Abstract
In this study, we examined how sexual objectification, abortion circumstance, and sexist attitudes interacted to affect abortion stigma and dehumanization of women who have abortions. A total of 237 participants were randomly assigned to a vignette depicting a woman who was either (a) sexually objectified or (b) not sexually objectified and who sought an abortion in either (a) a typical circumstance (i.e., ill timed) or (b) an exceptionalized circumstance (i.e., fetal anomaly). Results indicated that in typical abortion circumstances, women who have abortions are at risk of dehumanization, regardless of whether they are seen as sexual objects. In contrast, exceptionalized conditions may yield humanizing effects for women who have abortions, unless they are seen as sexual objects. Furthermore, participants with moderate and high levels of hostile sexism judged women who have abortions more harshly: they believed that the woman deserved more punishment for having an abortion for typical reasons compared to having an abortion for exceptionalized reasons. Results also suggested that women who have abortions for more typical reasons may be granted less moral patiency (i.e., moral concern), regardless of sexual objectification. However, under exceptionalized circumstances, sexual objectification diminished perceptions of moral patiency among participants with moderate and high levels of hostile sexism. In short, our findings suggest that sexual objectification, abortion circumstance, and hostile sexism operate concomitantly to affect how women who have abortions are evaluated and perceived.
On June 24, 2022, the United States’ Supreme Court upheld the decision from Dobbs v. Jackson Women's Health Organization (2022), a Mississippi law obstructing abortion after 15 gestational weeks except in cases of fetal anomaly and medical emergencies. In doing so, the Supreme Court returned abortion access to the purview of individual states and overturned Roe v. Wade, which previously established the right to have an abortion as part of the constitutional right to privacy (Watson & Germain, 2024). This ruling advanced pro-natalist sentiment that prioritizes women's reproductive and parenting capacity over personal liberty and autonomy, specifically that of white, heterosexual, cisgender women (Fikslin, 2021). In this article, we acknowledge that abortion and abortion stigma affect people across the gender spectrum, including nonbinary, transmasculine, and gender diverse people capable of childbearing. Where possible, we use the term “people who have abortions” to acknowledge the diversity of people affected by pro-natalism and anti-abortion stigma and legislation. Most commonly, we use the term “women who have abortions” to describe prior research involving samples of (presumably) cisgender women, theoretical frameworks applied to women, and the relations among our study variables.
Notably, the Dobbs decision contradicts public opinion about abortion, as most Americans support abortion access at least in some circumstances (Pew Research Center, 2025). Despite public support for abortion, this common pregnancy outcome (approximately one in four women by the age of 45; Guttmacher Institute, 2024) remains highly stigmatized and concealed (O'Shea & Watson, 2025). Recent statistics on people who have abortions revealed that 57% were in their ‘20's; 42% were non-Hispanic Black, 30% non-Hispanic White, and 22% Hispanic; and 43% had previous live births (Kortsmit et al., 2023).
However, the furtive nature of abortion often results in Americans relying upon the “abortion imaginary” to fill gaps in understanding about abortion and the people who have them (Bruce et al., 2024). These imaginings typically reflect stigmatizing and stereotypical representations of women and their sexuality, rooted in sexist, racist, and classist assumptions. For example, Bruce et al. (2024) found that participants in their study imagined “who has abortions” (p. 3) as young, irresponsible, incompetent, cisgender women who would inevitably become mothers, but were not yet emotionally or financially prepared. Additional imaginings referenced casual, sexual rendezvous that resulted in unwanted pregnancies among “promiscuous women,” and/or images of poor, Black women reliant upon social welfare supports. These findings suggest that people perceive women who have abortions negatively and fail to fully humanize them, but the roles of sexism, racism, classism (and other axes of oppression) driving these stigmatizing and dehumanizing perceptions remains underexplored.
Pregnant women considering abortion experience abortion stigma as do those who have had abortions (Biggs et al., 2020); however, there is limited research on factors that shape perceivers’ imposition of abortion stigma on women who have abortions (e.g., Pacilli et al., 2018, 2024). This study integrates theories of stigma (Link & Phelan, 2001), objectification (Fredrickson & Roberts, 1997; see also Dyer et al., 2023), and ambivalent sexism (Glick & Fiske, 1996) to test a framework predicting stigma and dehumanization of women who have abortions. Specifically, we investigated whether sexual objectification shapes how women who have abortions are perceived, and whether these perceptions interact with abortion circumstance and sexist ideologies to influence abortion stigma and dehumanization.
Abortion Stigma and the Transgression of Feminine Ideals
According to Link and Phelan (2001), stigma occurs at the nexus of several interrelated factors, including the detection of human differences that are socially devalued and negatively stereotyped, which ultimately result in a loss of status and discrimination. Further, Kumar et al. (2009) defined abortion stigma as “a negative attribute ascribed to women who seek to terminate a pregnancy that marks them, internally or externally, as inferior to ideals of womanhood” (p. 169). Abortion stigma is thought to arise because women who have abortions have transgressed three essentialist notions of femininity and womanhood: (a) perpetual fecundity, whereby women's sexuality and functioning are viewed as a conduit to and object for procreation; (b) motherhood as an inevitability; and (c) the notion that women are instinctually maternal (Kumar et al., 2009). In an experimental study, Pacilli et al. (2018) found that women who aborted their pregnancies were deemed less suitable for more gender stereotypical occupations (e.g., social work and educator) than those who did not, revealing how women who have abortions are thought to violate feminine ideals.
Abortion stigma also includes stereotyping people who have abortions as immoral, promiscuous, incompetent, selfish, and murderous (Kumar et al., 2009). Such stereotypes are often used as justification for punishment and mistreatment of women who have abortions (Baker et al., 2023; Link & Phelan, 2001). For example, Baker et al. (2023) found that participants believed that women who had abortions transgressed feminine ideals. These perceived transgressions were used to justify potential forms of punishment, including incarceration, forced sterilization, economic sanctions (e.g., fines), and religious consequences (e.g., karma and judgments from God).
However, other variables beyond abortion stigma affect how people who have abortions are treated (Dyer et al., 2023; Kumar, 2013). Indeed, Dyer et al. (2023) advanced a reproductive objectification framework in which both hostile and benevolent sexisms are posited to result in the objectification and dehumanization of people who have abortions. Thus, in expanding the reproductive objectification framework, we also integrated ambivalent sexism and objectification theory to better understand the experiences of women who have abortions.
Abortion Stigma, Dehumanization, and Sexual Objectification
Stigmatization of women who have abortions may also be connected to objectifying and dehumanizing views of pregnant women. Scholars have noted that anti-abortion rhetoric can be characterized as objectifying (Dyer et al., 2023), describing pregnant women as “instruments” (Moore, 2019) or “vessels” for developing fetuses (Dyer et al., 2023; Huang et al., 2014). According to objectification theory, women are frequently reduced to their attractiveness, sex appeal, or sexual and reproductive functions for use by others, leading to dehumanization (Fredrickson & Roberts, 1997; see Bernard et al., 2018a, 2018b; Roberts et al., 2018; Ward et al., 2023, for recent reviews). For example, when perceivers narrow attention to a woman's appearance, they perceive her as less competent and warm—two fundamental dimensions of humanness (Bernard et al., 2018a; Fiske, 2013; Heflick et al., 2011; Heflick & Goldenberg, 2009).
Sexually objectified women—those presented in ways that emphasize their bodies or appearance (e.g., in lingerie or bikinis)—are perceived more as sexual objects and less as capable, likeable individuals (e.g., Loughnan et al., 2010; Wollast et al., 2018). They are also denied moral status, placing them outside the realm of concern for harm inflicted on fellow human beings (Loughnan et al., 2010). For example, Wollast et al. (2018) found that, compared to nonsexually objectified women, sexually objectified women were seen as less capable and agentic, with participants also indicating that they would be more likely to shock them during a hypothetical experimental task.
