Multi­level gains of fat activism and their impact on sustained activism for fat justice

Previous research has indicated that outcomes of collective action can occur at the individual, group, and societal levels. Taken together, we argue that multi-level outcomes can influence sustained involvement in social movements. We aimed to examine the multi-level outcomes of fat activism across two studies. In our first study, we conducted semi-structured interviews with fat activists ( N = 20) to learn what they believe are the multi-level outcomes of fat activism. At the individual level, activists reported greater health, well- being, and self-esteem; at the group level, they reported a sense of community and increased clothing options; and at the societal level, they reported change in toxic cultures around dieting. By building on the findings of Study 1, Study 2 ( N = 464) aimed to understand how fat individuals’ past collective action participation may predict their future collective action participation through individual-, group-, and so- cietal-level gains. Results indicate that greater collective action participation in the past predicts greater willingness to engage in collective action through the pathway of higher beliefs in individual and societal gains of fat activism, but not through group-level gains, even after we control for identification with fat and fat activist identities. We discuss these findings in relation to the importance of multi-level outcomes in collective action and sustained involvement in social movements.


Introduction
Fat activism fights against anti-fat bias and works toward change in social attitudes on multiple dimensions (see (Cooper, 2016); (LeBesco, 2004). Fat activism has a strong online presence, with one large part of the activism focused on changing perceptions of fat bodies. Through these platforms, activists can also bring attention to diverse, intersectional issues related to fatness, including feminism, racism, classism, homophobia, and transphobia (Pausé, 2014).
This research addresses fat activism as both a relatively understudied kind of activism in social psychology and a movement that deliberately works toward achieving multiple goals. Our previous research has indicated that outcomes of collective action can occur at the individual, group, and societal levels (Uluğ & Acar, 2018). Individual outcomes can include identity politicization, a sense of empowerment, and increased subjective well-being (see Vestergren et al., 2017). Group-level change focuses not only on achievements for the ingroup, but also on how group members view one another (intragroup changes), and how groups view other groups (intergroup changes). Societal-level change (or policy change) is less addressed in psychological research but focuses on the way society can be transformed by collective action -especially relevant when the goals of collective action are to produce large-scale societal change (see, e.g., Louis, 2009). In the current contribution, we examine the effects of multi-level outcomes of fat activism across two studies. We focus on the impact that multi-level outcomes have for sustaining engagement with fat activism, creating the necessary momentum for fat activism to continue as a social movement working for positive change for fat individuals, fat people as a group, and society in general.
2012 ;Tarrow, 1988). While outcomes of collective action are underresearched (Louis, 2009), we believe that they are vital to understanding long-term, sustained commitment to collective action, and in particular social movements. The individual-level outcomes of collective action are the most clearly noted, including changes in identity, knowledge, and self-reported well-being (see Vestergren et al., 2017) for a review of outcomes of participation).
Group-level outcomes have also been noted. Intragroup changes may occur; some groups may politicize after participating in protest (Simon & Klandermans, 2001), with group members forging newer, stronger bonds. Intergroup changes may also occur; especially if they are participating in a protest with other groups, they may develop intimacy with other groups participating in the protest. As a result, trust and respect can increase, and the groups are more likely to help and support one another.
In addition to these important outcomes, our previous work (Uluğ & Acar, 2018) has focused on individual-level, group-level, and societal-level changes in the context of the Gezi Park protests in Turkey. Activists noted individual-level changes, such as feeling empowered, higher well-being, and identity politicization. Activists mentioned group-level changes either as (a) intragroup relations (i.e., changes within a group, such as changes in attitudes towards other protesters within the same movement), or as (b) intergroup changes (i.e., changes in attitudes towards other disadvantaged groups or social movements). Societal-level changes are often related to policy-level outcomes of collective action. We also posit that societal-level changes, such as a change in the way society as a whole views a particular group (e.g., through a change in stereotyping, prejudice, and discrimination toward the group), are also relevant at this level. In the case of individual-, group-, and societallevel changes, protesters at Gezi Park discussed changes in terms of gains, what sorts of net positives they experienced as a result of their participation, as well as losses, describing what, in the long run, they felt they had lost due to their participation.

