Introduction
In June of 2022, the Supreme Court decided Dobbs v. Jackson Women’s Health Organization and, in that 6-3 decision, overturned decades of precedent regarding reproductive rights. The majority opinion in Dobbs states six times that this decision returned the question of the legality of abortion to the people’s elected representatives, “Roe and Casey must be overruled, and the authority to regulate abortion must be returned to the people and their elected representatives,” and return this issue to the states it did.Footnote 1 The decision kicked off a flurry of legislation with twenty states and Washington DC adding new protections for abortion or expanded access as of February 2024. However, in the same time period, 21 other states have enacted outright bans or restricted abortion in a way that would have been prohibited under Roe v. Wade and Planned Parenthood of Southeast Pennsylvania v. Casey.
While some may have been shocked by the Dobbs decision, the reality was that this outcome was expected. The subsequent legislation coming from the states was not much of a surprise to those who followed the legal and political landscape around abortion rights either. Rather, the decision gave pro-life advocates the space to enact the restrictions or bans on abortion with which they have been experimenting and discussing for decades. Between 2010 and 2020, state legislatures have passed over 100 regulatory abortion bills, a number which far surpasses any other 10-year period since its legalization in 1973 (Nash et al., Reference Nash2018).Footnote 2 Further, while state legislatures have been active in passing abortion restrictions both before and after the Dobbs decision, a more subtle but significant change has also been occurring alongside it. Specifically, the language used to promote these restrictions has undergone a transformation.
It is well understood that issue frames can be influential in shaping public opinion, especially when employed by political elites (Chong and Druckman Reference Chong and Druckman2007; Druckman Reference Druckman2001). What is also interesting is that, as a rhetorical tool, frames are dynamic and can transform if the frame did not elicit the desired effect. For decades, the pro-life movement had used the frame of “fetal rights” in their efforts to ban or restrict abortion. However, scholars have shown the framing of the pro-life movement to have shifted from the right to life of the fetus — or fetal personhood frames — to a pro-woman frame (Gee Reference Gee2007; Roberti Reference Roberti2021; Rose Reference Rose2011; Siegel Reference Siegel2007, Reference Siegel2008). While the fetal rights frame still exists, scholars have documented a rise of a “pro-woman” frame. The pro-woman frame asserts that regulations on abortion will both protect women from potential harms of abortion and empower their decision-making through further information on those harms. This frame uses feminist-adjacent language that can be helpful in obscuring and insulating pro-life policy from the scrutiny of the public, which is supportive of abortion rights.Footnote 3
Pro-life advocates historically and contemporarily have employed a variety of frames; fetal personhood, pro-woman, states’ rights, and taxpayer rights, to name a few (Saurette and Gordon Reference Saurette and Gordon2016). However, the pro-woman frame has come to dominate the pro-life movement’s rhetoric inside and outside the halls of government since the early 2000s. Though scholars have acknowledged this frame transformation from fetal personhood to pro-woman (Reingold et al. Reference Reingold, Kreitzer, Osborn and Swers2015; Roberti Reference Roberti2021; Rose Reference Rose2011; Saurette and Gordon Reference Saurette and Gordon2016; Siegel Reference Siegel2008), no study has tested how this framing is evaluated by the public. The data presented in this paper offers a novel way to understand the power of these frames and provides a level of experimental control that the real world does not. While this topic has been important and on the radar of many for some time now, Dobbs has brought its importance into stark focus. It is necessary to understand the persuasiveness of the pro-woman frame on the unpopular policy standpoint of pro-life legislation.
In this study, we hypothesize that a pro-woman framing of pro-life policies will be evaluated more positively compared to the more traditional fetal personhood frame; however, these evaluations will be influenced by the party and gender of those using the frame. This hypothesis points to a strategic employment of the pro-woman frame by policymakers and activists for several distinct reasons: to soften or lessen the backlash of their unpopular policy positions (or at least muddle views of their positions), participate in “frame extension” (Rose Reference Rose2011) which allows them to capture more support for their issue, and to provide important representational claims-making that they are authentically concerned with women’s best interests. Additionally, as reproductive politics is seen as a “women’s issue,” we assess the gender dynamics of the use of frames in pro-life rhetoric. We manipulate the gender of the lawmaker using the frame to test the well-known adage of scholars of women and politics that women lawmakers are seen as more credible and knowledgeable about women’s issues legislation (Carroll Reference Carroll2001; Carroll, Dodson, and Mandel Reference Carroll, Dodson and Mandel1991). Finally, we also manipulate the partisanship of our source cue as abortion has been a hallmark issue of the culture wars and the growing partisan divide in our current political climate. Further, these pro-woman frames have been employed by both Republican women and more conservative Democratic women when they have proposed legislation that would regulate abortion access (Roberti Reference Roberti2021). The combination of frame, gender, and party therefore may help to explain the mismatch between public attitudes and legislative agendas regarding abortion.