This link between sexual objectification and dehumanization may also apply to abortion stigma, although this has been understudied to date. Dyer et al. (2023) theorized that pregnant people, and those who consider or have abortions, may face a distinct form of sexual objectification tied to their reproductive capacity. Drawing upon Nussbaum's (1995) and Langton's (2009) conceptualizations of objectification, Dyer et al. (2023) argued that the purported “pro-woman” anti-abortion rhetoric involved in anti-abortion legislation may be considered objectifying because abortion seekers are silenced and denied autonomy and agency, as though they are incapable of relying upon their own subjective experiences to make informed choices. Thus, paternalistic anti-abortion legislation seeks to “protect” pregnant people from the so-called damaging effects of abortion. Additionally, “pro-woman” abortion rhetoric prioritizes pregnant people's reproductive capacity, valuing their bodies as instruments or vessels for fetal viability. In addition, like sexually objectified women, people who have abortions are often reduced to their sexual and reproductive behavior (e.g., perceived as “irresponsible sluts,” Tanenbaum, 2017). In a recent study, Moradi et al. (2025) found that multiple indicators of sexually objectifying views and behaviors toward women were indirectly related to less support for abortion access in both typical and exceptionalized abortion 1 circumstances through benevolent and hostile sexisms. Further, in a qualitative study regarding cisgender women's experiences of the Dobb's ruling, participants described objectifying and dehumanizing consequences, noting that they were denied bodily autonomy and experienced their bodies as a state-surveilled site (Nolen et al., 2024).
Despite these important studies, little empirical research has examined sexual objectification in conjunction with abortion stigma and dehumanization (Dyer et al., 2023). One reason for this gap in the literature is that most sexual objectification research has focused on women who appear to conform to feminine gender roles. Thus, it remains unclear whether sexually objectified women who transgress feminine ideals, such as those who have abortions, are also subject to dehumanizing perceptions (Bernard & Wollast, 2019; Gray et al., 2011; Haslam et al., 2013; Loughnan et al., 2010).
Although no studies to date have examined the role of sexual objectification for women who have abortions, related literature suggests that sexually objectified women who have abortions may be dehumanized and stigmatized. For example, studies show that sexual objectification increases dehumanization and victim blame and reduces perceptions of perpetrator responsibility in cases of stranger and acquaintance rape (Bernard et al., 2015; Bevens & Loughnan, 2019; Loughnan et al., 2013). Although survivors of sexual assault and those who have abortions differ in myriad ways, both face disproportionate scrutiny and negative evaluations of their sexual behavior, including harmful stereotypes that they are promiscuous and irresponsible (Buddie & Miller, 2001; Kumar et al., 2009). Relatedly, prior research found that women who had an abortion were dehumanized (i.e., seen as less warm similar to machines or less competent similar to animals) more than those who did not have an abortion (Pacilli et al., 2018). Together, these findings suggest that abortion stigma and dehumanization may be connected to sexual objectification.
One potential moderating factor in the link from sexual objectification to stigma and dehumanization is the circumstances under which an abortion is sought. Pacilli et al. (2024) found that women who chose to abort their pregnancies for typical (e.g., economic concerns and educational pursuits) versus exceptionalized (e.g., rape, incest, and maternal life endangerment) reasons were attributed less human characteristics, and hostile behaviors toward them were viewed as less problematic. Additionally, moral outrage mediated the relation between abortion for typical reasons and dehumanization; abortion for more typical reasons elicited greater moral outrage, which in turn increased dehumanization (Pacilli et al., 2024). One explanation is that exceptionalized cases facilitate empathy toward women who have abortions (e.g., Hunt, 2019), effectively humanizing them and increasing moral concern, even as they challenge traditional gender roles.
Taken together, research has linked sexual objectification of women with dehumanization, particularly in the context of highly personal and stigmatizing experiences (e.g., rape). Additionally, perceivers may apply stereotypes and imaginings of promiscuity to women who have abortions, particularly for more typical reasons (e.g., Bruce et al., 2024), ultimately resulting in greater stigmatization and dehumanization of women who have abortions. Further, these posited relations may be especially strong among perceivers with greater sexist attitudes toward women.
Ambivalent Sexism, Sexual Objectification, and Abortion Circumstance
Glick and Fiske (1996) described two distinct but interrelated forms of sexism: hostile and benevolent. Hostile sexism refers to overtly negative and stereotypical attitudes toward women (e.g., women are manipulative, overly emotional, revenge seeking, and untrustworthy). Benevolent sexism is often characterized by superficially positive yet paternalistic attitudes toward women—particularly those who conform to traditional feminine norms—that inevitably maintain patriarchy (e.g., women are precious and need to be protected by men). These two forms of sexism are correlated with one another, suggesting they operate concomitantly to affect women's lived experiences (Glick & Fiske, 1996).
The view that women who have abortions are selfish, irresponsible, and murderous illustrates how hostile sexism promulgates abortion stigma (Dyer et al., 2023; Kumar et al., 2009). On the other hand, the sentiment that women are emotionally fragile and must be protected from the purported “traumatic” effects of abortion, as commonly seen in supposed “pro-woman” anti-abortion rhetoric, is characteristic of benevolent sexism (Dyer et al., 2023; Roberti, 2021). Indeed, both benevolent and hostile sexisms were related to lower levels of abortion support (Begun & Walls, 2015; Huang et al., 2014; Osborne & Davies, 2012). However, the relation between ambivalent sexism and abortion support is complicated by the circumstance surrounding abortion. Specifically, past research shows that benevolent sexism was negatively correlated with support for abortions labeled as both “traumatic” (e.g., when the woman's life is in danger) and “elective” (i.e., seeking an abortion for any reason outside of rape, incest, fetal anomaly, or life endangerment). 2 However, hostile sexism was only negatively related to abortion support for “traumatic” abortion (Huang et al., 2014). These findings suggest that, although both forms of sexism affect negative perceptions and deleterious treatment of women seeking abortions, benevolent sexism, despite its ostensibly positive stance towards women, may be particularly pernicious.
Because most people are not rigidly pro-abortion or pro-life, but rather situationists, it is important to examine how the purpose or reason for seeking an abortion may affect how people seeking abortions are perceived (Diamant & Mohamed, 2023; Osborne et al., 2022; Rye & Underhill, 2020). Although public support for abortion access across circumstances is high, it is greater in exceptionalized versus typical circumstances (Jozkowski et al., 2021; Osborne et al., 2022; Smith & Son, 2013). In prior research, typical abortion circumstances are operationalized to include that a pregnant person does not want children, has a low income, and/or is unmarried and does not want to marry the impregnating person. By contrast, a defining feature of exceptionalized abortions is that they are viewed as outside of the person's control (e.g., fetal anomalies, life endangerment, and rape or incest), often involving health risk to the pregnant person and/or fetus.
A contemporary example of exceptionalized abortions that meets all these conditions is denying women abortion care when they are having a miscarriage. Several widely publicized cases described the dire consequences of denying women emergency abortion care for miscarriage (e.g., Candi Miller, Amber Nicole Thurman, and Porsha Ngezumi). Jaci Statton, an Oklahoman woman, began showing signs of a miscarriage resulting from a partial molar pregnancy. Because fetal cardiac activity was detected by sonogram, she was denied an abortion and told to wait in the parking lot of the hospital until near death (Forman, 2023). Similarly, after a year and a half of fertility treatments, Amanda Eid became pregnant. When her water broke prematurely and she began to miscarry, her doctors informed her that they were unable to provide an emergency abortion because fetal heartbeat was still detected (Cohen & Bonifield, 2023). Both women nearly lost their lives and experienced significant trauma. Nevertheless, people who hold the most ardent anti-abortion attitude oppose abortion in all cases, regardless of circumstance (15%; Pew Research Center, 2025). In addition, abortion for miscarriage involves identical medical procedures as abortion for other reasons and it is viewed as abortion in federal guidance (e.g., Emergency Medical Treatment and Labor Act [EMTALA]). Thus, in addition to sexual objectification and ambivalent sexism, we sought to better understand the role of circumstance surrounding the abortion and how it relates to the stigmatization and dehumanization of women.