Sustained involvement in social movements
Sustained involvement is essential for any social movement to succeed. It is essential for any social movement to find ways to motivate individuals to continuously engage in collective action on behalf of the social movement. Participation can forge new relationships and have a bolstering effect for identity, and individuals may take on various types of roles (e.g., organizing) as they get further politicized and involved in a social movement (see, e.g., Saunders et al., 2012). As well, feelings of empowerment can be an outcome of protest participation, as well as motivate for future participation (Drury and Reicher, 2005). Interestingly, though, much of the focus of collective action research in social psychology is focused on the motivations or antecedents for participation (e.g., Thomas et al., 2012;van Zomeren et al., 2008). We believe this is due to collective action being taken as singular events, rather than multiple, long-term events toward a singular goal, as they would be in a social movement. In this research, we consider the way that different types of outcomes -and specifically gains -of past participation can predict future participation in collective action and social movements.
Previous research has indicated that psychological change experienced during collective action can motivate sustained and extended involvement (see Vestergren et al., 2018), meaning that participation can motivate continued involvement in the same, or sometimes in other causes and struggles. We add to the discussion on sustained involvement in collective action by examining the ways that different outcomes to collective action participation can relate to or motivate continued participation. Specifically, we do so through fat activism for the fat acceptance movement.

Fat acceptance and fat activism
The fat acceptance movement began in the late 1960 s with groups such as the National Association to Advance Fat Acceptance and the more radical Fat Underground (Wann, 2009). While there is no single perspective in fat activism, fat acceptance aims for just that; the acceptance of fat bodies in societies where they are otherwise medicalized (e.g., the "obesity epidemic") and "moralized" (i.e., being fat is related to moral deficit, see Fikkan & Rothblum, 2012). Fat activism also recognizes that fat oppression, similar to other oppressions, impacts all people regardless of size, as it determines and expects people to adhere to a constructed "normal" (McCrindle, 2018). Therefore, fat activism recognizes that as a form of oppression, fat oppression needs to be tackled from different levels -changing the way that individual fat people see and understand their bodies, changing the way that fat people are treated as a group, and changing the way society constructs a "normal" body, and dehumanizes those larger than the purported normal (see (Cooper, 2016).
Previous research into the "fatosphere," or fat acceptance blogging community, suggests that engaging with fat acceptance created a sense of empowerment, social connectedness, and support in dealing with discrimination (Dickins et al., 2016(Dickins et al., , 2011. With a social psychological perspective, we aim to add to the discussion around tackling fat oppression by understanding how this multi-level approach can sustain involvement in activism for fat justice.

The current study
The current research has two aims. First, it seeks to understand the way fat activism shapes multi-level outcomes from the perspective of fat activists. While previous research has indicated that there are both positive and negative outcomes that can come from participation in collective action (see Uluğ and Acar, 2018), for the purpose of this study, we focus on the positive outcomes, or gains, from participation, especially as we seek to understand the positive motivations for continued participation. Second, this research aims to examine how participation in fat activism may predict future collective action participation for fat justice among fat individuals through the endorsement of fat individuals' gains at the (1) individual-, (2) group-, and (3) societal-level after their protest participation. In order to address these aims, in our first study, we conducted interviews with fat activists to learn about perspectives on the outcomes of their activism. Using the findings of the first study, Study 2 examined the impact of multi-level outcomes as they relate to past and future collective action participation. Controlling for fat (activist) identity, we hypothesized that past collective action participation in fat activism would predict willingness to engage in future collective action through greater gains at multi-levels. Together, both studies are expected to contribute to the literature by understanding the multi-level outcomes of fat activism in particular, and the influence of these outcomes on (future) participation in social movements in general.

Participants and procedure
We received IRB approval for this research from the University of Massachusetts Amherst. Interviews were conducted with 20 activists who live in the US, the UK, Canada, and New Zealand. One of 20 participants self-identified as male and 12 as female. Seven participants self-identified in other ways. Ages of participants ranged from 22 to 42 (M = 34.10; SD = 8.45). Six participants identified as straight, four bisexual, six queer, and four pansexual. In terms of socio-economic status, five selected lower class, two lower-middle class, seven middle class, two upper-middle class, and four did not respond to this question.
We reached out to participants via email and social media; a snowball method was used to reach further participants. All participants were interviewed individually online via Skype. Participants were asked to reflect on the outcomes of fat activism at the individual level. They first answered these questions: How have you personally benefited from this activism? What have you felt change for you, physically, mentally, emotionally? What have you learned? What tools does this network/movement provide you to help you deal with fat oppression? Second, they were asked to reflect on how fat people benefit from fat activism. Last, they were asked to answer how society benefits from fat activism. After the study was over, the participants were debriefed, thanked and compensated for their participation ($25). As in our previous research with activists (Acar & Uluğ, 2016), we continued to conduct interviews until the data was saturated (Matthes et al., 2017), and no new themes were identified. We focused here especially on the dynamics between individual-, group-and societal-level outcomes of fat activism, as these were our most important variables. Fat activists started to converge in their depictions of these variables and began to add little to no new insight during the last interviews. It was at this point of convergence that we stopped sampling further participants. The anonymized raw data, coding frame, and all questions used to collect data are publicly available via the Open Society Framework (OSF) webpage: https:// osf.io/yav4n/?view_only=b662e59adec74cb384044d7796528318.

