Abstract
Negative body talk is a normative behavior among United States women and is an important area of intervention for women's health and well-being. Identifying as a feminist and/or endorsing feminist beliefs, which can promote a resistant stance toward cultural appearance standards, may protect against negative body talk. We conducted a mixed-methods study with an online sample of 447 predominantly White (81.1%) United States women between the ages of 18–73 years (M = 41.25, SD = 12.54). We used Consensual Qualitative Research-Modified to systematically code open-ended survey data comprising responses to negative body talk and assessed whether liberal feminist attitudes or feminist identity relate to response categories. We categorized women's responses to negative body talk along with a feminist spectrum which ranged from explicit reinforcement of the thin ideal to a resistant stance toward beauty standards. Liberal feminism was negatively correlated with explicit reinforcement of the thin ideal, and both liberal feminism and feminist identity were positively correlated with a resistant stance. Therapists and practitioners who work with women on issues related to body image disturbance might consider supporting their clients in developing a feminist identity that can critically filter cultural messages that further objectify women.
“Fat talk” is a normative social behavior in Western cultures that consist of self-disparaging conversations about one's weight or shape commonly exchanged among peers (Nichter & Vuckovic, 1994). This dialogue is imbued with cultural meaning about the value and internalization of the White thin ideal, where girls and women feel pressure to achieve a smaller and/or proportionate figure in a society that sexually objectifies women (Arroyo et al., 2014) and pathologizes fatness (e.g., Strings, 2019). Although the term fat talk has primarily been used in previous research, negative body talk represents conversations about weight and shape in addition to other body concerns such as physical aging, eating and exercise behaviors, verbal appearance comparisons, and muscle talk (Becker et al., 2013a; Engeln-Maddox et al., 2012; Sladek et al., 2014). Negative body talk is often a behavioral extension of body shame (Clarke et al., 2010; Gapinksi et al., 2003) in which women seek validation and solidarity with others to “feel better about feeling bad about their bodies” (Shannon & Mills, 2015, p. 164). Meta-analytic evidence suggests that negative body talk can be a risk factor for body image disturbance, making it an important area of intervention for women's health and well-being (Mills & Fuller-Tyszkiewicz, 2017).
Research indicates that women who are younger, exhibit higher body dissatisfaction and/or dieting behavior, lower self-esteem, and higher negative affect are more likely to engage in negative body talk with their peers (for reviews, see Shannon & Mills, 2015; Sharpe et al., 2013). Although previous work yielded minimal differences in the prevalence of self-reported fat talk among Black, White, and Latina women (Engeln & Salk, 2016), making simplistic racial/ethnic body image comparisons does not acknowledge the unique oppression that women of color, and Black women, in particular, experience at the intersection of gender and race (Watson et al., 2019; see also Stanton et al., 2022). Indeed, sexual objectification of the body has historically been rooted in anti-Black racism and the moralization of fatness (Strings, 2019). Therefore, in addition to assessing what protects against negative body talk, exploring which protective factors promote a critical stance to systemic objectification narratives is crucial.
Content of Negative Body Talk
A few studies have investigated the qualitative content of negative body talk in college-aged and community-based samples of women. For example, a United States (U.S.) sample of predominantly White undergraduate women were asked to write a hypothetical conversation with a friend who began the conversation with the cue, “Ugh, I feel fat” (Salk & Engeln-Maddox, 2011). Denying that their friend was fat was the most common response among participants, followed by offering empathy, proposing to make plans to work out or diet together, or making self-disparaging comments about one's own weight/shape to make a friend feel more confident. Participants also identified potential reasons and cause for why one may “feel fat” (e.g., “I don't fit in my jeans anymore”; “I ate so much last night”; Salk & Engeln-Maddox, 2011, p. 22). Most commonly, respondents wanted their friend to deny their self-criticism or for their friend to offer an appearance-related compliment (Salk & Engeln-Maddox, 2011). These response categories were also prevalent in negative body talk responses in a predominantly White sample of women residing in the U.S., United Kingdom, and Australia (Mage = 36.80; SD = 13.48; Becker et al., 2013a). “Old talk” was also present in the scripted dialogue, underscoring the function of youthful beauty standards in appearance anxiety (Becker et al., 2013a). Although denial, compliments, and comparisons are common reactions to negative body talk, they still reinforce the idea that being thin is desirable and achievable. Even positive appearance appraisal from peers can be a form of self-objectification and is associated with body shame (e.g., Tiggemann & Boundy, 2008). Furthermore, making a disparaging comment about one's own body to make someone else feel better about their body is associated with increased state-level body dissatisfaction (Mills & Fuller-Tyszkiewicz, 2018).