Current Study
The current vignette study utilized an experimental design to examine the effects of sexual objectification, abortion circumstance, and sexist attitudes on abortion stigma and dehumanization. Within the context of this study, abortion stigma was conceptualized as viewing those who have abortions as transgressing feminine ideals and deserving punishment, consistent with prior theory (e.g., Kumar et al., 2009). Given that dehumanization has frequently been assessed through moral patiency (i.e., deservingness of moral treatment; Loughnan et al., 2010) and perceived warmth, competence, and moral agency (i.e., capacity to act morally; Loughnan et al., 2010), we examined these variables as indicators of dehumanization (e.g., Bernard & Wollast, 2019; Gray et al., 2011; Heflick & Goldenberg, 2009; Loughnan et al., 2010).
Participants were exposed to one of four experimental conditions, in which they viewed an image of a woman either clothed (i.e., not sexually objectified) or dressed in a bra (i.e., sexually objectified) and then read about her abortion experience, which was described as either occurring for an exceptionalized reason (i.e., fetal anomaly, beginning of a miscarriage, and threat to life) or typical reason (i.e., decided the timing was not right; see Supplemental materials). Aside from these variations, the vignettes were identical. The current study examined the following hypotheses.
The sexually objectified woman who received an abortion would elicit greater abortion stigma and dehumanization than the nonobjectified woman who received an abortion. The woman who received an abortion for a typical reason would elicit greater abortion stigma and dehumanization than the woman who received an abortion for an exceptionalized reason (i.e., miscarriage management, fetal anomaly, and maternal life endangerment). The sexually objectified woman who received an abortion for a typical reason would elicit more abortion stigma and dehumanization than the nonobjectified woman who received an abortion for an exceptionalized reason.
Across all hypotheses, we anticipated that the observed effects would be stronger for participants with higher (versus lower) levels of hostile and benevolent sexism.
Method
Participants
A power analysis estimating a small (.03) effect, with power of .80 and alpha of .05 indicated a minimum of 192 participants to detect significant main and interaction effects. A total of 237 participants took part in this study; therefore, the study was adequately powered to detect significant effects. Participants were predominantly cisgender women (n = 117, 49.4%) and cisgender men (n = 113, 47.7%). In terms of race, most participants were white (n = 180, 75.9%), followed by Black or African American (n = 19, 8.0%), Latine (n = 15, 6.3%), Asian American (n = 12, 5.1%), multiracial (n = 7, 3.0%), Native Hawaiian or Pacific Islander (n = 4, 1.3%), and Native American (n = 3, 0.9%). Participants were also predominantly heterosexual (n = 177, 74.7%), followed by bisexual (n = 39, 16.5%), pansexual (n = 7, 3.0%), asexual (n = 7, 3.0%), gay (n = 4, 1.7%), lesbian (n = 4, 1.7%), and queer (n = 4, 1.7%); see Table 1 for additional demographic data.
Participants’ Demographics.
Procedure
This study was approved by a university Institutional Review Board. To take part in the study, participants had to be 18+ years of age and live in the United States. Participants were recruited through Prolific, a research recruitment platform that connects researchers with a verified pool of human participants with fraud detection capabilities (Haycraft Mee, 2025). They were informed that the purpose of the study was to examine abortion-related attitudes and involved completing a survey through Qualtrics.com. Participants began by reading a consent form, and after giving their consent and indicating that they met the inclusion criteria, they completed a demographics measure, the ambivalent sexism inventory, and a measure assessing attitudes toward women (i.e., Attitudes Towards Women scale [Spence et al., 1973], which was incorporated to assess validity of scores on the dependent variable measures); these three inventories were randomized. After completing these measures, participants were randomly and evenly assigned to one of the four different experimental conditions: objectified and typical circumstance, objectified and exceptionalized circumstance, nonobjectified and typical circumstance, nonobjectified and exceptionalized circumstance. Participants were not allowed to advance to the remainder of the study until at least 35 seconds had elapsed, in order to facilitate full reading and comprehension of the vignette. After exposure to the experimental conditions, participants completed manipulation check items and then several brief inventories assessing the dependent variables; the measures assessing the dependent variables were presented in a randomized order. Following study completion, participants were paid $2 through Prolific. Study hypotheses were not pre-registered. Data from this study may be accessed by contacting the first author.
Materials and Measures
Sexual Objectification
In the objectification literature, sexual objectification is often manipulated by presenting images of women fully clothed or scarcely dressed (e.g., in bikinis or lingerie; Bernard et al., 2018b). However, such research has often used nonstandardized images of objectified women, resulting in difficulty interpreting results across studies. To standardize and validate these stimuli, Ruzzante et al. (2022) recently published the Sexual OBjectification and EMotion (SOBEM) database for use by the scientific community. The SOBEM database is a stimulus set of 280 high-resolution images of objectified and nonobjectified women with varying facial expressions and intensity. Images were standardized by resizing for symmetry and equating coloring. Validity for these images was supported through findings that objectifying images of women were viewed as more attractive and object-like (i.e., less competent, trustworthy, and likeable than nonobjectified women); these relations did not vary by neutral facial expression (Ruzzante et al., 2022). In the current study, the objectified image depicted an attractive white woman in a bra and the nonobjectified image depicted the same woman with a long sleeve shirt. We used an image of a woman with a neutral facial expression (see Supplemental Appendix).
Abortion Circumstance
Two separate vignettes were created to examine the effects of an exceptionalized (i.e., abortion occurring as a result of miscarriage management) or typical abortion circumstance. In each circumstance, the target (Laura) was described as a 26-year-old woman in graduate school who modeled part-time to pay for her education. The part-time job as a model was used to rationalize her presentation in a bra, as denoted above in the manipulation of sexual objectification. She was married for three years, during which she and her partner merely discussed having children. In the exceptionalized condition, Laura became pregnant and learned of an anomaly at eight weeks that would make it challenging for the fetus to survive. At 12 weeks, Laura began to miscarry and was told that she would risk getting a life-threatening infection if she did not have an abortion. These vignettes were based on data and real-world examples of typical and exceptionalized abortions (e.g., Jaci Statton and Amanda Eid; Cohen & Bonifield, 2023; Diamant & Mohamed, 2023). In the typical abortion condition, all details were the same except the pregnancy appeared healthy. However, Laura decided that “it was best” that she had an abortion at this time. Both vignettes ended with the statement “Thus, at 12-weeks, Laura had an abortion,” clearly indicating that the women had abortions despite the different circumstances.
Manipulation Checks
Based on objectification theory (Fredrickson & Roberts, 1997), we examined the extent to which participants reduced the target to her appearance and sexual functioning using two items, “Laura is very sexy” and “Laura is attractive.” Additionally, we assessed the degree to which participants stereotyped the target based on abortion circumstance using two items, “Laura is promiscuous,” and “Laura had a traumatic experience”; the former item was used to assess the stereotype that people who have typical abortions are sexually promiscuous, while the latter was used to examine the assumption that exceptionalized abortion experiences are traumatic and more emotionally challenging than typical abortions. Participants responded to each of these items on a 1 (strongly disagree) to 7 (strongly agree) scale.