Analysis
All the transcriptions were analyzed using Qualitative Content Analysis (QCA; (Schreier, 2012). QCA has some benefits. First, it "provides for a deep, contextual, thick reading of content, and helps us recognize the nuanced relationship among variables" (Ostertag & Ortiz, 2014; p. 16). Second, it helps "to find patterns, themes, and categories that are important to social concerns" (Kumar et al., 2020), p. 5).
We followed the eight steps of QCA (Schreier, 2012): (1) selecting the material, (2) building a coding frame, (3) dividing the material into units of coding, (4) trying out the coding frame with a second coder, (5) evaluating the coding frame, (6) modifying the coding frame, (7) carrying out the main analysis, and (8) interpreting and presenting the findings. After selecting the relevant responses, such as a word or paragraph related to the answers to the research question, we systematically described the selected data. Coding was accomplished by a team of research assistants. Team meetings were carried out to establish consistency and consensus (see also (Creswell, 2013).

Results
Results below are presented based on the responses to interview questions. Each main category is an aspect of the data that we wanted to focus on during analysis. The main categories contain subsequent subcategories, which are what participants had to say about each of the main categories. The individual-level code was used for instances where people explicitly referred to themselves. We used the group-level code for every reference to fat people in society. Using this approach, references to "I gained a sense of community" would be coded as individual-level whereas references to "fat people gained a sense of community" would be coded as group-level. However, references to individuals (e.g., "I think it is the exposure that you [fat people] have permission, you [fat people] have permission to not want to change your body") to explain the situation of fat people would be coded as group-level as the emphasis is on fat people as a group. Similarly, we used the system-level code for any general reference to society (see (van Bezouw et al., 2019) for a similar application).

Main Category 1: Outcomes of fat activism for individuals
All participants talked about the different ways they individually benefited from fat activism. We coded these benefits under individual-level outcomes if they benefited from fat activism as an individual. These benefits included (1) gaining a sense of self (frequency 6), and (2) gaining a sense of community (frequency 18), (3) feeling healthier (frequency 1), (4) getting gratification and being grateful (frequency 2), (5) increased awareness (frequency 12), (6) building strength and confidence and feeling empowerment (frequency 12), (7) increased self and body acceptance (frequency 8), (8) becoming influential (frequency 9), (9) eating without guilt (frequency 2), and (10) learning new things (frequency 3). For example, participant #2 mentioned how fat activism benefited themselves by saying, "first and foremost [fat activism] provides me inspiration to be myself." Similarly, participant #3 thinks that fat activism helps people love themselves the way they are: "…there is like this beautiful intimacy, and this like self-love that comes as a result of that intimacy." Participant #18 also mentioned feeling overall healthier: I feel more different as far as my health. … So I think afterwards I am still figuring out what health was outside of what I am told about my weight, I started to take myself more seriously as far as my mental health, my physical health, and started to actually take it seriously.
Another important outcome of fat activism for individuals was gaining strength and confidence. As can be seen in Participant #11′s words, gaining strength and confidence was one of the most frequent benefits of fat activism: … there are aspects to not just like me personally becoming [a] more confident and stronger person, but also like physically I think I see it in myself like when I am in the community, but I find myself advocating for myself a lot more and that I am not as scared as I used to be and that if there are ever to be issues that come up from you know trying to advocate for myself to have like, to be in certain kinds of spaces that, you know, if something were to happen I feel like I could take it back and, kind of like get advice from people on how to move forward.