Literature Review
Pro-Woman and Fetal Rights Frame Transformation
Abortion has been framed by a variety of actors and in numerous ways since its appearance in US political discourse, yet the persuasiveness of the frames remains understudied. Scholarship has traced the different framing of abortion in the media (Ferree Reference Ferree2002, Reference Ferree2003; Terkildsen, Schnell, and Ling Reference Terkildsen, Schnell and Ling1998), in social movements (Cannold Reference Cannold2002; Lambert, Hackworth, and Billings Reference Lambert, Hackworth and Billings2023; Martin et al. Reference Martin, Beacken, Trauthig and Woolley2024; McCaffrey and Keys Reference McCaffrey and Keys2000; Rohlinger Reference Rohlinger2002; Rose Reference Rose2011; Siegel Reference Siegel2008), and in legislation (Reingold et al. Reference Reingold, Kreitzer, Osborn and Swers2015; Roberti Reference Roberti2021). For example, Freedman (Reference Freedman1999) finds that social movement actors can effectively manipulate the public’s attitudes toward abortion if they employ alternative frames. Similarly, Kalla, Levine, and Brockman (Reference Kalla, Levine and Brockman2022) find that a “personalized moral reframing” (1240) — one that matches the values of a person — can be employed to persuade people on the issue of abortion. Additionally, a wealth of scholarship has studied the varied ways in which abortion attitudes are formed (Wilcox Reference Wilcox1992) including religious influence (Strickler and Danigelis Reference Strickler and Danigelis2002) and race (Wilcox Reference Wilcox1990).
Opponents of abortion have framed their standpoints in numerous ways: as a matter of taxpayer rights following the passage of the Hyde Amendment, religiously based morality, fetal rights, and states’ rights (Saurette and Gordon Reference Saurette and Gordon2016).Footnote 4 However, the major frame that has dominated the pro-life side of the abortion debate since its legalization in 1973 has been a fetal-centric frame based on the right to life of the fetus. The fetal rights frame was useful to the pro-life movement, as it was simple: the fetus is no different than a child, and abortion is akin to murder. Rooted in a moral standpoint of fighting for the rights of the unborn, who are voiceless and without real representation in the body politic, this frame was the “moral center” of the pro-life movement for decades (Doan and Ehrlich Reference Doan, Ehrlich, Campo-Engelstein and Burcher2017, 128).
A main feature of rhetorical frames is that they can be fluid; elites can use different frames for specific audiences, making a stronger psychological connection to a particular group of people (Tarrow Reference Tarrow2011). Frames also can be multidimensional, employing numerous frames, or shifting in response to the opposition’s arguments (Mucciaroni, Ferraiolo, and Rubado Reference Mucciaroni, Ferraiolo and Rubado2019). When the understanding of an issue begins to shift, or an “organization’s perspective on an issue does not resonate with, or is antithetical to, conventional understandings of the world” (Trumpy Reference Trumpy2014, 167), that organization may change its framing to suit the current paradigm. As such, issues can be transformed from traditional to new because a frame shifted the value dimension (Chong and Druckman Reference Chong and Druckman2007). In abortion politics, this could not be more evident.
At the latter end of the 1980s, the pro-life movement was becoming synonymous with hostility toward women and extreme: dramatic protest tactics, heavily Evangelical discourse of groups such as Operation Rescue, and violence against clinics, staff, and providers (Cohen and Connon Reference Cohen and Connon2015; Huff Reference Huff2014; Maxwell Reference Maxwell2002; Rose Reference Rose2011; Saurette and Gordon Reference Saurette and Gordon2016). Furthermore, it became clear that the public sentiment on abortion was not changing, as polls indicated that the increasingly vivid portrayals of “unborn victims of abortion” remained unpersuasive to the general public — in fact, from 1988 to 1992, support for abortion under any circumstance enjoyed a 10-point increase (from 24% to 34% support).Footnote 5 Finally, the pro-life movement largely considered the ruling in Casey to be a loss, as the case failed to dismantle Roe completely (Von Hagel and Mansbach Reference Von Hagel and Mansbach2016). As any successful movement would do, the pro-life movement shifted its rhetoric to meet the moment.
A slow-moving shift in rhetoric began to occur, motivated by numerous well-known pro-life activists and groups, to center women and the harm abortion causes them. Women Exploited by Abortion (WEBA) was one such group: beginning in 1982, women who had abortions both before and after Roe began chapters throughout the US to discuss abortion as harmful beyond “killing babies” (Diamond Reference Diamond1989). In 1985, at the height of WEBA, Pro-life activist David C. Reardon conducted interviews with over 200 women who belonged to WEBA about their experiences with abortion. Far from an ethically sound qualitative study, it culminated in a book entitled Aborted Women: Silent No More (Reference Reardon1987), where he argued that women are harmed physically and psychologically by abortion, leading to a disorder called “post-abortive syndrome.” Proponents of the shift in focus to a pro-woman frame were also inspired by language in the US Supreme Court’s opinion in Webster v. Reproductive Health Services (Saurette and Gordon Reference Saurette and Gordon2016), and the diligent advocacy by pro-life women activists working in crisis pregnancy centers (Doan and Ehrlich Reference Doan, Ehrlich, Campo-Engelstein and Burcher2017; Kelly Reference Kelly2014).