Ambivalent Sexism
The 12-item short version of the Ambivalent Sexism Inventory (Glick & Fiske, 1996; Rollero et al., 2014) was used to assess hostile sexism (six items, e.g., “Women seek to gain power by gaining control over men”) and benevolent sexism (six items, e.g., “Many women have a quality of purity that few men possess). Items were rated on a 0 (strongly disagree) to 5 (strongly agree) scale and averaged to compute subscale scores, with higher scores indicating greater sexist attitudes. Validity has been supported by significant correlations from hostile sexism (
Abortion Stigma
Few studies have assessed abortion stigma from the stigma holders’ perspective (e.g., Pacilli et al., 2018, 2024). Drawing from the available studies, we developed measures to assess abortion stigma in the context of our study's vignettes. Specifically, we generated 38 items to assess the extent to which participants attributed theorized aspects of abortion stigma to the vignette target, including that Laura was denying her maternal essence, was immoral and selfish, should be ashamed, and should be punished (e.g., Baker et al., 2023; Kumar et al., 2009). These items were specifically derived from scholarship on abortion stigma (e.g., Baker et al., 2023; Cockrill et al., 2013; Hanschmidt et al., 2016) and Kumar et al.'s (2009) conceptualization of abortion stigma as arising from transgressing feminine ideals. We utilized principal axis factoring with direct oblimin rotation to determine the factor structure of responses to these items (see Supplemental materials). Kaiser–Meyer–Olkin (KMO) criteria and Bartlett's test of sphericity were examined to assess adequacy of sample size and factorability, respectively. Data indicated that the sample size was adequate, KMO = .941, and suitable for factor analysis, X2(325) = 7,374.986, p < .001. Factor structure was guided by the following criteria: eigenvalues >1, scree plots, interpretability of factors, and at least three items per potential subscale (Field, 2013). In determining item retention, items had to load at .45 and above and not have cross loadings ≥ |.32|, resulting in the removal of 12 items (Field, 2013). Based on these criteria, the items yielded two factors comprising what we labeled (1) Transgression of Feminine Ideals and (2) Deserves Punishment, which were consistent with prior theory about components of abortion stigma (Baker et al., 2023; Kumar et al., 2009). Together, these two factors explained 68.570% of the variance in the abortion stigma items (see Table 2).
Items, Loadings, Communalities, and Descriptive Statistics of Abortion Stigma.
Note. *Reverse scored item.
Bold values indicates item loadings to their respective factors.
Transgression of Feminine Ideals. Consistent with Kumar's (2009) conceptualization of abortion stigma, 18 of our abortion stigma items assessed the extent to which the target was perceived to transgress essentialized notions of ideal womanhood. Sample items included, “Laura is denying her potential as a mother” and “Laura is selfish.” Items were rated on a 1 (strongly disagree) to 7 (strongly agree) scale and averaged to attain scale scores, with higher scores indicating greater endorsement of transgression of feminine ideals. In addition, as a validity check, we examined the degree to which scores on this measure correlated with the Attitudes Towards Women scale (Spence et al., 1973). The significant positive relation (r = .47, p < .001) supported validity, suggesting that transgression of feminine ideals was associated with more restrictive attitudes about the rights and roles of women. Responses across transgression items were highly consistent,
Deserves Punishment. Consistent with Baker et al.'s (2023) findings on imagined punishment for women who have abortions, eight of our abortion stigma items assessed participants’ views that the target should be punished for having an abortion. Example items included “Laura should face jail time” and “Laura should pay a monetary fine.” Items were rated on a 1 (strongly disagree) to 7 (strongly agree) scale and averaged to attain scale scores, with higher scores indicating more endorsement of punishment. The Deserves Punishment scores were correlated positively with scores on the Attitudes Towards Women scale, r = .483, p < .001, supporting validity. Responses across the Deserves Punishment items yielded excellent internal consistency (
Dehumanization
Prior research indicates that perceived warmth, competence, moral agency, and moral patiency (i.e., moral concern) for targets are key components of perceiving others as human versus object (Fiske, 2013; Haslam, 2006; Haslam et al., 2013; Loughnan et al., 2010, 2013). Accordingly, we operationalized dehumanization with items established in prior studies to assess these domains (e.g., Bernard & Wollast, 2019; Heflick et al., 2011; Holland & Haslam,2013; Loughnan et al., 2010). Specifically, we used nine items from prior research (e.g., Bernard & Wollast, 2019; Heflick et al., 2011) assessing warmth, competence, and moral agency of the target and four items from prior research (Loughnan et al., 2010) assessing moral patiency for the target. Importantly, these specific items have been used to examine the effect of sexual objectification on dehumanization in prior research (e.g., Bernard & Wollast, 2019; Bernard et al., 2018b; Gray et al., 2011; Heflick et al., 2011; Heflick & Goldenberg, 2009; Loughnan et al., 2010).
We conducted a principal axis analysis with direct oblimin rotation to determine the structure of responses to the dehumanization items. Results suggested that the sample size was adequate, KMO = .919, and the data were suitable for factorability, X2(66) = 3,059.698, p < .001. We used the same criteria as described above for determining the number of factors and item retention (Field, 2013). The data yielded two factors, which we titled humanization (attributing competence, warmth, and moral agency to the target) and moral patiency (expressing moral concern for the target). This two-subscale structure explained 77.861% of the variance in dehumanization (see Table 3).
Items, Loadings, Communalities, and Descriptive Statistics of Dehumanization.
Note. Bold values indicates item loadings to their respective factors.
Humanization. Eight items reflected the degree to which participants humanized the target. These included items such as “How capable do you find Laura” and “How trustworthy do you find Laura.” The items were rated on a 1 (not at all) to 4 (extremely) scale. Item ratings were averaged to compute total scores, with higher scores indicating greater levels of humanization. Humanization scores yielded moderate negative relations with the Attitudes Towards Women Scale, r = −.309, p < .001, demonstrating evidence of validity (i.e., humanization scores are associated with, but distinct from negative attitudes toward women). Responses across the Humanization items were highly consistent,
Moral Patiency. Four items assessed moral patiency, or having moral concern for the target. Sample items included “How bad would you feel if you heard Laura had been hurt” and “How bad would you feel if you heard Laura had been taken advantage of?” Items were rated on a 1 (not at all) to 4 (extremely) scale and averaged to achieve total scores, with higher scores indicating greater levels of moral patiency. As initial evidence of validity, this measure yielded a significant, negative associated with the Attitudes Towards Women scale, r = −.329, p < .001, suggesting that moral patiency is associated with, but distinct from, negative attitudes toward women. Responses across the moral patiency items demonstrated excellent internal consistency (
Results
Data Cleaning and Analytic Plan
A total of 265 people accessed the study's link, but 13 did not meet the inclusion criteria and therefore did not advance to the survey. Potentially fraudulent responses were examined by Qualtrics bot detectors, including reCAPTCHA scores, duplicate entries, and fraud scores; five respondents were removed due to suspicious responses. Next, attention check items (e.g., “To make sure you are paying attention, please select 1 for this response”) were examined, as well as responses to a nonsensical item (e.g., “I swim across the Atlantic Ocean daily to get to work”). Participants had to respond correctly to three out of four validity check items and correctly answer the nonsensical item to be retained; no cases were omitted based on these criteria.
Before examining study hypotheses, data were cleaned and assessed for assumptions of multivariate analysis of variance (MANOVA). Univariate outliers were examined by z-scores (
Chi-square analyses and analyses of variance (ANOVAs) were conducted to ensure that there were no demographic differences across study conditions. Notably, nonbinary participants were excluded from analyses involving gender due to sample size restrictions. There were no significant differences across study conditions based on race (white vs. people of color; X2[3] = 1.92, p = .59), sexual identity (heterosexual vs. sexual minority; X2[3] = 3.79, p = .39), gender (man vs. woman, transgender inclusive; X2[3] = 2.67, p = .45), feminist identity (no, yes, unsure; X2[3] = 9.94, p = .13), knowledge of someone who has had an abortion (yes, no, unsure; X2[3] = .61, p = .90), personal abortion history (X2[3] = 2.75, p = .43), religiosity (F[3, 233]) = 1.54, p = .21), or age (F[3, 233]) = .74, p = .53). Relations between potential demographic controls and dependent variables were examined via bivariate correlations. Significant relations emerged on several dependent variables based on gender and sexual identity; however, these variables were nonsignificant in the main analyses, did not change the overall pattern of results, and were therefore excluded.