Main category 2: Outcomes of fat activism for fat people
Participants talked about the different ways fat people as a group benefited from fat activism. These benefits were 1) exposure to the idea that being fat is okay (frequency 5), (2) gaining a sense of community (frequency 3), (3) providing other narratives (frequency 3), (4) providing clothing options (frequency 8), (5) providing a platform/validation (frequency 6) and (6) apolitical fat people are left behind (frequency 2). For example, participant #4 argues that, through fat activism, fat people get exposed to the idea that being fat and loving oneself are not mutually exclusive: "I think it is the exposure that you [fat people] have permission, you [fat people] have permission to not want to change your body." We should note that this exposure is not always positive, though, as it sometimes comes with fatphobia: "… so I think that it definitely gives them [fat people] more insights like what the real deal is, on like how fatphobia, honestly hurts people, and even sometimes can kill people" (Participant #5).
Fat activism also provided a sense of community to fat people. Other benefits included but were not limited to providing other narratives, clothing options as well as a platform/validation. However, one negative outcome associated with fat activism is the feeling that apolitical fat people are left behind.

Main Category 3: Outcomes of fat activism for society
When asked about outcomes of fat activism for society, the participants' responses overlapped to some extent. These outcomes included (1) body liberation (frequency 7), (2) ending toxic cultures such as dieting and cosmetic surgery (frequency 6), (3) people feeling less fear about being/becoming fat (frequency 4), and (4) trying to devastate the impact of fatphobia (frequency 1). Participants mentioned that fat activism showed people that being fat, unapologetic, and successful is possible and creates a different public imagination model for fat people. For example, in participant #20′s own opinion, "fat activism showed people that fat people could live amazing, adorable, cute, thriving lives." Other positive outcomes of fat activism were ending toxic cultures such as dieting, cosmetic surgery, and social media advertising, as mentioned again by participant #20: … Weight Watchers is slowly, publicly, like, falling into its own grave. I don't think that would be happening without fat activism. … Before fat activism was visible on social media, then the nation and, like, at large was completely ignorant of the fact that you did not have to diet.
Similarly, one participant (#18) stated that fat activism has changed "how people fear fatness so much" because "there is the fear and the hatred that poisons people and poisons relationships and … makes life worse for everyone" (participant # 9). However, even though a vast majority of activists discussed different outcomes of fat activism changing norms in society, we should also note that two participants (#1 & #5) argued that fat activism has not yet influenced society.

Discussion
Study 1 allowed us to understand the way that fat activists themselves understand multi-level outcomes of fat activism. Fat activists described their outcomes mostly in terms of "gains" rather than any negative outcomes they may experience from their activism. They noted positive outcomes at the individual-, group-, and societal-level. In line with our expectations, participants pointed out that fat activism has improved their health, well-being, and selfesteem. Notably, participants described fewer outcomes at the group level and included some losses as well. Gains included a sense of community and an increase in clothing options, but some participants noted that apolitical fat people can be left behind. While participants noted the community they have gained by connecting to other self-identified fat people, they also noted that fatphobia as a construct has prevented many apolitical fat people from connecting with others and changing their views of themselves. Societal gains included body liberation and fighting against diet culture. In fighting against toxic cultures such as diet culture, they saw fat activism as removing barriers for everyone, not just fat people, when it came to fears around their bodies that prevented them from living their lives fully.

Study 2
In Study 2, we aimed to understand the relationship between multi-level outcomes and fat activism participation. Based on responses to Study 1, in Study 2, we chose to describe multi-level outcomes as multi-level gains, and examined how these gains may be predicted by fat individuals' past collective action participation in fat activism, which in turn, may predict their continued participation in fat activism in the future. In Study 2, we also controlled for participants' identification with (a) fat identity and (b) fat activist identity given prior research has shown that members of disadvantaged groups are usually more motivated to engage in collective actions that would benefit their group (e.g., Uluğ et al., 2020; van Zomeren et al., 2008) and this is even more likely when this disadvantaged identity is politicized (e.g., Simon and Klandermans, 2001).