Though the notion of “post-abortive syndrome” has been thoroughly repudiated by major US medical institutions including the American College of Gynecologists and Obstetricians, the evolution of the frame was in motion. In 1997, the former president of the National Right to Life, John Willke, and his wife Barbara — also an avid pro-life activist and leader of the local Cincinnati Right to Life group — authored Why Can’t We Love Them Both? (Reference Willke and Willke1997), which detailed a focus group-tested strategy to center women in pro-life rhetoric in order to capture the “mushy middle” of the public (Wilke and Wilke Reference Wilke and Wilke1997). The pro-woman frame was further enmeshed in pro-life rhetoric following the Gonzales v. Carhart (2007) case, where Justice Kennedy cited an amicus brief from pro-life group Operation Outcry when he wrote, without evidence, that it is “unexceptionable to conclude that some women come to regret their choice to abort the infant life they once created and sustained” (159). In an effort to revive their public image, dispel the myths of being anti-woman, and win over new audiences, this frame became ubiquitous in pro-life rhetoric (Doan, Candal, and Sylvester Reference Doan, Candal and Sylvester2018; Roberti Reference Roberti2021).Footnote 6
A small but growing literature has identified, defined, and analyzed the pro-woman frame employed by the pro-life movement as one that situates abortion as harmful for women, thus necessitating its regulation or criminalization (Doan, Candal, and Sylvester Reference Doan, Candal and Sylvester2018; Doan and Ehrlich Reference Doan, Ehrlich, Campo-Engelstein and Burcher2017; Siegel Reference Siegel2008).Footnote 7 The pro-woman frame is different than other frames found within pro-life rhetoric as it centers the woman as the subject (Cannold Reference Cannold2002), or the victim of abortion, instead of the fetus.Footnote 8 This frame has gone by different names in scholarship--pro-woman, pro-woman pro-life, woman-protective — but it always focuses on women and the harm that can befall women who obtain an abortion. Gee (Reference Gee2007) deems the pro-woman frame a “feminization” of abortion rhetoric, where there was a “repositioning those opposed to abortion as protectors, rather than critics, of women” (982).
Expanding on the “protective” nature of this frame, Siegel (Reference Siegel2008) uncovers the frame’s link between abortion and physical and mental injury. Women are often cast as victims of abortion and abortion providers, or of their own decisions in that they later regret their choice (Doan, Candal, and Sylvester Reference Doan, Candal and Sylvester2018; Doan and Ehrlich Reference Doan, Ehrlich, Campo-Engelstein and Burcher2017; Siegel Reference Siegel2008). In a more recent analysis, Roberti (Reference Roberti2022) shows how the frame is more nuanced, containing elements of protective language as well as feminist or empowering language: “[t]he ‘pro-woman’ frame is a rhetorical device that, when used in abortion politics, situates the state as empowering women through regulatory abortion policy: educational informed consent policies, ultrasound mandates, and anything that might inform reproductive decision-making” (140). Using “choice” and “opportunity” language — as in women’s choices would be more informed, or they would have an opportunity to hear a description of the ultrasound — specifically reinforces a façade of autonomous decision-making by the woman. However, this is still predicated on receiving the right kind of information, which the state decides, in order to make a rational “autonomous” decision (Denbow Reference Denbow2015). The co-option of feminist language is a strategic choice to further counteract the narrative that the pro-life movement is anti-woman. As a result, we expect that this type of language will contribute to a greater receptivity to the pro-woman framing compared to the pro-fetus framing.
The pro-woman frame can be useful to the pro-life political elite; by casting abortion as harmful to women, both psychologically and physically, a pro-life politician can soften the appearance of their unpopular standpoint on abortion. Women, they claim, would be protected from the harms of abortion and empowered in their decision-making on such a difficult matter through their legislation. This frame, though greeted with skepticism at first (Rose Reference Rose2011), resonated with enough elites both inside and outside of government that its usage has steadily increased since the 1990s to today, where it is commonplace in current abortion debates and legislation. The pro-woman frame thus shifts the abortion discussion from being a tension between woman and fetus to one of protecting women and, by implicit or explicit extension, the fetus too. As Mucciaroni, Ferraiolo, and Rubado (Reference Mucciaroni, Ferraiolo and Rubado2019) argue, this change from abortion as a binary debate between “life” and “choice” is important as advocates seek to move those in the target audience who remained undecided on the issue. Williams (Reference Williams2013), for example, shows how the pro-life movement evolved its framing and arguments in an effort to attract support, or at the least, increase public openness to their policy goals. However, the question remains: to what extent is this frame persuasive to the public? We expect that:
H1: Regardless of the gender or party cue the pro-woman frame will be viewed as doing more to protect women, respondents will agree more with it than the fetal right frame, and it will be viewed as stronger, more believable than the fetal rights frame.
Gender, Partisanship, and Framing Effects
Because we hypothesize that who uses the frame is as important as the substance of the frame, it is necessary to theoretically situate the gender and party dynamics of representation and issue framing. Though scholarship has shown that women lawmakers have highly diverse legislative portfolios (Atkinson and Windett Reference Atkinson and Windett2019), a wealth of scholarship has also indicated that women lawmakers are more likely to be seen as experts on “women’s issues” legislation such as health, welfare, education, civil rights, and policies on children and families (Carroll Reference Carroll2001; Dodson Reference Dodson2006; Dodson and Carroll Reference Dodson and Carroll1991; Reingold Reference Reingold2000; Reingold et al. Reference Reingold, Kreitzer, Osborn and Swers2015; Saint-Germain Reference Saint-Germain1989; Swers Reference Swers2002), are more likely than their male counterparts to bring forth such legislation (Volden, Wiseman, and Wittmer Reference Volden, Wiseman and Wittmer2013; Reingold Reference Reingold1992), and are more likely to champion it as it is implemented (Lowande, Ritchie, and Lauterbach Reference Lowande, Ritchie and Lauterbach2019).Footnote 9 Abortion is a prime “women’s issue” as it deals most intimately and directly with the reproductive capacity of women and their freedom of bodily autonomy as determined by the state. However, abortion has also been raised most prominently by women — Democratic women — in politics for decades (Swers Reference Swers2002; Volden, Wiseman, and Wittmer Reference Volden, Wiseman and Wittmer2013). The public may thus see women lawmakers speaking on abortion, from any standpoint, as automatically more credible by nature of their gender identity.