To test our study hypotheses, we adopted a two-stage approach in which we first conducted multivariate tests examining the influence of the experimental conditions (i.e., main and interaction effects of sexual objectification and abortion circumstance) on the dependent variables of abortion stigma and dehumanization. In a subsequent analysis, we examined whether additional variance in study outcomes could be explained by ambivalent sexism. We adopted this two-stage approach to isolate the unique effects of sexual objectification and abortion circumstance on abortion stigma and dehumanization, given that research has not experimentally examined the effects of objectification and circumstance on these outcomes. Also, we anticipated ambivalent sexism would share significant overlapping variance with sexual objectification and abortion circumstance on stigma and dehumanization (Bareket & Fiske, 2023; Huang et al., 2014; Pacilli et al., 2018), and we initially sought to minimize its potential influence.
Manipulation Checks
Independent samples t-tests revealed that, compared to those in the nonsexually objectified condition, participants in the sexually objectified condition were more likely to view the target as attractive, t(235) = −2.802, p = .003, and sexy, t(235) = −3.585, p < .001 (see Table 4). Moreover, participants in the exceptionalized abortion condition were more likely than those in the typical abortion condition to believe the target had a traumatic experience, t(235) = 16.602, p < .001. Last, participants in the typical abortion condition were more likely than those in the exceptionalized abortion condition to view the target as promiscuous, t(235) = −2.997, p = .002 (see Table 5). Thus, we concluded that the prompts successfully manipulated sexual objectification and abortion circumstance.
Manipulation Check Assessing the Effects of Sexual Objectification.
Note. The same superscript denotes mean differences at p < .05.
Manipulation Check Assessing the Effects of Circumstance.
Note. The same superscript denotes mean differences at p < .05.
Bivariate Correlations
Sexual objectification (nonobjectified = 0; objectified = 1) was significantly positively correlated with punishment and significantly negatively correlated with moral patiency, demonstrating small effect sizes. Abortion circumstance (exceptionalized = 0; typical = 1) was significantly positively correlated with transgression of feminine ideals and punishment, and significantly negatively correlated with humanization and moral patiency, yielding small to near-large effect sizes. Hostile sexism was significantly positively correlated with transgression of feminine ideals and punishment, and significantly negatively correlated with humanization and moral patiency, demonstrating moderate to near-large effect sizes. Benevolent sexism was significantly positively correlated with transgression of feminine ideals and punishment, yielding near-moderate effect sizes (see Table 6).
Bivariate Correlations and Descriptive Statistics.
Note. *p < .05, **p < .01, ***p < .001.
Sexual Objectification and Abortion Circumstance
As stated in our analytic plan, we first assessed the extent to which the experimental manipulation of sexual objectification and abortion circumstance predicted abortion stigma and dehumanization using a 2 (abortion circumstance: typical vs. exceptionalized) × 2 (objectified vs. nonobjectified) MANOVA. The assumption of homogeneity of covariance matrices was violated (Box's M = 190.795, F[30; 148,742.229] = 6.163, p < .001) and Levene's test was significant for each dependent variable (ps
Post hoc analyses were examined applying a Bonferroni correction for multiple comparisons (see Table 7). Between-subjects effects revealed a significant univariate effect of sexual objectification on punishment and moral patiency; participants in the sexually objectified condition believed the woman deserved greater punishment and granted her lower moral patiency compared to those in the nonobjectified condition. Univariate effects based on objectification status were not observed for humanization and transgression of feminine ideals.
Follow-Up Univariate Analyses of Variance Predicting Abortion Stigma and Dehumanization.
Note. *p < .05, **p < .01, ***p < .001.
Significant univariate effects were also observed for abortion circumstance on transgression of feminine ideals, punishment, humanization, and moral patiency. Specifically, compared to those in the exceptionalized abortion condition, participants in the typical abortion circumstance rated the woman as higher in transgressing feminine ideals and deserving of punishment, while also granting her lower humanity and moral patiency (see Table 8).
Mean Differences by Experimental Conditions.
Note. The same superscripts denote mean differences at p < .05.
Last, a significant univariate interaction effect was observed between sexual objectification and abortion circumstance on humanization (see Figure 1 and Table 9), qualifying the main effect of abortion circumstance. To deconstruct the interaction, we estimated pairwise comparisons between the exceptionalized and typical reasons in the nonobjectified and objectified conditions. Specifically, in the nonobjectified condition, the woman elicited greater humanization when the abortion occurred for an exceptionalized reason than for a typical reason, F(1, 233) = 27.170, p < .001, Interaction Between Abortion Circumstance and Objectification as It Predicts Humanization. Means and Standard Deviations for Dependent Variables for Circumstance by Objectification. Note. The same superscripts denote mean differences at p < .05.
Taken together, these findings suggest that women who have abortions for more typical reasons are likely to experience less humanization than those whose abortions occur for exceptionalized reasons, and this may occur regardless of whether or not women are seen as sexual objects. However, the humanizing effects of exceptionalized abortions may be weakened when women are, in fact, seen as sexual objects. Sexual objectification and abortion circumstance did not significantly interact to predict punishment, moral patiency, or transgression of feminine ideals. Next, we sought to further examine how these relations may be affected by sexist attitudes.
The Moderating Role of Ambivalent Sexism
To examine the potential moderating role of ambivalent sexism for hypotheses 1 to 3, we next conducted a 2 × 2 MANOVA. The two- and three-way interactions involving the continuous predictors (i.e., hostile and benevolent sexism) were added to the model. Multivariate effects emerged for hostile sexism (centered), benevolent sexism (centered), the two-way interaction between abortion circumstance (0 = exceptionalized; 1 = typical) and hostile sexism, and the three-way interaction between sexual objectification (0 = nonobjectified; 1 = objectified), abortion circumstance, and hostile sexism.
No main multivariate effects were observed for sexual objectification or abortion circumstance in this analysis (see Table 10). At the multivariate level, objectification did not vary as a function of abortion circumstance, hostile sexism, or benevolent sexism, and abortion circumstance did not vary based on benevolent sexism. No significant multivariate effects were observed for the three-way interaction between objectification, circumstance, and benevolent sexism. Below, we describe the observed univariate main effects, two-way interaction, and three-way interaction.
Multivariate Effects Examining Objectification, Abortion Circumstance, Ambivalent Sexism, and Their Interactions.
Note. *p < .05, ***p < .001. BS = benevolent sexism; HS = hostile sexism.
Main Effects
Follow-up univariate analyses revealed that hostile sexism was related to greater transgression of feminine ideals (B = .390, SE = .123) and deserved punishment (B = .394, SE = .078), and lower humanization (B = −.079, SE = .081) and moral patiency (B = −.137, SE = .072, see Table 11). Benevolent sexism was associated with greater moral patiency (B = .075, SE = .071), but not significantly related to humanization, punishment, or transgression of feminine ideals.
Univariate Effects Examining Objectification, Circumstance, Ambivalent Sexism, and Their Interactions.
Note. *p < .05, **p < .01, ***p < .001. BS = benevolent sexism; HS = hostile sexism.
Two-Way Interaction
The two-way interaction between abortion circumstance and hostile sexism significantly predicted punishment (see Figure 2), suggesting that abortion circumstance qualified the positive association between hostile sexism and punishment. Simple slopes analyses, using PROCESS, Model 1 (Hayes, 2022), were used to interpret the interaction terms. Specifically, hostile sexism was associated with greater punishment in both exceptionalized (B = .119, SE = .041, t = 2.928, p = .004, 95% CI = .039 to .199) and typical abortion circumstances (B = .345, SE = .096, t = 3.589, p < .001, 95% CI = .156 to .534), although the effect was notably weaker among participants in the exceptionalized condition compared to the typical circumstance. Additionally, abortion circumstance did not predict punishment at low levels of hostile sexism (B = .068, SE = .071, t = .948, p = .344, 95% CI = −.073 to .208), but did so at moderate (B = .317, SE = .075, t = 4.230, p < .001, 95% CI = .170 to .465) and high levels of hostile sexism (B = .567, SE = .181, t = 3.131, p = .002, 95% CI = .210 to .924). Abortion circumstance did not vary as a function of hostile sexism to predict humanization, moral patiency, or transgression of feminine ideals.