Participants and procedure
We received ethical approval for this research from the University of Dundee. We collected data using snowball and convenience samplings. Some activists who participated in Study 1 helped us reach out to a larger sample by sharing the link to the survey on their social media platforms. The original sample consisted of 479 participants from different parts of the world including the US, the UK, the Netherlands, Sweden, Canada, and Australia. Participants who were younger than 18 years old (n = 3) and 12 univariate outliers were identified for each variable (5 on individuallevel gains; 1 on group-level gains; 1 on societal-level gains; 5 on willingness to engage in collective action) and dropped from the dataset. The remaining sample consisted of 464 participants. Twenty-six respondents self-identified as male, 314 as female, 118 preferred to self-describe as non-binary, gender-fluid, and agender, and six did not respond. Respondents' ages ranged from 18 to 60 years (M = 29.92; SD = 7.96).
To make sure that we reached out to fat individuals as well as an activist sample similar to Study 1, we asked whether (a) being fat and (b) being a fat activist is something that reflects who they are. Out of 464 participants, only 45 (10.4 %) participants surveyed (completely) disagreed with the item "being fat is something that reflects who I am" and only 77 (14.5 %) participants surveyed (completely) disagreed with the item "being a fat activist is something that reflects who I am." So, we believe we could reach out to self-identified fat individuals who tend to be politicized as fat activists.

Measures 3.1.2.1. Past collective action participation for fat justice.
We used five items adapted and inspired by Acar (Acar, 2018) and Odağ et al. (2016): These items were to protest prejudice against fat people, (1) I shared a statement on social media, (2) I shared a photo on social media, (3) I joined a social media campaign, (4) I boycotted a store/ company, and (5) I attended a peaceful demonstration/meeting. Responses to these items ranged from 1 (never) to 5 (always; α = 0.70).

Individual-level gains.
We created five items 1 based on the findings of Study 1: When I think about fat activism's impact on me as an individual, (1) I think I gained a community, (2) There have been improvements to my well-being/health, (3) I feel more confident, (4) I feel more accepting of my body, and (5) I feel empowered. Responses to these items ranged from 1 (completely disagree) to 7 (completely agree; α = 0.90).

Group-level gains.
We created three items based on the findings of Study 1: When I think about fat activism's impact on fat people as a group, I think fat people have (1) gained a sense of community, (2) improved options for clothing, and (3) improved treatment from medical professionals. As the first item lowered the internal consistency of the scale substantially, we dropped it. Responses to the two items were moderately associated (Spearman-Brown coefficient = 0.64) and averaged prior to data analysis. Responses to these items ranged from 1 (completely disagree) to 7 (completely agree).  We created four items 2 based on the findings of Study 1: When I think about fat activism's impact on society, I think fat activism has (1) helped to liberate fat bodies, (2) raised awareness on toxic cultures, such as diet culture, cosmetic surgery, or body oppression, (3) helped fight against the patriarchy and (4) helped fight against capitalism. Responses to these items ranged from 1 (completely disagree) to 7 (completely agree; α = 0.81).
3.1.2.5. Willingness to engage in collective action for fat justice. We used the same five items we used to measure past collective action participation by changing the wording (e.g., I would share a statement on social media to protest prejudice against fat people). Responses to these items ranged from 1 (completely disagree) to 7 (completely agree; α = 0.80).
3.1.2.6. Identification with fat identity. Using five items adapted from Hains et al. (1997), we measured participants' identification with fat identity to be used as a control variable in data analysis. These items were "being fat is something that reflects who I am," "being fat is important to me," "I have a lot in common with other fat people," "I have a strong tie with other fat people," and "being fat is an important part of my self-image." Responses to these items ranged from 1 (completely disagree) to 7 (completely agree; α = 0.82).

Identification with fat activist identity.
Using three items adapted from Hains et al. (1997), we measured participants' identification with fat activist identity to be used as a control variable in data analysis. These items were "being a fat activist is something that reflects who I am," "being a fat activist is important to me," and "being a fat activist is an important part of my self-image." Responses to these items ranged from 1 (completely disagree) to 7 (completely agree; α = 0.93).