In the past, Republican male lawmakers were the most active in introducing specifically pro-life legislation (Osborn Reference Osborn2012), with Republican women shying away from being the voice of the issue to avoid being seen as niche (Swers Reference Swers2023). Republican women were also simply more moderate on abortion as well, holding pro-choice views, or even simply disagreeing with abortion being brought to the floor as an issue (Wineinger Reference Wineinger2022).Footnote 10 However, scholarship has indicated that since the 1990s, Republican women have become more conservative, deviating less on floor votes and aligning ideologically more closely to other conservative Republicans (Rolfes-Haase and Swers Reference Rolfes-Haase and Swers2022). The Tea Party movement in 2010 additionally ushered in a more conservative and vocally pro-life contingent of Republican women in various state and federal offices.
At present, Republican women lawmakers are overrepresented in introducing legislation regulating access to abortion and will do so more often than their male colleagues (Reingold et al. Reference Reingold, Kreitzer, Osborn and Swers2015; Roberti Reference Roberti2021; Swers Reference Swers2023), thus becoming the public face of the pro-life movement in government. As they have increasingly become the messengers of pro-life views, Republican women lawmakers have increasingly employed pro-woman framing of their pro-life legislation (Cannold Reference Cannold2002; Reingold et al. Reference Reingold, Kreitzer, Osborn and Swers2015; Roberti Reference Roberti2021). Wineinger (Reference Wineinger2022) finds evidence that the gendered nature of who delivers the pro-life message is indeed a strategic tactic of women in the Republican Party. By employing pro-woman frames, Republican women can eschew the unease the public may feel — especially women — with men in elected office talking about women’s bodies.
This is also a way to link their substantive and descriptive representation. When women lawmakers assert that they act both for women and as women in their support of pro-life legislation, they follow a long-employed strategy of conservative women’s groups. Literature on anti-feminist women’s groups has found that they often use the feminist-adjacent rhetoric in their justifications of a variety of anti-feminist legislation in addition to abortion, such as in their opposition to the Equal Rights Amendment (Roberti Reference Roberti2022; Schreiber Reference Schreiber2012; Smith Reference Smith2014). In this way, the pro-woman frame serves as a vehicle for claims-making; pro-life women lawmakers can claim that they are more authentically concerned for the health and welfare of women — often in comparison to the early feminist movement (Trumpy Reference Trumpy2014) — and boost their expertise and credibility on abortion as a “women’s issue.”
This is also useful for Democratic women. As Democratic women lawmakers are more likely to be pro-choice, pro-life Democrats “trespass” on their party’s typical issue position, and face an enhanced likelihood of citizen misperception of their standpoint (Koch Reference Koch2001). On an issue as contentious as abortion, with the public’s attitudes only becoming more favorable to abortion since 2019, a frame that subverts the prototypical abortion rhetoric and co-opts a feminist rhetoric that citizens may expect of a Democratic woman lawmaker could be strategically cogent. Democratic women thus have two identities on their side: their gender and their partisanship. The public may expect them to focus on a woman’s issue like abortion and may be persuaded by a message that sounds feminist, even if the policy is not (Swers and Larson Reference Swers and Larson2005). Because women legislators are seen as experts on women’s issues, and because the pro-woman frame employs feminist-sounding rhetoric of “protecting” women or “empowering” their decisions, we expect that:
H2a: Looking at the pro-woman frame, specifically women lawmakers will be seen as more credible, more believable and as sending a stronger message in contrast to their male counterparts.
However, because there is also a strong party dimension as well, we assert the following:
H2b: Democratic women will be seen as the most credible, most believable, and as sending the strongest message both overall and when looking specifically at the pro-woman frame.
Persuasiveness to the Public
Though frames can be erroneously interpreted as talking points, in which elites engage in rhetoric without truly being engaged with each other, scholars have found that frames are most effective when they avoid “talking past” the rhetoric of the opposition (Ansolabehere and Iyengar Reference Ansolabehere and Iyengar1994; Jerit Reference Jerit2008). Again, the rhetorical framing of the abortion debate is a perfect example of this effect. The pro-woman frame has the potential to speak directly to arguments made by reproductive rights policy actors by asserting that to be pro-life is to be more “authentically” feminist (Roberti Reference Roberti2021, Reference Roberti2022; Trumpy Reference Trumpy2014). This also highlights the importance for pro-choice policy makers and activists to be attentive, and responsive, when this language is deployed by pro-life actors in policy debates.