Interaction Between Abortion Circumstance and Hostile Sexism as It Predicts Deserves Punishment.
Three-Way Interaction
Additionally, a three-way interaction between sexual objectification, abortion circumstance, and hostile sexism emerged on moral patiency, suggesting that abortion circumstance and sexual objectification qualified the association between more hostile sexism and less moral patiency. However, this interaction did not predict the other three dependent variables: humanization, punishment, or transgression of feminine ideals. Hayes's PROCESS macro, Model 3, was used to decompose the significant three-way interaction term. We first began by examining the effect of objectification condition as it varied by abortion circumstance and hostile sexism (see Figure 3). Regardless of whether the target was objectified or nonobjectified, the target elicited similar levels of moral patiency in the exceptionalized (B = −.102, SE = .120, t = −.850, p = .396, 95% CI = −.340 to .135) condition and in the typical condition (B = −.287, SE = .168, t = −1.705, p = .090, 95% CI = −.618 to .045) among participants with low levels of hostile sexism.
Influence of Objectification Status on Moral Patiency as It Varies by Abortion Circumstance and Hostile Sexism. Note. 0 = Exceptionalized; 1 = Typical.
Among participants with moderate levels of hostile sexism, the target in the exceptionalized condition elicited significantly less moral patiency when objectified than when nonobjectified (B = −.278, SE = .089, t = −3.105, p = .002, 95% CI = −.454 to −.101); however, for the target in the typical circumstance, moral patiency did not vary according to whether the target was objectified or not (B = −.061, SE = .116, t = −.524, p = .601, 95% CI = −.288 to .167). In a similar pattern, among participants with high levels of hostile sexism, the target in the exceptionalized condition elicited significantly less moral patiency when objectified than when nonobjectified (B = −.453, SE = .119, t = −2.274, p = .024, 95% CI = −.845 to −.061); again, in the typical condition, the target elicited similar levels of moral patiency regardless of objectification status (B = .166, SE = .211, t = .784, p = .434, 95% CI = −.250 to .581).
Additionally, we examined the effect of abortion circumstance as it varied by objectification status and hostile sexism to further understand this interaction. At low levels of hostile sexism, the degree of moral patiency extended to the nonobjectified target did not vary according to abortion circumstance (B = −.172, SE = .137, t = −1.255, p = .211, 95% CI = −.443 to .098); by contrast, for the objectified target participants, participants with lower hostile sexism extended less moral patiency when the abortion occurred for a typical versus exceptionalized reason (B = −.356, SE = .155, t = −2.304, p = .022, 95% CI = −.661 to −.052, see Figure 4). Among participants with moderate levels of hostile sexism, the nonobjectified target elicited significantly less moral patiency in the typical condition than the exceptionalized condition (B = −.334, SE = .105, t = −3.180, p = .002, 95% CI = −.542 to −.127). And, the objectified target elicited similar levels of moral patiency in both the typical and exceptionalized conditions (B = −.117, SE = .101, t = −1.156, p = .249, 95% CI = −.317 to .083). Similarly, among participants with high levels of hostile sexism, the nonobjectified target elicited significantly less moral patiency in the typical circumstance than in the exceptionalized circumstance (B = −.497, SE = .214, t = −2.316, p = .021, 95% CI = −.919 to −.074), but the objectified target elicited similar levels of moral patiency in both the typical and exceptionalized conditions (B = .122, SE = .196, t = .624, p = .534, 95% CI = −.263 to .507; see Figure 4).
Influence of Abortion Circumstance on Moral Patiency as It Varies by Objectification Status and Hostile Sexism. Note. Objectification: 0 = Non-Objectified; 1 = Objectified.
Finally, examining the influence of hostile sexism on moral patiency for each of the different conditions further elucidated the role of hostile sexism in the context of sexual objectification and abortion circumstance. For the nonobjectified condition, greater hostile sexism did not predict moral patiency under exceptionalized circumstances (B = −.155, SE = .097, t = −1.601, p = .111, 95% CI = −.345 to .036), but predicted lower moral patiency under typical circumstances (B = −.302, SE = .090, t = −3.348, p = .001, 95% CI = −.479 to −.124). For the objectified condition, greater hostile sexism predicted less moral patiency in the exceptionalized circumstance (B = −.313, SE = .079, t = −3.956, p < .001, 95% CI = −.470 to −.157), but not in the typical circumstance (B = −.097, SE = .104, t = −.936, p = .350, 95% CI = −.301 to .107). Hostile sexism was not significantly related to moral patiency for the objectified woman who had an abortion under typical circumstances because moral patiency was already low (see Figure 5). This may suggest that objectification and typical abortion contexts independently dehumanize women to such an extent that additional prejudice from hostile sexism has little room to exert further influence.
Influence of Hostile Sexism on Moral Patiency as It Varies by Abortion Circumstance and Objectification Status. Note. Circumstance: 0 = Exceptionalized; 1 = Typical.
Discussion
The purpose of this study was to integrate and extend scholarship on abortion stigma, sexual objectification, dehumanization, and ambivalent sexism. We theorized that stereotypes about women's sexuality would be invoked under typical abortion circumstances, in particular, and consequently affect abortion stigma and dehumanization. We believed this relation would be especially strong when women were sexually objectified and among perceivers with greater sexist attitudes toward women. Results of our study revealed how the sexual objectification of women, abortion circumstance, and ambivalent sexism contribute independently and in conjunction to affect perceptions of women who have abortions. Below, we outline the findings of the study and address implications and limitations of the research.
Sexual Objectification and Abortion Circumstance
We hypothesized that a sexually objectified woman who had an abortion would elicit greater abortion stigma and dehumanization than a nonobjectified woman who had an abortion, and this hypothesis was partially supported. Results revealed a significant effect of sexual objectification on moral patiency, such that the sexually objectified target was granted less moral concern than the nonobjectified target. Additionally, participants in the sexual objectification condition believed that the woman deserved greater punishment for having an abortion compared to participants in the nonobjectification condition. These findings align with prior research showing that a sexually objectified rape survivor was granted less moral patiency than a nonobjectified rape survivor, which resulted in greater rape blame and attribution of responsibility to the sexually objectified survivor (Loughnan et al., 2013). Our findings reveal similar harmful consequences for women who have abortions under sexually objectifying conditions. That is, reducing women who have abortions to their bodies and sexual functioning can heighten the degree to which others view them as deserving of punishment, and diminish the degree to which others care about their wellbeing or the harm that they may incur. Although conceptual work has extended objectification theory to abortion and reproductive decision making (Dyer et al., 2023), to our knowledge, this is one of the first empirical studies to examine this theory, providing initial support for the idea that objectified women are similarly stigmatized and denied moral concern in the context of abortion.
We were also interested in examining how abortion circumstance may affect stigma and dehumanization. Although miscarriage management and abortion care involve identical medical procedures and are not always fully distinct, we were especially interested in how abortion occurring for miscarriage management, fetal anomaly, and maternal life endangerment may be viewed compared to abortion occurring for more typical reasons (i.e., ill-timed). Because miscarriage may be more likely to be perceived as outside of a person's control, we hypothesized that a woman seeking an abortion for a typical reason would elicit greater abortion stigma and dehumanization compared to a woman seeking an abortion for this exceptionalized reason; this hypothesis was fully supported. Mirroring prior research (Pacilli et al., 2018, 2024), robust main effects emerged for abortion circumstance on all dependent variables. Compared to those in the exceptionalized abortion condition, participants in the typical abortion circumstance viewed the target as more transgressive of feminine ideals, more deserving of punishment, less humanized, and less morally patient. These results support Kumar et al.’s (2009) assertions that abortion stigma arises as a result of the perception that women who have abortions have transgressed feminine ideals, including the belief that women's sexuality is a vehicle for inevitable and instinctive motherhood. Such findings are especially troublesome, given that most women have abortions for typical reasons (Woodruff et al., 2024). Surveys suggest that, although most U.S. citizens support abortion access, there is somewhat less support for abortion under circumstances including financial insecurity, emotional readiness, relational instability, and other conditions viewed as within the pregnant person's control (Jozkowski et al., 2021; Pew Research Center, 2025). Such framing stigmatizes women who have abortions by ignoring the myriad intricate and interlocking social and political processes that affect sexual and reproductive health (North, 2019).