Most of the research to date has focused on the correlates and consequences of negative body talk (Shannon & Mills, 2015), though there has been some attention to protective factors, or constructs that may minimize or even negate the harmful effects of negative body talk. Notably, Webb et al. (2016) found that self-compassion ameliorated the relationship between antifat attitudes, body shame, and fat talk in a predominately White sample of college women. Furthermore, there is initial empirical support for the protective function of feminist approaches to negative body talk. In a sample of predominately White college women, Ambwani et al. (2017) found that participants exposed to feminist-oriented vignettes that challenged fat talk reported better mood and less engagement in fat talk when compared with participants exposed to a “typical” fat talk vignette. An exploratory thematic analysis of free-text responses asking how participants would respond to a friend making a negative comment about her own body revealed that positive body talk and feminist talk (“opposes fat talk, the thin ideal, and body objectification; promotes holistic health or self-worth;” p. 89) were common in both conditions with minimal group differences. Other common responses included fat talk and expressing empathy (Ambwani et al., 2017). It is possible that holding feminist attitudes, as opposed to simply being exposed to a feminist-oriented vignette, might protect against engaging in negative body talk. Furthermore, a more nuanced qualitative breakdown of body talk responses situated along with a feminist spectrum and across multiple scenarios is needed to further assess how body talk can reflect feminist values.
A Feminist Approach to Body Talk
Objectification theory posits that because women and girls experience sexual objectification, both interpersonally and through media messages, they are more likely to engage in habitual body monitoring, which can result in body shame and adverse health consequences, including disordered eating, disrupted sexual functioning, and negative affect (Fredrickson & Roberts, 1997). If women are constantly managing and monitoring their appearance, their social interactions (e.g., body talk) reflect and further emphasize this self-objectification instead of focusing on social change, political activism, or feminist empowerment (Calogero, 2017). Clarke et al. (2010) found that the use of fat talk was positively correlated with passive acceptance of traditional gender roles (e.g., related to work and family) and negatively correlated with feelings of personal empowerment. It is possible that holding more traditional attitudes toward women's societal roles puts women at increased risk of engaging in body-scrutinizing conversations, particularly if those women are not engaging in critical thinking regarding gender oppression (Clarke et al., 2010).
Although holding traditional gender role beliefs can put women at risk of self-objectification, holding feminist attitudes—rooted in recognizing and resisting sexist oppression—may protect against experiences of disembodiment, such as body dissatisfaction and appearance-focused behaviors like negative body talk (Henley et al., 1998; Hines, 2020; Hoskin et al., 2017; Smolak & Murnen, 2007). Furthermore, feminist identity, openly declaring oneself a feminist, may serve a protective function given evidence suggesting that feminist women are likely more protected against body shame due to an increased critical awareness of cultural objectification (Murnen & Smolak, 2009). Indeed, Piran (2017) proposed that developing a critical stance toward gendered beliefs is a necessary component of positive embodiment.
As Siegel and Calogero (2021) have recently noted, the measurement of feminist beliefs, identity, and attitudes has been inconsistent and warrants further research exploration. Self-identifying as a feminist is associated with more self-efficacy and body satisfaction, as well as more interest in participating in feminist activism (Borowsky et al., 2016; Eisele & Stake, 2008; Murnen & Smolak, 2009; Yoder et al., 2011). However, endorsing feminist attitudes and beliefs, even in the absence of feminist self-identification, does potentially give women a critical lens through which to view culturally based appearance pressures (Myers & Crowther, 2007; Myers et al., 2012; Rubin et al., 2004). There are numerous nuanced types of feminism that can be endorsed (Henley et al., 1998); however, the most common type of feminism endorsed in the U.S. is liberal feminist attitudes (Chrisler & McHugh, 2011; Myers, 2015). Liberal feminist attitudes include a focus on equal rights, an emphasis on rational arguments, ideas that issues of gender and other inequities can be best solved via our democratic processes by being codified into law, and beliefs that such laws should be limited to the public sphere and not interfere in individuals’ private lives (Henley et al., 1998; Morgan, 1996).
Despite the potential protective role of feminist beliefs, critical feminist women are not immune to these pressures and related appearance concerns (Myers & Crowther, 2007; Rubin et al., 2004). It is possible that when women – including feminist women – are aware of pressures to conform to this ideal, they may experience more appearance concerns. This awareness may mirror what happens when women view photoshopped images with added disclaimer labels. In these instances, instead of resulting in the images effecting women less, the labels seem to call attention to the manipulated parts of the images (Bury et al., 2014; Tiggemann et al., 2013). Experiencing negative body talk may likewise increase appearance concerns for all women, regardless of levels of feminist beliefs. However, despite feminist discourse on and analysis of body talk (e.g., Martz, 2019), additional research is needed to document the relationship between holding feminist attitudes and engaging—or not engaging—in negative body talk.
Present Study
Previous research has qualitatively examined the content of negative body talk among women (e.g., Ambwani et al., 2017; Becker et al., 2013a; Salk & Engeln-Maddox, 2011), providing crucial insight into the composition and function of body talk, but additional analysis is needed to understand the range of feminist or nonfeminist responses to negative body talk. We address these gaps by using an exploratory mixed methods design that incorporates both qualitative and quantitative methods. Qualitatively, we aimed to: (1) inductively categorize body talk responses across three hypothetical dialogue scenarios and (2) construct a feminist continuum of body talk responses. Using quantitative analyses, we aimed to: (3) assess whether liberal feminist attitudes or feminist identity relate to reduced engagement in negative body talk and increased engagement in feminist-related body talk.