Pro-choice activists have historically claimed to be supporters of women’s rights and health, and that the pro-life movement is hostile toward women.Footnote 11 To engage with and disrupt this pro-choice frame, the pro-life movement’s shift toward a pro-woman frame argues that abortion harms women and that the state, through regulations, can protect women from the negative physical and psychological effects of abortion (Cannold Reference Cannold2002; Rose Reference Rose2011; Siegel Reference Siegel2008). In this way, the pro-woman frame would allow the pro-life movement to claim that they are concerned with women’s health and wellbeing, and at the same time move away from the more violent or hostile factions of the pro-life movement or the idea that there is a “war on women” being waged by conservatives. Burke and Bernstein (Reference Burke and Bernstein2014) identify this type of counter-framing as a co-optation or appropriation of opposition movement rhetoric.
This tactic of lawmakers using feminist-adjacent rhetoric has been employed in other venues aside from abortion policy. However, it could be especially persuasive in the abortion context, where abortion rights and abortion restrictions are in the jurisdiction of the states and often on the ballot. When majorities of the public are in support of legal abortion, but a pro-life ballot measure is facing a vote, changing minds or even muddling the message at the margins can mean an election day win even for an unpopular measure.Footnote 12 For example, since Dobbs, the issue of abortion has been on the ballot in states via direct democracy measures 18 times.Footnote 13 In Nebraska, an amendment that would protect abortion access lost by slightly less than 2 percentage points — less than 19,000 votes — in 2024.Footnote 14 That same exact election, a pro-life initiative that banned abortion after the first trimester won by only nine points. Where many states are using direct democracy measures to codify abortion policy, the persuasiveness of a framing campaign is incredibly significant. The pro-life movement is hyper-aware of this and has written about the need to persuade the “mushy middle” of the public.Footnote 15
We seek to measure how persuasive the pro-woman frame is to the public in this incredibly potent political moment. Because pro-choice respondents are more likely to align or agree with feminist viewpoints, and the feminist-sounding nature of the pro-woman frame, we expect that:
H3a: Individuals who support abortion rights (based on the pre-treatment measures) will be more likely to agree with the pro-woman frame.
However, because pro-life respondents are likely to agree with either framing as the outcome of the proposed hypothetical legislation is the same: an abortion restriction, we expect that:
H3b: Opponents of abortion (based on the pre-treatment measures) will be statistically indistinguishable in their level of agreement with the frames.
The intended audience of the frame is not the pro-life community; the audience is those who do not identify as pro-life and possibly those who identify as pro-choice. As Tarrow (Reference Tarrow2011) argues, the frames are intended to be more effective with a specific audience; thus, in this case, the expectation is that the deployment of the pro-woman frame will resonate with abortion supporters, rather than influence the views of those already in agreement with the overall goal. Future work should examine the comparative mobilizing power of these frames within the pro-life community; however, that is beyond the scope of this project.
Methods
We fielded this survey from December 4 to 13, 2023, using Lucid Theorem. Lucid takes many steps to ensure high-quality data, including bot blocking, thorough participant screening, and the implementation of fraud prevention and quality detection services.Footnote 16 In a recent study, Coppock and McClellan (Reference Coppock and McClellan2019) assess Lucid’s suitability for online survey experimental research and find it to be an acceptable replacement for Amazon’s MTurk with subjects that are less professionalized and are closer to “US national benchmarks in terms of their demographic, political, and psychological profiles” (12). Despite this, 593 of 2,307 responses were terminated for failing the attention check.Footnote 17 Another 12 respondents were removed from the sample due to other quality issues, resulting in a final sample size of 1,702.Footnote 18 Once low-quality responses were filtered, the remaining sample was 51.6% female, 48.4% male, and had an average age of 47.0 years old. 72.1% of the sample identified as white, 12.1% as African American, 8.4% as Hispanic, and 7.4% as some other race. The average respondent reported slightly less than a 2-year degree. The sample leaned slightly Democratic with 42.1% of respondents identifying as such, 17.65% identifying as independent, and 40.4% identifying as Republicans.
Following much of the framing literature, we employ a survey experiment to evaluate these two major frames around abortion: the fetal personhood and pro-woman frames.Footnote 19 This methodology is, obviously, not the only way to approach this topic, but it does offer some unique advantages. Rather than relying on public opinion or observational data, experiments allow us an important level of control and the ability to isolate different variables of interest. For example, having control over the gender and party cue that accompanies the message is something that we believe is important to understanding how our messages are evaluated. Additionally, through randomization, we are able to account for external factors that may influence attitudes and therefore draw a clearer causal argument, as any differences we detect can be attributed to the different manipulations rather than demographics or other exogenous factors.
To create a pre-treatment measure of support for abortion, we first asked respondents questions measuring their attitudes toward two types of proposed legislation. The first would make it illegal for a woman to get and fill a prescription from an out-of-state provider for medication that would induce an abortion, and the second would make abortion legal in all cases. Respondents were also asked about support for the Equal Rights Amendment, the death penalty, gun laws, and reparations for slavery and racial discrimination to lessen the likelihood of the abortion questions having any priming effects. After this, respondents were randomly assigned to one of eight treatment conditions. All subjects received a combination of the following: a gender cue (man or woman), a partisan source cue (Democrat or Republican), and finally a frame justifying the fictional legislation regulating abortion (pro-woman or fetal rights). This provides us with a fully randomized 2×2×2 design. The text below shows the vignettes for the two frames, with the bolded and bracketed portions showing what was manipulated based on the gender and partisan cues.