We also hypothesized an interaction between sexual objectification and abortion circumstance. Indeed, sexual objectification qualified the main effect of abortion circumstance on humanization. The target elicited greater humanization (i.e., perceived warmth, competence, and moral agency) in the exceptionalized abortion condition than in the typical abortion condition. Although this effect occurred regardless of objectification status, the effect was significantly weaker when the target was objectified than when nonobjectified. These findings suggest that women who have abortions under more exceptionalized circumstances are seen as more human than those who have abortion for more typical reasons. However, even in exceptionalized circumstances involving miscarriage, reducing women to sexual objects may result in less humanization. This finding builds on past work showing that sexual objectification causes dehumanization (Heflick et al., 2011; Heflick & Goldenberg, 2009; Loughnan et al., 2010) by extending these findings to the understudied context of women who transgress feminine ideals through abortion.
Taken together, these findings suggest that tropes and stereotypes of women's sexuality are invoked by varying sexual objectification and abortion circumstance. As indicated by our manipulation checks, participants in the typical condition viewed the target as more promiscuous than those in the exceptionalized condition, and accordingly they granted the target fewer human qualities. In contrast, in the exceptionalized condition, the target elicited greater humanization, although these humanizing effects were weaker when she was sexually objectified. Because prior research has also examined ambivalent sexism in connection with abortion under varying circumstances (Osborne et al., 2022), we sought to further understand how sexism may shape abortion stigma and dehumanization of women who have abortions when they are reduced to a sexual object (Glick & Fiske, 1996).
Ambivalent Sexism, Stigma, and Dehumanization
We hypothesized that both hostile and benevolent sexism would uniquely relate to and interact with sexual objectification and abortion circumstance to affect abortion stigma and dehumanization, and this hypothesis was partially supported. Notably, when adding ambivalent sexism into the model, the main effects of sexual objectification and abortion circumstance were no longer significant, suggesting that ambivalent sexism, and specifically hostile sexism, accounted for a large degree of variance in abortion stigma and dehumanization. Indeed, hostile sexism was uniquely related to each dependent variable, including greater perceived transgression of feminine ideals and punishment, as well as less humanization and moral patiency. In contrast, benevolent sexism was only related to greater moral patiency. Such benevolent sexist attitudes are ostensibly supportive of women, yet can advance harmful narratives that abortion is inherently traumatizing and that pregnant people are incapable of truly understanding the repercussions of abortion (Reardon, 2018; Roberti, 2021). These paternalistic views suggest that pregnant people lack competency and intelligence—key aspects of human uniqueness (e.g., Dyer et al., 2023; Haslam, 2006).
Although the main effect of abortion circumstance became nonsignificant in this model, abortion circumstance interacted with hostile sexism to predict punishment, suggesting that the effects of hostile sexism on punishment depended on abortion circumstance. That is, at low levels of hostile sexism, abortion circumstance did not predict punishment. However, at moderate and high levels of hostile sexism, participants endorsed greater punishment for the woman, Laura, although these effects were stronger in the typical condition compared to the exceptionalized condition. These findings are not merely theoretical or without consequence. Several states (e.g., Texas, Idaho, and Oklahoma) have enacted bounty hunter laws whereby private citizens are incentivized to sue those who perform or “aid and abet” abortion. Additionally, in the year following the Dobbs decision, over 200 pregnant people faced criminal charges in connection with their pregnancies—the highest number documented in a single year (Bach & Wasilczuk, 2024). Most of these individuals were from low-income backgrounds, with information obtained from a medical system or provider. Additionally, the majority of the charges (n = 191) did not require proof of harm to a fetus or infant. Although only five of these cases involved allegations of abortion or attempted abortion, in the current sociopolitical context, pregnant people may be at risk of facing dire consequences.
Moreover, the main effect of hostile sexism on less moral patiency was affected by abortion circumstance and sexual objectification, as evidenced by the three-way interaction. In the typical abortion condition, moral patiency was similarly low and did not differ based on sexual objectification status among those with moderate and high levels of hostile sexism. However, in the exceptionalized condition, the sexually objectified target elicited less moral concern (similar to women in the typical circumstance) than the nonobjectified target. Nonobjectified targets elicited less moral patiency under typical abortion circumstances compared to exceptionalized abortion circumstances, with objectified targets generally eliciting similar levels of moral patiency (i.e., lower)—regardless of abortion circumstance—among those with moderate and high levels of hostile sexism. The dehumanizing effects of objectification also occurred among those with low levels of hostile sexism, but the pattern was different; participants were less likely to extend moral patiency to the objectified target under typical abortion circumstances versus exceptionalized circumstances.
Together, these results suggest that moral responsiveness to women who have abortions is shaped by hostile sexism, objectification, and abortion circumstance. Individuals higher in hostile sexism are generally less willing to extend moral patiency to women who have an abortion for typical versus exceptionalized reasons under nonobjectifying circumstances. However, any humanizing effects of exceptionalized abortions among these individuals diminish when women are viewed as sexual objects. In other words, even in the context of abortion occurring as a result of miscarriage management, fetal anomaly, and maternal life endangerment, individuals who endorse high levels of hostile sexism are less likely to extend moral patiency to women receiving abortion care when they are seen as sexual objects. In addition, even among individuals who report low levels of hostile attitudes toward women, they may be less likely to extend moral concern to women they see as sexual objects when the abortions occur for typical versus exceptionalized reasons. These patterns highlight how both sexual objectification and abortion circumstance constrain who is seen as morally worthy, particularly among those with strong animus toward women, but also among those whose animus is less pronounced.
These findings align with prior research and theorizing. Dyer et al. (2023) found that women-centered anti-abortion rhetoric often uses benevolently sexist and objectifying language, portraying pregnant people as “precious objects.” However, according to Dyer et al., once someone has an abortion, they will be regarded as a “useless object.” Consistent with this shift, participants in the present study rated the objectified woman who had an abortion under typical compared to exceptionalized circumstances as less human, a pattern that aligns with traditional understandings of abortion stigma (Kumar et al., 2009) and dehumanization (Pacilli et al., 2024). Relatedly, Moradi et al. (2025) found that objectifying views of women were related to lower support for access to abortion under exceptionalized and typical circumstances due to sexist attitudes toward women (Moradi et al., 2025). Additionally, our results support findings by Pacilli et al. (2024) who found that women who have abortions for typical reasons are more likely to be met with moral outrage than those who have abortions for more exceptionalized reasons. Our results extend these lines of work, however, by demonstrating that, even under exceptionalized circumstances, women viewed as sexual objects may be denied moral patiency.
Strengths, Limitations, and Future Research
Using an experimental design, we were able to draw causal inferences about the ways in which sexual objectification, abortion circumstance, and ambivalent sexism affect abortion stigma and dehumanization. Many threats to internal validity were managed as a result of the experimental design and random assignment to experimental conditions. Yet, it is important to consider the context under which these data were collected, which occurred after the overturning of Roe v. Wade by the U.S. Supreme Court. As such, participants’ reactions to the overturning of this historic federal right may have affected responses to survey items. Subsequent research on this topic may wish to better understand how such reactions may affect relations among variables, as more adverse reactions to the Supreme Court's decision have been shown to enhance psychological distress among cisgender women and people assigned female at birth (Watson & Germain, 2024).