Results and discussion
Correlations among the variables and their means and standard deviations are presented in Table 1. All study variables were positively associated except the association between group-level gains and willingness to engage in future collective action, identification with fat and fat activist identities.
We conducted a mediation analysis using PROCESS Model 4 with 5000 bootstrapped samples (see Hayes, 2013) in order to test whether (a) individual-level gains, (b) group-level gains, and (c) societallevel gains mediate the relationship between past collective action participation and willingness to engage in collective action. We conducted this analysis both with and without including identification with (a) fat identity and (b) fat activist identity as covariates in the model. As these tests yielded similar results, we have reported the full model that does not include these two variables as covariates in the model below. Results indicated that collective action participation in past protests was a significant predictor of (a) individuallevel gains, b = 0.385, SE = 0.062, p < .001, and (c) societal-level gains, b = 0.302, SE = 0.057, p < .001, but not of b) group-level gains, b = 0.038, SE = 0.066, p = .420.
The results also showed that (a) individual-level gains, b = 0.241, SE = 0.036, p < .001, (b) group-3 level gains, b = −0.172, SE = 0.033, p < .001, and (c) societal-level gains, b = 0.128, SE = 0.040, p = .003, were significant predictors of willingness to engage in future collective action. Past collective action participation remained a significant predictor of willingness to engage in collective action even after all types of gains were included in the model, b = 0.435, SE = 0.048, p < .001.
The indirect effect was tested using a bootstrap estimation approach with 5000 samples; results indicated a significant indirect association between collective action participation in past protests and willingness to engage in collective action, through ( As an exploratory analysis, we further conducted a moderated mediation analysis using PROCESS Model 58 with 5000 bootstrapped samples. Specifically, we investigated the moderating effect of fat identification on the relationships between past collective action participation on the one hand and individual-, group-and societal-level gains on the other. With respect to individual-level gains, the results indicated that the interaction between past collective action and fat identification was not significant, b = −0.071, SE = 0.041, p = .083, suggesting fat identification does not moderate the effect of past collective action participation on individual-level gains. With respect to group-level gains, the results indicated that the interaction between past collective action and fat identification was significant, b = 0.083, SE = 0.039, p = .032, suggesting fat identification moderates the effect of past collective action participation on group-level gains. Past collective action participation was not significant at − 1 SD of fat identification (b = −0.002, t(453) = −0.028, p = .978), but was significant at + 1 SD of fat identification (b = 0.209, t(453) = 2.87, p = .004). The relationship between past collective action participation and group-level gains was strongest among strong fat identifiers, but not among weak fat identifiers. With respect to societal-level gains, the results indicated that the interaction between past collective action and fat identification was not significant, b = −0.040, SE = 0.041, p = .326, suggesting fat identification does not moderate the effect of past collective action participation on societal-level gains.
We also investigated the moderating effect of fat identification on the relationships between individual-, group-and societal-level gains on the one hand and willingness to engage in collective action for fat justice on the other. The results indicated that the interactions between (a) individual-level gains and fat identification (b = −0.012, SE = 0.028, p = .677) and (b) between group-level gains and fat identification (b = 0.049, SE = 0.034, p = .147) were not significant suggesting fat identification does not moderate the effect of individual-and group-level gains on willingness to engage in collective action for fat justice. However, the results showed that the interaction between societal-level gains and fat identification was significant (b = −0.074, SE =0.034, p = .030), suggesting fat identification moderates the effect of societal-level gains on willingness to engage in collective action for fat justice. Perception of societal-level gains was significant at − 1 SD of fat identification (b = 0.207, t (453) = 3.43, p < .001), but was not significant at + 1 SD of fat identification (b = 0.021, t(453) = 0.350, p = .726). The relationship between societal-level gains and willingness to engage in collective action was strongest among weak fat identifiers, but not among strong fat identifiers.
Study 2 offered support for our hypothesis that greater participation in collective action in the past to protest prejudice against fat people would predict greater willingness to participate in collective action for fat justice in the future among fat individuals. The findings could be explained in part through the pathway of fat individuals' gains at the individual and societal level, but not at the group level, after their protest participation for fat justice.