Pro-Woman Frame
At a recent press conference, a group of [Democratic/Republican] [men/women] discussed legislation that will come up for vote in the legislature. A Spokes[man/woman] for the group had the following to say; “soon this chamber will be voting on an important bill that ensures women are protected from the dangers associated with abortion. Women across the country have suffered from the physical risks the abortion procedure and also the mental health consequences later in life, we are advancing this legislation to protect against those harms. For too long doctors have presented women with incomplete information regarding the risks of abortion, and the options she may have, this legislation intends to remedy that. We believe that it is a critical role of the state to ensure that women, during a confusing and despairing time, are making the most informed decision possible.”
Fetal Rights Frame
At a recent press conference, a group of [Democratic/Republican] [men/women] discussed legislation focused on abortion that will come up for vote in the legislature. A Spokes[man/woman] for the group had the following to say; “soon this chamber will be voting on an important bill that ensures the sacredness of human life. Abortion, at its core is the taking of a human life and we cannot stand aside and allow this horror to continue. We intend to promote a culture of life in our state because all life deserves to be protected. That is exactly what this legislation intends to do. Every person, born or unborn, has dignity and is entitled to the full protection of the law. We intend to do everything within our power to ensure that this legislation protecting the sanctity of life becomes law in our state.”
A randomization check was performed on standard demographic variables (age, gender, race, education level, and party identification), and there were no statistically significant differences across treatment groups for any of these variables. Given the nature of abortion as a political issue, we also ran randomization checks for pre-treatment abortion attitudes, religion, and identification as a Born Again or Evangelical Christian. There were no significant differences across treatment groups for the pre-treatment attitudes toward abortion or religion. However, there was a statistically significant difference across treatment groups for the variable for Born Again or Evangelical Christians. Adding this variable as a control to all of our statistical tests did not affect the overall results, so we only report the bivariate regressions here.
The language used in the restriction frames was designed to mimic the language used both in legislation itself (Denbow Reference Denbow2015; Reingold Reference Reingold2000; Roberti Reference Roberti2021) and in the broader public debate over limiting abortion rights (Cannold Reference Cannold2002; Rose Reference Rose2011; Saurette and Gordon Reference Saurette and Gordon2016; Siegel Reference Siegel2008; Trumpy Reference Trumpy2014). For example, the homepage of the organization National Right to LifeFootnote 20 contains a passage using both frames; “… [they] promote legislative efforts to provide legal protection to unborn children.” Then goes on to say: “Pro-life education and legislative efforts are making an impact on our culture and in the lives of women facing unexpected pregnancies.”
In the section of its site specific to abortion, Americans United for LifeFootnote 21 also employs both frames. In one section, they say:
In opposing abortion, we acknowledge the humanity of the child in the womb which fuels our effort to protect the pre-born child’s life. From conception, the preborn human being has a unique and complete genetic composition … . As early as five (5) weeks’ gestation, the preborn human being’s heart begins beating. The preborn human being begins to move about in the womb at approximately eight (8) weeks’ gestation.
In a later section, their rhetoric shifts to focusing on women, saying, “In fact, abortion is what harms not only the pre-born child but also the mother, both physically and emotionally. Women who undergo abortions are at increased risk for a range of health complications, including breast cancer, infertility, and psychological trauma.”
As a result, this design provides us with a high level of external validity, since this language mimics what has been used by advocates on both sides, around limiting access to abortion (Reingold et al. Reference Reingold, Kreitzer, Osborn and Swers2015; Roberti Reference Roberti2021). The language used in these statements by pro-life advocates is often intentionally provocative, for example, calling the fetus a “preborn child.” While impossible to capture all of the rhetorical tactics used by advocates, the language in the two experimental conditions is drawn from the language employed on the ground. The value of also testing these two frames against each other is that they broadly represent the shifting evolution of the abortion debate. For example, the final frame presented above, the fetal rights frame, is the classic “protecting the rights of the unborn” argument that has become a hallmark of the abortion debate, while the pro-woman frame is a more recent addition to the ways in which the pro-life movement has couched its arguments. The pro-woman frame holds that abortion is inherently unsafe for the woman, both physically and psychologically, and therefore, regulations seek to protect women.Footnote 22
After receiving the treatments, all subjects were then given a series of questions that served as our dependent variables in the analysis. Respondents were asked the extent to which they agreed or disagreed with the passage, the strength of the argument in the passage, the credibility of those quoted in the passage, the extent to which they believe the proposed legislation would have the effect of protecting women, the extent to which they believe the proposed legislation would have the effect of protecting fetuses, and the believability of the message. The exact wording of the questions can be found in the Appendix; responses were measured on a 5-point Likert scale and then re-scaled from 0 to 1. These different dependent variables allow us to evaluate responses to these frames in several ways, as well as the extent to which gender and partisanship influence those evaluations.
Hypotheses
In light of the previous scholarship and the political environment and past and future rhetoric surrounding reproductive rights, the experiment was designed to test several different hypotheses. Broadly speaking, we are interested in three main topics: how does the pro-woman frame compare against the fetal rights frame, how does the gender and partisanship of the speaker affect their credibility and the strength and believability of their message, and who is receptive to the various frames. To recap, our expectations are as follows:
H1: Regardless of the gender or party cue the pro-woman frame will be viewed as doing more to protect women, respondents will agree more with it than the fetal right frame, and it will be viewed as stronger, more believable than the fetal rights frame.