Method
Participants
After excluding participants who gave nonsense (i.e., indiscernible in meaning) responses to the open-ended questions (n = 60, five of whom also failed an attention check item), and an additional three who uniquely failed the attention check item, there were 447 women participants with a range of ages between 18 and 73 years (M = 41.25, SD = 12.54) residing in the U.S. With respect to race/ethnicity, 8.8% (n = 39) identified as African-American or Black, 5.4% (n = 24) Asian-American, 2.5% (n = 11) Latina, 1.8% (n = 8) Native American, 81.1% (n = 360) White, and 0.5% (n = 2) mixed race or other. In terms of sexual orientation, 0.4% (n = 2) identified as asexual, 10.1% (n = 45) bisexual/pansexual/queer, 82.1% (n = 367) heterosexual, and 6.7% (n = 30) lesbian. This was a highly educated sample with 22.0% (n = 98) having earned a high school education or less, 5.4% (n = 24) a two-year college degree, 17.9% (n = 80) a four-year college degree, and 54.6% (n = 244) a graduate or professional degree.
Procedure
This study was approved by the Institutional Review Board at Kenyon College We administered the online Qualtrics survey through Amazon Mechanical Turk (MTurk). MTurk (www.mturk.com) pays participants a small fee for completing surveys or other tasks, which allows for efficient and cost-effective data collection. “Turkers” sign up for MTurk either to earn additional income or because they are interested in completing the tasks. Participants tend to represent a wider age range than undergraduate samples (Buhrmester et al., 2011). MTurk data has similar reliability to that of in-person data or other online sites both in general (Buhrmester et al., 2011) and for body image-specific data (Gardner et al., 2012). Participants recruited via MTurk also follow directions better than those recruited in more traditional ways (Hauser & Schwarz, 2016). For the current study, we paid participants $2.25 each for their participation. Participants saw an advertisement posted on MTurk and followed it to a Qualtrics survey. No identifying information was collected via Qualtrics, so participants’ responses could not be linked back to their identities.
We used a convergent mixed methods approach (Fetters et al., 2013), whereby we simultaneously collected quantitative and qualitative data via an online survey and then integrated the data at the analysis stage. Given the exploratory nature of our research design, we emphasize the qualitative strand to allow for the construction of themes and a conceptual spectrum of body talk responses, which were our first two study objectives. Secondarily, we used quantitative analyses to test the third objective of assessing the relations between body talk response categories and feminist measures (Creswell & Plano Clark, 2011).
Materials
Participants filled out a variety of scales related to body image which we did not examine in the present study. Participants responded to negative body talk items we examined immediately after providing demographic information, then to the various body image scales, followed by feminist-related measures.
Body Talk Scenarios
We based the negative body talk scenarios on a measure by Clarke et al. (2010). Participants were told that, “Below are a few examples of ‘weight talk’ that some people engage in. Imagine that you are in each scenario with a female friend who is of average weight and height for her age.” After each scenario participants indicated the frequency of prior experience in response to two questions (i.e., “Have you heard a comment like this before from female friends?”) using a scale of 0 (never) to 4 (very often). Next, we instructed participants to indicate a “typical way” they might respond with a text box in which they typed their response. These responses represent the qualitative data in the present study. The scenarios are as follows:
Scenario 1: Your friend looks in the mirror and says, “I really need to start working out again. Honestly, I am so flabby.” Scenario 2: Your friend expresses remorse about eating a high-calorie dessert item, describing herself as a “huge fatty.” Scenario 3: Your friend wears a new article of clothing, expressing insecurity that it “makes her look fat.”
Liberal Feminist Attitudes
The Liberal Feminist Attitude and Ideology Scale – Short-Form (LFAIS; Morgan, 1996) was used to measure feminist attitudes. This version of the scale contains 11 items measuring agreement with the goals of liberal feminism and the extent to which the participant has progressive gender role attitudes. This measure uses a six-point scale, ranging from 1 (strongly disagree) to 6 (strongly agree). An example item is, “A woman should have the same job opportunities as a man.” A mean score across items was computed. This scale is used consistently as a measure of liberal feminism by researchers in multiple fields, including social work (Charter & Mogro-Wilson, 2018) and psychology (Liss & Erchull, 2013). The alpha coefficient of internal consistency in this sample was 0.89. There were very few missing values on this scale (0.22% of responses), and scores did not depart much from normality with skewness and kurtosis values less than 1. The range of scores was between 1.83 and 6.00. Scores on this scale were not significantly correlated with age (r = −.06, p = .184); or with level of education (r = −.01, p = .886) and they did not differ by racial groups (African American, Asian American, Latina, and White women compared; F(3,440) = 0.95, p = .418 or by sexual orientation groups (sexual minority women vs. nonminority women; t(442) = 0.36, p = .719). In terms of construct validity, in the original sample scores on this scale were higher among avowed feminists than in a sample of undergraduate students, and they were higher in women than in men.
