The positive association between witnessing gender discrimination and willingness to engage in collective action for gender justice was stronger among women who perceived less (vs. more) female support for mobilization around women's gender rights issues.

| GENERAL DISCUSSION
Even though there has been some progress in establishing women's rights in different domains (Kelan, 2020; Menasce Horowitz & Igielnik, 2020), many women still do not necessarily see or accept that gender discrimination against their gender group exists or take action to challenge gender inequality (Maxmen, 2018). Therefore, we argued that witnessing gender discrimination can encourage women to become more willing to promote gender justice. In the current research, we sought to contribute to the collective action literature on gender discrimination (e.g., Foster & Matheson, 1998;Friedman & Leaper, 2010;Sweetman, 2013;Ulu g et al., 2020) by examining whether witnessing incidents of gender discrimination among women predicts their willingness to engage in feminist collective action. We also examined whether (a) this association was moderated by women's perceived availability of female support from other close women who stand for equal gender rights and (b) remained significant when controlling for women's personal experiences of gender discrimination.
Our findings from three studies consistently showed that witnessing gender discrimination was a significant predictor of women's willingness to engage in collective action for gender justice not only among general women samples in the U.S. (Study 1) and Ukraine (Study 2), but also among self-identified feminist women in Turkey (Study 3).
We should also note that the direct association remained significant after controlling for women's personal experiences of gender discrimination across three studies.
The current research supported the previous findings that witnessing social injustices in the public sphere can instigate individuals' collective action tendencies (e.g., Mizock & Page, 2016;Tropp & Ulu g, 2019;Ulu g & Tropp, 2021). The unique contribution of our studies was that perceived female support moderates the relation between witnessing and willingness to take action such that it becomes stronger for women that reported to have lower (vs. higher) female support in their lives. Our results should not be interpreted as perceived female support hindering women's desire to take action against gender discrimination. We rather show that even though female support (both at low & high levels) motivates women to take action against gender discrimination, the link between witnessing gender discrimination and willingness to engage in collective action for gender justice is stronger when perceived female support is lower. A potential explanation for this result may be related to a possible ceiling effect in the values of willingness to engage in collective action when women receive high female support. There may already be high willingness to take action against gender discrimination, especially among feminists, which may make it difficult to increase significantly by the effect of witnessing gender discrimination.
Interestingly enough, the moderation effect was found to be generalized to three different cultural contexts: U.S., Ukraine, and Turkey. Even though these countries differ in terms of gender equality and democracy levels, gender inequality is still a global problem in established and developing democracies, let alone autocratic contexts.
Numerous studies conducted on collective action have included participants from Western, educated, industrialized, rich, and democratic (WEIRD) societies, such as the U.S., the Netherlands, the U.K., and Germany (Henrich, Heine, & Norenzayan, 2010), all of which constitute democracies. However, to understand the barriers to and facilitators of social change, it is also essential to examine collective action in hybrid democracies (e.g., Turkey) and autocracies (e.g., Russia, Venezuela), in which collective action participation has been associated with high costs and high risks (e.g., Ayanian et al., 2021; Oda g, Ulu g, Kanık, & Maganic, 2022). Therefore, future studies should focus on how F I G U R E 3 The interaction of witnessing gender discrimination and female support on willingness to engage in collective action for gender justice in Study 3 perceived costs and risks may play an important role in motivating women to take action against gender discrimination in more risky contexts.
Taken together, these findings call for a further investigation of the role of perceived availability of both peer and family support for grassroots social activism to better understand when and why perceived (lack of) support can motivate collective action (e.g., Cohen & Wills, 1985;Dennehy & Dasgupta, 2017;Gee et al., 2006;Holahan et al., 1995;Mossakowski & Zhang, 2014;Noh & Kaspar, 2003), rather than impede it (e.g., Gil de Zúñiga & Valenzuela, 2011;McCarthy & Zald, 2001;Smets & van Ham, 2013;Son & Lin, 2008). Women may be more mobilized and willing to take action against gender discrimination after they witness gender discrimination. However, that is especially the case when the support they receive from other women, who supposedly stand for equal gender equal rights, is relatively lower. We believe that these results may be used by non-governmental gender organizations and feminist movements that aim to recruit more women. They may target women who do not have female support in their lives and provide information about how their organizations and movements can provide this support to women to better fight against gender discrimination (e.g., "you are not alone campaigns").