H2a: Looking at the pro-woman frame, specifically women lawmakers will be seen as more credible, more believable and as sending a stronger message in contrast to their male counterparts.
H2b: Democratic women will be seen as the most credible, most believable, and as sending the strongest message both overall and when looking specifically at the pro-woman frame.
H3a: Individuals who support abortion rights (based on the pre-treatment measures) will be more likely to agree with the pro-woman frame.
H3b: Opponents of abortion (based on the pre-treatment measures) will be statistically indistinguishable in their level of agreement with the frames.
Findings
Effects of the Pro-Woman Frame
Regardless of the gender party cue, H1 expects that the pro-woman frame would be viewed as doing more to protect women, be more persuasive, and be viewed as stronger and more believable than the fetal right frame. Table 1 shows the bivariate regression results for the dependent variables measuring each of these things. While the difference between the two frames is not statistically significant for either the strength of the argument or the believability of the message, the mean level of agreement that the legislation protects women is 11.7% points higher in the pro-woman frame than in the fetal rights frame, and agreement with the overall passage was 9.7% points higher in the pro-woman frame versus the fetal rights frame.Footnote 23
Table 1. Pro-woman frame ratings
Note: *p < 0.05; **p < 0.01; ***p < 0.001.
This provides us with preliminary evidence that respondents are looking at these two frames differently. While they may not have seen them as different in terms of strength or believability, the fact that differences emerged in terms of the belief in whether the legislation protects women and agreement with the passage overall shows that the frames at their most basic level are seen as distinct. Further, it presents a potential explanation for the rhetorical shift in the pro-life movement. Respondents here believed that the legislation discussed would protect women and expressed a higher degree of agreement with the pro-woman frame. As we explain in the discussion, this increased level of agreement is important because it suggests that the shift is not an accident. The evolution and choice of frames is a deliberate one by activists and elites, with one possible explanation for the adoption of the pro-woman frame being that voters are more receptive to this language, even on the highly contentious issue of abortion. By shifting the attention from fetal rights, a contentious topic, to one where there is broader agreement, specifically protecting women and empowerment, this may contribute to the increased levels of agreement that we observed between the frames.
Gender and Partisan Effects Within the Pro-Woman Frame
To assess how gender and party might affect respondents’ perceptions, we conducted regressions with dependent variables measuring credibility of the speaker as well as the believability and strength of message. Each regression included a full three-way interaction between experimental factors. The model was run on the full sample to preserve the random assignment structure and allow for causal interpretation of interaction effects. We had several expectations as to how the gender of the speaker would affect respondents’ perceptions within the pro-woman frame. H2a expected that, within the pro-woman frame, women would be viewed as more credible, more believable, and their message viewed as stronger when compared to their male counterparts. Figure 1 isolates the subset of respondents assigned to the pro-woman frame in order to test this. Each panel of the plot corresponds to one of the three dependent variables and depicts the estimated marginal means and their standard errors. While we did see a statistically significant difference between men and women in the full sample (see Appendix for plot), among those assigned to the pro-woman frame, we find null effects across all dependent variables.
Figure 1. Gender effects within the pro-woman frame.
While we did not have a formal hypothesis about how partisanship would affect respondents’ perceptions, we conducted an exploratory analysis similar to the one we did for gender looking at party evaluations. Figure 2 illustrates that, once again, we do not see any statistically significant differences (see Appendix for plot of comparison by party across frames).
Figure 2. Party effects within the pro-woman frame.
Hypothesis H2b expected that, within the pro-woman frame, Democratic women would be seen as the most credible and that their message would be seen as the most believable and the strongest. Figure 3 again isolates respondents assigned to the pro-woman frame, only this time we compare Democratic women, Democratic men, Republican women, and Republican men. Here we see that it is actually Republican women that are evaluated the highest in all categories. Further, there are statistically significant differences among the categories. The credibility of Republican women is rated 4.7% points higher than the credibility of Republican men. Messages coming from Republican women are significantly more believable than all others (rated 5.1% points higher than Democrat men, 6.0% points higher than Democrat women, and 6.9% points higher than Republican men). Finally, messages coming from Republican women are rated 4.7% points stronger than messages coming from Republican men.
Figure 3. Gender × party effects in the pro-woman frame.
Contrary to our hypothesis, respondents are not convinced by Democratic women using the pro-woman frame. We theorized earlier that the traditional association of Democratic women with abortion policy may lead to a greater credibility on the issue, it may be that the association to pro-choice policy is too strong and, as a result, our respondents were skeptical of the passage when it cut so hard against what most would expect the position of Democratic women to be, regardless of the feminist-sounding language used. Instead, consistently it is Republican women who are viewed more authoritatively than their peers within the pro-woman frame.
Who Is Convinced?
We used our pre-treatment measure of abortion attitudes to create dummy variables for “abortion rights supporter” and “abortion rights opponent.” To be an “abortion rights supporter,” respondents must have either somewhat or strongly opposed both types of legislation proposed in the pre-treatment measures (making abortion pills illegal and making abortion illegal in all cases). To be an “abortion rights opponent,” respondents must have either somewhat or strongly supported both. As a result, this sets the bar for finding effects quite high, this analysis is not looking at individuals who had mixed feelings about abortion in different circumstances, rather, they are individuals who have a consistent feeling one way or the other, and as a result, movement among these individuals is particularly significant to note. We expected abortion right supporters to be more likely to agree with the pro-woman frame versus the fetal rights frame, but we did not expect to find a difference among abortion rights opponent. Table 2 shows the bivariate regression results among subsets of supporters and opponent, and provides support for H3.Footnote 24 Among abortion rights supporters, the mean level of agreement with the pro-woman frame was 19.0% higher than agreement with the fetal rights frame.