Feminist Identity Measure
We used a one-item continuous measure of feminist identity developed by Morgan (1996). Participants chose a statement that described them. Higher scores indicated more feminist responses. The values were 1 = “I do not consider myself a feminist at all and I believe that feminists are harmful to family life and undermine relations between men and women;” 2 = “I do not consider myself a feminist at all. I am quite traditional;” 3 = “I agree with some of the objectives of the feminist movement, but tend to be somewhat traditional;” 4 = “I agree with most of the objectives of the feminist movement, but do not consider myself a feminist;” 5 = “I agree with all of the objectives of the feminist movement, but do not consider myself a feminist;” 6 = “feminist;” 7 = “a committed feminist;” and 8 = “a committed feminist currently active in the Women's Movement” (Morgan, 1996). This item was chosen because it has more nuance than the typical yes/no question used in previous studies for feminist identity (e.g., Myers, 2022). This item requires participants to understand the term “feminist” and be willing to self-identify with it, as opposed to the LFAIS, which only asks participants to agree with statements that many liberal individuals would endorse regardless of willingness to identify as feminist. We found that the whole range of scores occurred in the data and skewness and kurtosis indexes were less than 1. Only one person did not answer this question (missing data = 0.22% of responses). Scores on this measure correlated slightly negatively with age (r = −.12, p = .013) and positively correlated with education (r = .20, p <.001). In addition, they differed by racial groups, with Latina women having a higher score (M = 6.55, SD = 1.04) than African American (M = 4.26, SD = 1.52), Asian American (M = 4.83, SD = 1.34), or White women (M = 4.82, SD = 1.95), who did not differ from each other; F(3,429) = 4.29, p = .005, followed by LSD post hoc test). Sexual minority women had a higher score (M = 5.66, SD = 1.88) than nonminority women (M = 4.66, SD = 1.86), t(441) = 4.27, p <.001).
Qualitative Analysis
We used Consensual Qualitative Research-Modified (CQR-M) to conduct the qualitative portion of the study. CQR-M is particularly suited to analyzing large volumes of brief qualitative text, such as open-ended survey responses (Spangler et al., 2012). Consensual qualitative research is also fitting for feminist-oriented researchers as it values collaboration and consensual decision-making during the analytic process (and therefore does not employ inter-rater reliability procedures; Hill et al., 1997). We chose an inductive qualitative method (as opposed to deductive or relying on previously reported coding schemes) so that we grounded themes in the data and to reflect the nuance of the three separate negative body talk scenarios. CQR-M has overlap with codebook-based thematic analysis, particularly with its emphasis on early codebook development. Braun and Clarke (2020) situate thematic analysis approaches along an epistemological spectrum: on one end is the reflexive analysis typically characterized by a subjective, interpretivist stance, and on the other end is the coder-reliability approach most often applied within a post-positivist framework. We view CQR-M as a situated between these epistemological stances (Hill et al., 2005), emphasizing both the importance of reflexivity (and therefore the subjectivity of the researcher) on one hand, and a more structured coding process with preliminary theme development on the other (Braun & Clarke, 2020; Hill et al., 2005). Further, given this is a mixed methods study, we tap into both constructivist (theme development) and post-positivist (quantitative analysis) epistemologies throughout the analytic process.
The coding team and authors of this paper consisted of one Ph.D. student in social work (first author), two Ph.D.-level professors, one in clinical psychology (second author) and one in social psychology (fourth author), and one undergraduate student in psychology (third author). Each of us identifies as feminist, White, cisgender women. We each grew up in the U.S. and are highly accustomed to the use and function of body talk in our daily lives. Our positionality, both individually and collectively as an authorship, shapes the analysis and interpretation of our findings, and we acknowledge that owning and reflecting on our own lived experiences and biases is essential to trustworthy qualitative research. We engaged in reflexivity through the analysis and revision processes, which consisted of personal reflection and group discussions, to support critical awareness and mutual accountability.
The team analyzed one scenario at a time, starting with the scenario focused on exercise. The first author reviewed a subset of responses in Scenario 1 and developed an initial list of codes with operational definitions and textual examples which comprised a preliminary codebook. The team met together to review the preliminary codebook and discussed modifications. All four team members then independently coded the data in Scenario 1. We came back together as a group in a series of videoconference meetings to discuss our coding decisions line by line and to come to an agreement on code application. This process was repeated for Scenarios 2 and 3, with team members taking turns developing the initial list of codes for each scenario. We applied relevant codes from Scenario 1 to the second and third scenarios but also developed new codes specific to Scenarios 2 and 3. After we applied the codes to all textual responses, we conducted a second cycle of coding in which we met to (a) discuss how to group codes together into broader domains, (b) categorize the data along with a feminist-inspired conceptual spectrum, and (c) facilitate comparisons across scenarios (see Table 1).
Table of Domains, Subthemes, and Exemplary Responses.
Note. N = 447. Counts for each scenario represent the number of times that code applied to a free-text response, and multiple codes could be applied to each response.
Data were also labeled as ambiguous or bad data but were excluded from analysis. If the cell is blank, the code was not relevant to that scenario.
To ensure the trustworthiness of our analysis, we created an audit trail (Birks et al., 2008), which consisted of step-by-step documentation of our coding process, decisions and rationale related to the development of codes, and notes documenting our revisions of codes and domains. Because we applied codes to the data independently before coming together as a group to discuss the final code application, there were some differences in opinion on coding decisions. Identifying and resolving these discrepancies allowed us to clarify our code definitions and to discuss different potential meanings of the text. We resolved these differences as a group until we achieved consensus, with the third author often suggesting a final decision. As a final data quality check, the first author reviewed the team's coding notes to identify where we needed to recode or double-check coding decisions based on revised definitions of codes. We then went back into the data to ensure that we accurately applied the new or revised codes across all three scenarios.