| Limitations and future directions
Our studies have a few limitations. First, we collected data using Amazon's Mechanical Turk (MTurk) in Study 1. Even though the sample on this platform is relatively diverse in terms of age and income, it is still not representative of the U.S. population as a whole (Ross, Zaldivar, Irani, & Tomlinson, 2010). Moreover, in recent years, there has been some concern over scholars' reliance on the data from MTurk, especially with regard to participants falsely reporting their gender (e.g., women instead of men) to qualify for a study (Sharpe Wessling, Huber, & Netzer, 2017). To account for these concerns, we took a conservative approach by using filter questions in Study 1. We collected data from community samples in Studies 2 and 3 with a cross-sectional design; however, we should note that these samples are not representative either. Therefore, the cultural generalization of our results should be interpreted with caution. Cross-cultural research employing representative samples may therefore offer more robust information.
Second, the cross-sectional design of our studies does not allow us to make any causal inferences about our findings. Therefore, future experimental or longitudinal research should examine whether women's personal adverse experiences of gender discrimination (i.e., experiences of victimization) precede their greater awareness of genderbased inequalities. It is also possible that women's adherence to values congruent with feminist ideology (i.e., gender-based egalitarianism) and their identification with social categories such as women and feminists (e.g., Uysal et al., 2022;van Breen, Spears, Kuppens, & de Lemus, 2017) can act as a boundary condition for greater perceived awareness of gender inequalities and greater willingness to engage in grassroots activism.
Third, as mentioned earlier, the measures that were presented to participants were not randomized. Even though the items within each scale were randomly presented, we presented the witnessing gender discrimination scale, female support scale, and willingness to engage in collective action scale in the same order (except in Study 1).
The order effects might have contributed to people's willingness to engage in collective action in these correlational studies (see Şahin, 2021, for a recent discussion on item order). Future studies should randomize not only the items within the scales but also the scales themselves.
Fourth, we did not consider the role of intersectional discrimination in women's protests tendencies. Women from marginalized backgrounds may not be able to distinguish the ways in which they are discriminated against on the basis of gender from the ways they are discriminated against on the basis of race, class, disability, etc. In our studies, the way Black women in the U.S. or Kurdish women in Turkey see and interpret discrimination may have differed from White women and Turkish women in the same contexts. Using the intersectionality approach, future research should focus on how multiple and overlapping social identities shape both one's capacity to identify, acknowledge, and condemn gender discrimination and willingness to take action aimed at eroding it in our societies (e.g., Greenwood, 2008).
Last, we quantified perceived female support by asking respondents to what extent they perceived support from other women in their lives who stand for gender equality. Female support received from feminist-minded significant others should not be considered a proxy for general social support stemming from other women in society. This concept rather suspects that women might be surrounded by other females who think it is important to redress gender inequality (however, in the current research, we did not include the measure allowing us to detect whether this group of women constitutes a minority or majority of respondents' female contacts). Future research in this line should therefore examine whether the perceived consensus about the need to intervene among feminist-minded friends when witnessing discrimination can act as a catalyzer of collective action. In our own research, we conceptualized perceived female support as support women receive from their social networks to take action to promote gender justice. This conceptualization, however, might pose some limitations for our understanding of whether (a) participants personally received support when they needed it and (b) significant female others recognized the need and importance of challenging gender inequalities. Future research should also account for the disempowering effects of women's adherence to sexist ideologies and system-justifying beliefs that were shown to impede a wide range of social and political behaviours (e.g., Chayinska et al., 2021;Jost & Kay, 2005;Radke et al., 2018). Future qualitative studies may be helpful in understanding what female support may entail for women that represent different social and cultural backgrounds.

| Conclusion
The cases of gender-based discrimination, subjugation, and exploitation can be normalized and further feed and breed gender inequality in our societies if gender inequality is not instantly and adequately addressed through persuasive collective action. Our findings highlight the role of perceived female support as an important factor that may strengthen or weaken the relationship between witnessing gender discrimination and collective action.
Before we conclude the paper, we would like to highlight that it is not only the feminists' or women's responsibility to dismantle systemic oppression and empower themselves to create change for gender justice, but also men's responsibility as the perpetrators of gender inequality who benefit from it. We hope that our results show how female support may have the potential to shape women's movement in general as well as specific collective actions against gender discrimination initiated by different groups in particular.