Although the experimental design enhanced internal validity, the controlled nature of experiments may not necessarily generalize to real-life contexts. In the present study, we sought to mitigate this concern by using vignettes that paralleled real-world and naturally occurring examples of women seeking abortions (e.g., Cohen & Bonifield, 2023; Diamant & Mohamed, 2023). Accordingly, the typical abortion vignette noted the woman's volition that having an abortion was the best decision for her. By contrast, the exceptionalized abortion vignette included miscarriage with health risk to the pregnant person and the fetus, reflecting prominent coverage of real-life cases of denying women abortion care under these conditions (e.g., Jaci Statton; Cohen & Bonifield, 2023). Both vignettes ended with a direct statement that Laura had an abortion. It is possible that despite this clear statement, the mention of miscarriage in the exceptionalized condition may have led some participants to view miscarriage management as distinct from abortion, perhaps because the miscarriage was considered outside the realm of the woman's control whereas the abortion was considered “elective.” Importantly, it is precisely this perception of unavoidable/involuntary versus avoidable/voluntary abortion circumstances that is theorized to shape greater stigma toward those who have typical versus exceptionalized abortions (Jozkowski et al., 2021; Osborne et al., 2022; Smith & Son, 2013). Notably, the objective of experimental research is not always to generalize findings, but rather to assess generalizations and whether they occur in controlled conditions (Mook, 1983). Thus, it will be useful to evaluate the present effects under different conditions that do not involve miscarriage, such as other exceptionalized (e.g., rape and incest) and typical (e.g., financial constraint) circumstances.
Additionally, our findings may not be representative of a wide range of attitudes toward abortion. Our participants who were recruited through Prolific generally reported pro-choice beliefs, though there was also substantial variability. Given the nature of this study, some participants may have been incentivized to downplay stigmatizing attitudes. Although controlling for social desirability rarely yields different results, particularly in contexts not characterized by high-stakes decision making (Tracey, 2016), it is possible that examining implicit bias toward people who have abortions may be helpful in better understanding predictors of abortion stigma and dehumanization. However, most participants in this study espoused pro-choice beliefs and results still elucidated stigmatizing attitudes toward women who have abortions. These findings parallel those by Baker et al. (2023) in which those who purportedly held pro-choice beliefs still openly recommended punishment for women who have abortions. In other words, people are not necessarily highly motivated to conceal their anti-abortion beliefs and bias. Moreover, most participants in this study were white and heterosexual with some college background. Thus, additional studies that seek to better understand how these variables may be affected among those with more diverse viewpoints and identities may be helpful.
Given that this was a vignette study, we created and adapted measures to assess context-specific aspects of abortion stigma and dehumanization of the target. To this end, we used face valid items of abortion stigma (e.g., Kumar et al., 2009) and dehumanization measures that have been used in past experimental research on sexual objectification (e.g., Heflick et al., 2011; Loughnan et al., 2010). We also garnered strong psychometric support for the measures in the present study, which fosters confidence in our results. Most work on abortion stigma using validated measures assesses experienced abortion stigma from people who have had abortions (Hanschmidt et al., 2016), not enacted stigma from the perceiver perspective—thus, this addition to the literature could represent a methodological contribution in and of itself. Future research is needed to further validate these measures. Finally, our vignette included an image of a white woman, although race and other forms of stigma and oppression affect the degree to which women are objectified and encouraged to reproduce (SisterSong, n.d.; Ross, 2020). Thus, we encourage additional research that examines the ways in which abortion stigma and dehumanization may be affected by additional oppressive ideologies. These perceptions may also be shaped by demographic characteristics of the perceivers, such as race, transgender and nonbinary identities, and ability status. Such research could elucidate how additional forms of oppression—beyond sexism—may operate in shaping abortion stigma and dehumanization of women who have abortions.
Practical Implications
There are multiple considerations based on the findings from this study. First, clinicians should carefully examine their own attitudes, feelings, and ideas about abortion so as not to dehumanize and stigmatize their clients who may have had or may seek abortions. Indeed, mental health providers have nuanced and at times complicated feelings about abortion (Dyer et al., 2025) and are also influenced by abortion misinformation (Mollen et al., 2018). Mental health providers should also understand that clients who seek and have abortions are likely to encounter stigmatizing attitudes and behaviors, regardless of the circumstances surrounding their abortions, such as the belief that they have transgressed feminine ideals and deserve punishment. Our findings reveal that this is particularly the case among those who hold sexist attitudes toward women and when abortion is sought for more typical reasons. This merits great consideration for practitioners in states with restrictive laws against abortion—where women and even those assisting them may experience punishment and criminalization.
In terms of dehumanization, sexual objectification and hostile sexism appear to play a particularly deleterious role. Even under exceptionalized circumstances involving miscarriage management, women who are viewed as sexual objects may be granted less moral concern among those with moderate and high levels of hostile sexism. For women living in the 21 states with complete abortion bans or severe restrictions, these findings suggest that women may face intensely dehumanizing evaluation for an abortion, regardless of the situation surrounding their pregnancy. With the reversal of EMTALA's application to emergency abortion care, women seeking abortions for miscarriage management may be at risk of not being extended moral concern, particularly if they are viewed in an objectifying manner and by someone with hostile sexist ideology. Indeed, several women have lost their lives as a result of being denied emergency abortion care for miscarriage (e.g., Candi Miller, Amber Nicole Thurman, Nevaeh Crain, and Josseli Barnica). Understanding women's experiences with abortion in the context of sexist sociopolitical context that seeks reproductive control is imperative for effective care among practitioners.
Our findings also yielded important insights into the perception of women who have abortions, which can serve as important points of intervention. For example, interventions focused on reducing hostile sexism, abortion myths, and the sexual objectification of women may reduce abortion stigma and dehumanization of those who have abortions. Developing humanizing and accurate legislation and policy may also play a role in shifting abortion attitudes and prevent the enactment of abortion stigma before it occurs.
Conclusion
In this study, we sought to empirically examine how sexual objectification, abortion circumstance, and ambivalent sexism combine to shape abortion stigma and dehumanization of women seeking abortion. Multivariate analyses revealed that sexual objectification was related to less moral patiency and perceptions of greater deserved punishment, and also interacted with abortion circumstance to affect humanization. Our findings suggested that women who seek abortions for typical reasons are dehumanized to a greater degree, regardless of whether they are seen as sexual objects. In contrast, women who have abortions for exceptionalized reasons (i.e., miscarriage management, fetal anomaly, and maternal life endangerment) are humanized, unless they are seen as sexual objects. These findings also extended to moral patiency, as women seeking abortions for exceptionalized reasons were granted less moral patiency when presented in an objectified (versus non-objectified) manner among those with moderate and high levels of hostile sexism. Together, our findings reveal multiple factors that may affect the degree to which women who seek abortions are stigmatized and dehumanized.
Supplemental Material
sj-docx-1-pwq-10.1177_03616843261437756 - Supplemental material for Abortion Stigma and Dehumanization as Reproductive Control: The Roles of Sexual Objectification, Abortion Circumstance, and Sexism
Supplemental material, sj-docx-1-pwq-10.1177_03616843261437756 for Abortion Stigma and Dehumanization as Reproductive Control: The Roles of Sexual Objectification, Abortion Circumstance, and Sexism by Laurel B. Watson, Sarah J. Gervais, Bonnie Moradi, Jacob M. Germain, Tessa R. Graf, Kathleen O'Shea1 and Taylor D. Michl in Psychology of Women Quarterly
Footnotes
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Dalee fund and Elizabeth Noble Fund.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. This manuscript was reviewed by a guest editor and associate editor, external to the PWQ review board. The authors thank the guest editors for their work and upholding the rigor and integrity of the peer review process.
Data Availability Statement
Availability of supporting data is available by contacting the corresponding author. Authorship order is reflective of the nature and degree of contribution.
Supplemental Material
Supplemental material for this article is available online.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