Results
Qualitative results consisted of six domains which we conceptualized along with a spectrum, with the first domain consisting of explicit reinforcement of body standards, and the last domain consisting of a rejection of such standards. Conversations about the body seldom reflect a singular belief or point of view but are often complex and vary from context to context. For this reason, we double-coded many responses and view this spectrum as a fluid one. The purpose of this spectrum is not to assign value to these responses, but rather to categorize responses in a way that allows readers to identify nuances in how language can reinforce or reject beauty ideals. Domains applied across all three scenarios, but we allowed subthemes to vary by scenario (see Table 1 for domains, subthemes, and example quotes, and Figure 1). Placating and implicit reinforcement of the thin ideal were the most prevalent domains across all three scenarios and rejecting societal standards was the least common.

Feminist Spectrum of Responses to Negative Body Talk.
Consensual Qualitative Findings
Explicit Reinforcement of Thin Ideal Standards
The first domain consisted of comments that directly referenced or emphasized achieving standards related to being thin, attractive, or fit. This domain contained two subthemes, including skinny talk and calorie talk. Skinny talk consisted of explicit advice on ways to become thinner, toned, to lose weight, or a general comment about the value of being thin. We coded the second subtheme, calorie talk, when participants made a comment about how their friend can “undo” the calories, or criticized their friend's decision to indulge in a treat (e.g., “high calorie food will increase your belly and you will feel for that later”). Skinny talk was evident across all three scenarios, but skinny talk, and more specifically calorie talk, were most common in responses to the scenario about indulging in dessert (Scenario 2).
Implicit Reinforcement of the Thin Ideal
Whereas many participants provided comments that explicitly complied with appearance-related standards associated with being thin, other comments were less explicit but still bought into the skinny narrative. This domain consisted of comments which indirectly referenced complying with appearance standards and included the subthemes let's exercise together, commiseration, self-deprecation, and do it in moderation. We coded let's exercise together in Scenario 1 when a participant indicated that she would work out with her friend as a means of accountability or support. Participants offered commiserating responses, indicating that they knew where their friend was coming from and/or felt similarly about their own body. Some participants also provided self-deprecating responses, commenting negatively about their own body or desire to manage their own appearance-related behavior. Commiseration and self-deprecation were evident across all three scenarios but were less common in the clothing scenario (Scenario 3). We coded do it in moderation in the food scenario when the participant indicated it is okay to eat dessert if she does it only occasionally (e.g., “one dessert doesn't matter if you are otherwise eating right; tomorrow is another day”).
Ignore
The third domain consisted of comments that attempted to either move the conversation in a different direction or gloss over the seriousness of their friend's concern. Some participants chose to redirect or ignore by changing the subject, and others brushed it off entirely by dismissing their friend's concern about appearance, clothing, or behavior. These subthemes are applied across all three scenarios and function as a form of denial, though this is more explicit in the next domain (placate). A few participants in the food scenario (Scenario 2) laughed or made a joke in response to their friend's negative body talk, likely attempting to make light of the conversation or to deflect discomfort about the topic at hand (e.g., “I try to joke with them about how much we ate”).
Placate
Placate, defined broadly as comments intended to make a friend feel better about herself and/or her appearance, was the most prevalent domain across scenarios. Positive appearance affirmation was the most common and consisted of constructive compliments about a friend's physical appearance. This subtheme was less common in the dessert scenario. In addition to commenting on the positive aspects of appearance, many participants also explicitly denied or disagreed with their friend's negative appearance comment, which we coded as negative appearance denial. We often double-coded these with positive appearance affirmation across all three scenarios. In Scenario 3, participants offered fashion compliments, indicating that their friend looked great in that piece of clothing. Some participants provided fashion advice by suggesting what to wear instead to make her feel better (e.g., “maybe a different color would be more flattering”). Although these comments tended to be more reassuring than the previous domains, they were still largely appearance focused.
You are Better Than That
The next to last domain consisted of responses that began to challenge the assumptions the participant's friend may have held about her body, albeit more indirectly than the final domain (see reject societal ideals below). You’re better than that consisted generally of comments that not only assuaged their friend's concerns but also provided their friend the “permission” to be who they are, to care for themselves, or to indulge in what feels good. The subthemes included you do you—comments on how her friend should do whatever feels good or right for her; compassion—supportive responses encouraging a friend to not be so hard on herself; and treat yourself—indication that it is okay to indulge and splurge. One participant said, “Hey, I don't play like that. Eat it or don't, but don't talk about yourself like that!” These responses were permissive in nature. They gave their friend permission to indulge; permission to not be so harsh on oneself; permission to be who they want to be.