General discussion
Our research had two aims. First, we sought to understand the way engaging in fat activism shapes multi-level outcomes from the perspective of fat activists. Second, we aimed to examine how participation in fat activism may predict future collective action participation for fat justice among fat individuals through endorsing fat individuals' gains at the (1) individual-, (2) group-, and (3) societallevel after their protest participation. We addressed these aims through two studies. The findings of Study 1 indicated important outcomes at the individual-, intra-, and societal-levels. In line with previous research (see Dickins et al., 2011;Dickins et al., 2016;Vestergren et al., 2017), we found that individuals involved in fat activism felt healthier, had greater well-being, and greater selfconfidence. Participants reported that fat people as a group gained a community from fat activism, in that they were able to connect with other people politicized around fatness, and learn more about other social movements, and that fat activism benefitted people at the societal-level as it pushes back against toxic diet culture and stigma around different kinds of bodies.
These findings helped us design our second study, where we wanted to understand the way that multi-level gains from collective action participation could predict continued participation. As previously noted, for the success of the movement, participation has to continue over time and in many different ways. As such, we considered the importance of multi-level outcomes as factors that may aid in the continuation of collective action. In particular, we aimed to examine if different types of outcomes would have different relationships with willingness to engage in future collective action for fat justice.
Results in Study 2 were partially as expected. The findings of Study 2 suggest that for fat activists, increased previous collective action participation was associated with both more individual-level gains and societal-level gains, but not with group-level gains. In line with previous research (see Vestergren et al., 2017), we found that collective action participation in past protests corresponded with greater willingness to engage in future collective action. In addition, individual-level gains were associated with higher group-level gains and societal-level gains as well as greater willingness to engage in future collective action, suggesting added support for the idea that individual psychological changes could be highly relevant in sustained collective action participation (see Thomas et al., 2009).
Notably, the relationship between past collective action participation and group-level gains was strongest among strong fat identifiers (as opposed to weak fat identifiers). This finding is in line with expectations, as fat identity is a group-level marker and would most likely be salient at the group level. Relatedly, the results also showed that fat identification did not moderate the effect of individual-and group-level gains on willingness to engage in collective action for fat justice, but was significant at the societal level for low identifiers.
There are a few potential reasons why this may be; it is possible that low identifiers need to see societal level gains in order to be motivated to take action for fat justice, or it could be that high identifiers are not significant at this level as they are already on board for fat justice based on the potential outcomes for them as individuals or for fat people as a group.
Results of Study 2 also indicated that fat activists' gains at the individual-and societal-level after participation predicted willingness to engage in collective action in the future. We did not find support for our hypothesis related to group-level gains. One explanation may be related to slightly different sample characteristics in both studies. Study 1 focused on known fat activists who have a strong online presence; therefore, the findings of this study may have highlighted the sense of community more strongly as it may be particularly pronounced among this circle. On the other hand, we focused on self-identified fat individuals in Study 2, and this may be one of the reasons why we do not see some of the group-level gains among this sample. Nevertheless, these results indicate the importance of the individual-level gains that participants experience, as well as the societal-level goals and aims of fat activism, are central to sustained involvement in collective action for fat acceptance.

Limitations and future directions
Our studies have a few limitations. As mentioned above, there is no singular goal or perspective within fat activism, meaning that while we have especially learned about the experiences and multilevel outcomes of some activists, we may be missing part of the picture. There continue to be discussions within fat activism about approaches, including discussions around body neutrality, acceptance, and liberation, as well as the relative importance of health in discussions of the worth and value of fat people (see LeBesco, 2004). Future research could examine sustained involvement based on different perspectives on how fat acceptance can be achieved.
In addition, our items for group-level gains could have been improved. It could be that these items could have been provided in more detail or with more context. For example, items were framed as being "in progress," or that while progress has been made on clothing options and medical stigma, these are ongoing issues that need to be addressed in the future. Future research should focus more on the aims of fat activism for fat people to understand why group-level gains are not predicted by past collective action participation for fat justice.
The results here show that fat acceptance has benefits not just at the individual-level, but also for fat people as a group and society as a whole, such as having increased clothing options and a push for weight-neutral healthcare (see Hunger et al., 2020). Engaging in fat activism encouraged participants to consider their views on other Y.G. Acar and Ö.M. Uluğ Body Image 43 (2022) 34-40 social groups, and in many ways connected them to these other social movements. As the current studies were cross-sectional, future research should engage in more longitudinal research with social movements, following the ups and downs that activists experience in continued and sustained involvement (e.g., Chayinska et al., 2021;Tausch and Becker, 2013). In particular, we believe there is much to say about how past participation and the multi-level outcomes people experience can influence solidarity and intersectionality as it is perceived within one's own social movement, as well as how much people begin to engage in others.

Conclusion
Within social psychological research, participation in collective action is oftentimes sterilized from context and ongoing social movements and diluted to reflect a single protest event. However, collective action is usually connected to larger movements around social justice, rather than single events. As such, we believe it is important for research to reflect this and understand that motivators in participation in collective action are oftentimes connected to the outcomes of previous participation. Social justice movements such as the fat acceptance movement provide ample evidence that ongoing, sustained engagement can produce changes not just in individuals, but in societies. It is our hope that with continued research, we can contribute to the discussion around just how to maintain sustained involvement in fat activism and thus create a just society for all body types.

Funding
European Association of Social Psychology Seed Corn Grant.