Society Should Be Better Than That
The final domain consisted of responses that attempted to redefine how the participant's friend might relate to her body beyond appearance satisfaction. This domain, while the least prevalent, represented a more critical stance toward the objectification narrative and included comments that encouraged their friend to exercise for health, reject societal ideals, and focus on clothing functionality. For example, some participants encouraged their friends to exercise for health benefits as opposed to exercising to control, manipulate, or punish the body. A few participants rejected societal ideals when they explicitly addressed the objectifying nature of beauty ideals and applied across all three scenarios, though this was not common (e.g., “Come on. Have we learned nothing from Lizzo?”; “I might try a brief discussion about how food carries no moral value; your body isn't representative of who you are.”). Finally, only a few participants in Scenario 3 made a comment that focused on the functions or capabilities of a piece of clothing, as opposed to solely addressing how it looked.
Quantitative Results
After conducting a qualitative analysis of negative body responses, we added up the responses from each domain across the three scenarios for each participant. Note that there could be more than one response per domain across the three scenarios. For example, if a participant wrote two placating comments in response to Scenario 1 (i.e., comment consisted of negative appearance denial and positive appearance denial), one placating comment in response to Scenario 2, and no placating comments for Scenario 3, their total score for placating responses across scenarios would be three. We ran a series of correlations to determine whether and to what extent liberal feminism and feminist identity correlated with the domains.
The correlations between liberal feminism, feminist identity, and the response categories for the negative body talk scenarios are presented in Table 2. Liberal feminism had a moderately negative correlation with explicit reinforcement of thin ideal (r = −.34; p <.001) and small positive correlations with placating (r = .24, p <.001), you are better than that (r = .21, p <.001), and society should do better (r = .19, p <.001). Thus, those who endorsed liberal feminist attitudes tended to want to help their friend to feel better about her appearance, to offer permission to be more accepting of herself, and/or outright challenge her friend's disparaging comments by placing them in a cultural context. Further, there were positive albeit small correlations between feminist identity and you are better than that (r = .12, p = .012) and society should be better than that (r = .12, p = .012).
Correlations Among Liberal Feminist Attitudes, Feminist Identity, and Body Talk Responses.
Note. *p < .05; **p < .01. We also conducted nonparametric Spearman's rho correlations for the feminist identity measure, which yielded similar results: explicit reinforcement (r = −.11; p = .027), implicit reinforcement (r = −.05; p = .324), ignore (r = −.03; p = .564), and placate (r = −.01; p = .794), you are better than that (r = .13; p = .009), and society is better than that (r = .10; p = .036).
Discussion
We build on previous qualitative work on body talk (e.g., Ambwani et al., 2017; Becker et al., 2013a; Salk & Engeln-Maddox, 2011) by developing a feminist spectrum of responses to negative body talk and assessing whether feminist attitudes or feminist identity relate to negative body talk responses in an online sample of women. We found that feminist attitudes/identity meaningfully correlated with multiple body talk response categories, though it should be noted that these correlations were small to moderate. Nonetheless, the pattern of significant findings is promising and offers initial evidence for the potential protective nature of feminist identity/attitudes in peer body talk. Participants who endorsed liberal feminist attitudes were more likely to critique the idea of negative body talk by either indicating that the person doing the negative body talk is better than that or by offering a critique of societal ideas about weight and shape. Those who self-identified as more feminist showed a similar pattern of critique of individual and cultural expectations surrounding the body. In addition, those who endorsed liberal feminist attitudes were more likely to attempt to placate their friend in the vignette by offering positive appearance affirmations, denying negative comments about the speaker's body, or focusing instead on fashion choices. Perhaps most importantly, those who endorsed liberal feminist beliefs were less likely to explicitly engage in negative body talk or reinforce the body talk from the vignette.
Feminism and Fewer Instances of Explicit Reinforcement of the Thin Ideal
Women who endorsed liberal feminist attitudes were less likely to offer comments that explicitly reinforced the thin ideal, or what is typically considered negative body talk. This finding suggests that feminist attitudes may be protective against these potentially damaging forms of communication and is consistent with Ojerholm and Rothblum (1999), who found that women with higher levels of feminist attitudes endorsed lower levels of antifat attitudes toward other women.
Feminism and Increased Instances of Support and Societal Critique
Those who held liberal feminist beliefs or who self-identified as more feminist were more likely to provide responses that critiqued the beauty ideals, indicating that their friend could rise above the thin ideal and criticisms of society's unrealistic expectations. Therefore, women who identified as feminist or those who agreed with liberal feminist beliefs were most likely to respond to instances of body talk with what might be deemed “feminist” responses. These participants supported and educated their friend by critiquing the societal forces at play that enforce the internalization of the thin ideal and the use of negative body talk, as illustrated by one participant's comment referencing Lizzo, a popular Black musical artist who has built a platform on self-confidence and body acceptance. This comment attests to the power of Lizzo's role in influencing global discourses on body positivity. However, it is important to note that the mainstream body positivity movement has been leveraged to primarily benefit White women and largely lacks a critique of how racism and fatphobia are interlinked to perpetuate the marginalization of Black women (Senyonga & Luna, 2021; Strings, 2019).
These critiquing categories were the only domains related to feminist self-identification in the current sample, which is consistent with previous research findings that self-identification as a feminist is associated with increased participation in feminist activism (Yoder et al., 2011), which may include everyday actions such as critique of societal ideals. However, these findings also illustrate the importance of examining both feminist attitudes and feminist self-identification, as feminist attitudes related to the use of other domains in the current study, including explicitly reinforcing the thin ideal and placating their friend.
The society should be better than that domain, a more feminist response to negative body talk, was the least prevalent category across scenarios, indicating substantial room for growth in practicing compassionate responses that challenge gender and beauty norms. Nonetheless, this domain was more common in the exercise scenario (Scenario 1), where participants offered comments that helped their friend consider exercising for the value of health and not appearance. Our findings stand in contrast to Ambwani et al. (2017) who found that “feminist talk” was relatively common in a sample of undergraduate women at a liberal arts college. A possible explanation for the difference in the frequency of feminist responses is that Ambwani et al. (2017) used a sample of undergraduate women, who may be enrolled in courses that explicitly discuss feminist ideas and critiques of societal messaging, such as psychology and women's and gender studies courses. In contrast, our sample went beyond undergraduate participants to consist of women diverse in age and level of education, albeit a generally highly educated sample. Being outside of the college environment may distance participants from the priming related to class content that Ambwani et al.'s (2017) sample may have experienced in addition to the explicit priming of feminist ideas during their laboratory study, and it is possible that we used a stricter criterion for what is considered a “feminist” response.
Feminism and Increased Instances of Placating Language
Those who endorsed liberal feminist attitudes were also more likely to use placating language in response to negative body talk. This domain included instances of telling friends they looked just fine, denying comments about negative appearance, and comments related to clothing. Placating was the most frequently employed conversation category, where participants provided supportive responses to their friends to help them to feel better about their physical appearance, including denying that their friend was “fat,” which were also common response categories in previous research (Salk & Engeln-Maddox, 2011). Ambwani et al. (2017) also reported that positive body talk (e.g., compliments about the body's appearance) was common but that there was no significant difference between those exposed to a feminist vignette that challenged fat talk and those who were not.
The placating domain is situated in the middle of this feminist spectrum because, while it is not explicitly reenforcing the thin ideal, neither does it challenge it. Although appearance compliments can lead to feeling better temporarily, they do not de-emphasize the importance of appearance and are still associated with body shame (Tiggemann & Boundy, 2008). The increased instances of placating responses among women who endorsed liberal feminist attitudes are consistent with the findings of Rubin et al. (2004), in which women who self-identified as feminist admitted that they struggled to reject the beauty ideal even though they knew that they should, based on their feminist beliefs. The increased use of placating responses among those with higher feminist attitudes is also consistent with the ambiguity that Rubin et al.'s (2004) feminist women expressed surrounding social comparison and judging the bodies of other women. It is possible that women with feminist attitudes are more aware of pressures to conform to this ideal, which may in turn lead to more appearance concerns.
Other Domains
Liberal feminist attitudes had no relation to implicitly reinforcing the thin ideal or to ignoring the body talk by changing the subject or brushing off the concerns of the friend in the scenario. It seems that all women are equally likely to use these responses.
Research has shown that common responses to negative body talk include suggestions for dieting and exercising together to support one another in ameliorating body shame (Becker et al., 2013a, Salk & Engeln-Maddox, 2011). Suggesting exercising together to feel better about one's size or appearance is implicitly buying into the thin ideal, even if it is empathetic or well-intentioned. It is still rooted in the idea that one needs to monitor and alter body shape/size through exercise and that this may be more achievable with an accountability partner. It is important to disentangle health and fitness from achieving beauty standards. Research suggests that although social media content about “#fitspiration” might use language to support women's empowerment to improve their health, it still centers on the White, thin ideal and can contribute to increased social comparison and negative body image (Prichard et al., 2018; Prichard et al, 2020; Tiggemann & Zaccardo, 2015). Alternatively, the exemplary response categorized under the feminist domain, “work out to improve your health, not to fit an aesthetic mold” rejects this ideal. These findings are supported by Swami et al. (2019) who found that feminist beliefs related positively to the appreciation of one's body's functionality. Appreciating one's body functionality is not limited to being able-bodied, but includes other aspects of being in and caring for a body, including bodily sensations and physical expression (Alleva et al., 2015; Alleva & Tylka, 2021). Discussing functionality can be an opportunity for women to appreciate what their bodies can do which has positive effects on psychological well-being (Alleva & Tylka, 2021).
Given the ubiquity of the thin ideal in Western culture (Kilbourne, 1999) and the frequency with which women engage in negative body talk (Jones et al., 2014; Salk & Engeln-Maddox, 2011), it is unsurprising that all women may be equally likely to either try to ignore friend's concerns or reinforce the thin ideal without realizing it. Similarly, when women engage in unfavorable appearance-based social comparisons, which are often inherent in body talk, they experience the same amount of body dissatisfaction regardless of levels of agreement with feminist beliefs (Myers et al., 2012). In addition, it is possible that even women who endorse feminist attitudes are reluctant to engage in difficult conversations and thus eager to change the subject. Some of the responses also seemed to hint at participants being “over” negative body talk. For example, one participant said, she would “ignore it so people stop talking about this stupid stuff.” We coded this domain in the middle of the feminist spectrum, but we recognize that although ignoring negative body talk can be a way to deflect discomfort, it can also be done strategically to change the conversation to something other than appearance.
Prevalence and Comparisons Across Body Talk Scenarios
Our study also provides further insight into the spectrum of possible feminist responses to negative body talk across multiple scenarios. Responses to negative body talk that reinforced beauty ideals were more prevalent in the scenarios about specific behaviors (e.g., eating and exercising) as opposed to clothing. For example, skinny talk (categorized under the domain, explicit reinforcement of the thin ideal) was most prevalent in the dessert scenario (Scenario 2), which contained many comments related to behavioral modification. We developed an additional subtheme, calorie talk, because this diet-related scenario also yielded harsher comments related to modifying the diet to comply with beauty standards. Additionally, participants wanted to change the subject more often in the dessert scenario, with several participants indicating discomfort with the topic at hand. In sum, conversations around food may be particularly triggering or contentious when it comes to navigating conversations about bodies, as it seemed to yield more extreme body talk. This is in line with previous research which suggests that negative body talk is associated with risky eating behaviors (Guertin et al., 2017). Nonetheless, compassionate or permissive (e.g., “treat yourself”) comments were also relatively common in the dessert scenario, which might suggest that because women's conversations are so often characterized as judgmental and disparaging, compassionate responses are also common. Lastly, society should be better than that, the final and more feminist domain, was most frequently coded in the exercise scenario indicating that feminist body talk may be more frequent, or even socially acceptable, in conversations about movement and functionality.
Limitations and Future Directions
This study is not without limitations. For example, we did not utilize an ecological momentary assessment that would capture real-time responses to negative body talk, which would allow for examination of reactions to negative body talk as it occurs in daily life (e.g., Jones et al., 2014), as opposed to hypothetical scenarios which may be susceptible to social desirability bias. Examining qualitative reactions to body talk in real time would allow for an accurate examination of real-time responses to body talk by friends, family, or acquaintances, which may differ from how they would respond to a hypothetical friend whose body talk is only displayed through text on the screen. Further, this study is based on cross-sectional data, prohibiting causal inferences.
Most importantly, our sample was limited in sociodemographic diversity, and our findings are therefore limited in generalizability. Recent research on sociodemographic representation of Amazon MTurk samples suggests that MTurk workers tend to have higher levels of education, are more likely to be white-collar workers than the general population, and people of color, particularly Black and African Americans are underrepresented (Buhrmester et al., 2011; Castille et al., 2019). An inclusive and intersectional feminist analysis of body-related dialogue is essential to critiquing and improving societal conditions that marginalize individuals based on gender, race, and size. Future research should examine body talk with marginalized and under-investigated samples, including Women of Color, queer, transgender, nonbinary persons, individuals with disabilities, and those living in larger bodies, and should contend with the unique oppression these individuals face.
Lastly, although Hauser and Schwarz (2016) found that Turkers have better adherence to instructions than participants on other online recruiting sites, uncertainties exist about the accuracy of MTurk data (Hauser et al., 2019; Wessling et al., 2017). We used both our attention check variable and nonsense answers to the qualitative questions as exclusion criteria due to these concerns. Future studies should add queries about the content of the survey in light of potential problems with traditional attention check items (Hauser et al., 2019).
Practice Implications
Peer-based conversations can negotiate a more feminist approach to relating to the body—one that resists objectification and promotes positive embodiment (Piran, 2017). Given that dialogue can function as a reflection of cultural values about the body, speaking about the body beyond its appearance aspects could help women experience positive embodiment, particularly when this is done in the context of supportive, feminist relationships (Piran, 2017). One of the key features of positive embodiment, according to Piran and Tylka (2019), is the development and use of voice, which is developed in social context, including interpersonal relationships. Therapists and practitioners who work with women on issues related to body image disturbance might consider supporting their clients in developing a feminist identity that can critically filter cultural messages that further objectify women. In addition, women can be encouraged to talk about their bodies from a subjective point of view—instead of focusing on the body's outer appearance, conversations can focus on body functionality and appreciation, bodily expression and care, and self-compassion (Alleva & Tylka, 2021; Piran, 2017; Webb et al., 2016). Clients could also benefit from direct psychoeducation on negative body talk and its consequences as well as practice how best to respond in scenarios like those used in our study. These techniques directly addressing negative body talk are used in The Body Project, a highly successful eating disorder prevention program focusing on body image (Becker et al., 2013b; Stice et al., 2017).
Conclusion
Our mixed methods study provides a nuanced assessment of body talk responses from a feminist perspective. We establish initial evidence for the role of feminist attitudes and feminist identity in shaping the way women converse with one another about their bodies in an online sample of predominately White, middle-aged women. This dialogue can function as a cultural reflection of how society objectifies bodies, but it can also function as an act of resistance; an opportunity to challenge beauty norms. Transforming negative body talk to feminist talk might involve supporting peers to appreciate their body's nonappearance-related characteristics, challenging the ways society leads us to believe appearance determines value, and encouraging peers to move, eat, and wear clothes that align with her values and not that of society.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors received funding support from the Department of Psychology at Kenyon College.