Table 2. Framing effect on pre-treatment supporters and opponents
Note: *p < 0.05; **p < 0.01; ***p < 0.001.
In terms of thinking through why the pro-life movement has made a shift to this pro-woman frame in their messaging, these results may be the best explanation. The rhetorical frame is less important to those already on your side, but what is important is how it influences your opponents. We found that among those who were supportive of abortion rights, or, put another way, those that would be thought of as most hostile to the legislation being discussed in the first place, there was a sizeable increase in agreement with the passage when it was presented as pro-woman as opposed to the fetal rights frame. While it is unlikely that overall attitudes toward abortion may shift, by using this frame, the pro-life advocates may be able to avoid mobilizing opposition or even gain tacit support for a bill that one would otherwise oppose. This idea of diminishing the opposition to legislation is important. In recent years, when states have moved to ban or restrict abortion outright, this has been met with protest and political backlash, and has resulted in successful countermovement policy.Footnote 25 Thus, if pro-life activists and leaders can frame their legislation in a way that maintains their existing support while also decreasing the backlash from opponents, they may be more successful with their pro-life legislation.
Discussion and Conclusion
At the start of this paper, we set out to explore a puzzle surrounding the rhetorical shift from the pro-life movement; after decades of the messaging around abortion restrictions focusing on the life of the fetus, there was a shift to adopting a pro-woman frame. This new frame emphasized protecting women from the purported dangers of abortion and sought to empower women to make their own informed decisions (Roberti Reference Roberti2021, Reference Roberti2022). The evidence presented here suggests that one explanation for this rhetorical shift is that the frameworks. While the analysis did not find that one frame was viewed as stronger or more believable than the other, there was a statistically significant difference in terms of agreement, with the pro-woman frame being seen as more agreeable.
This finding regarding agreeableness suggests that the shift in framing could be serving several purposes. As we found in Hypothesis 6, while there was not a difference between how abortion opponents viewed the frames, there was a significant difference (19%) among those who supported abortion finding the pro-woman frame more agreeable. Therefore, this rhetorical shift may serve the purpose of expanding the audience for the legislation. While the fetal rights frame, given its dominance in the abortion discussion for decades, has reached most of those who would be receptive to it, the emergent pro-woman frame may reach a new audience for those seeking to restrict abortion. Consistent with the literature showing that the public is receptive to framing that mirrors their values, a pro-woman frame is feminist-sounding enough to persuade people who are otherwise supportive of abortion.
Even if it does not change the underlying attitudes, an additional benefit of this frame may simply be that when it is deployed around legislation, it does not mobilize pro-choice advocates as much. Given the contentious politics around abortion, something that has only increased since the Dobbs decision, if opponents of abortion can minimize the public backlash to their legislation, they may be more able to get them through a legislature that is ever attentive to possible electoral consequences. Alternatively, a goal of this frame may also be to simply muddy the waters surrounding the abortion debate by reaching and drawing in an audience that would otherwise be supportive of abortion rights. Future research may explore this theory of de-mobilization/muddying. However, given the limitations of the experimental design and data collected, this project is unable to speak to that empirically.
This is also supported by the finding that the pro-woman frame was seen as protecting women more than the fetal rights frame. While this is an unsurprising result, it again can help explain the rhetorical shift. After decades of messaging around fetal rights, public opinion has only become more supportive of legal abortion. While obviously individual attitudes can and do shift, at this point in time, being exposed to the fetal rights frame another time is unlikely to move the opinions of those who were not persuaded already. The shift to a pro-woman frame, thus, can capture a new audience of people who are concerned about the welfare of women.Footnote 26 By speaking to those concerns, the pro-life movement can possibly capture a new segment of the population who were not persuaded by their focus on the fetus. The pro-life movement could also use this frame tactically to counteract the narrative that the movement is hostile toward women.
This shift also may serve the purpose of opening up individuals to a policy proposal that restricts abortion access, such as recent state proposals to enact 15-week bans, or less, or efforts to change the labeling around mifepristone, because they seem like a “reasonable middle ground.” These seemingly incremental shifts in policy, to which we argue the pro-woman framing increases responsiveness, are actually quite restrictive and move toward the overall goal of the pro-life movement. Thus, individuals may retain their pro-choice position while simultaneously expressing support for a pro-life policy position, even if they do not necessarily see it as such. Further research is needed into these rhetorical shifts and the policy consequences that may flow from them. Pro-choice policymakers and activists should also take note of this dynamic. It may not be enough to assume that a pro-choice majority in a state will automatically oppose legislation that will restrict abortion. Rather, when encountering these frames, these actors must take affirmative steps to challenge them due to their ability to reach and sway, otherwise, pro-choice individuals.
In terms of the source cues of gender and party, the results are equally interesting but slightly more muddled. Within the pro-woman frame, while there were no significant differences in perceived credibility, believability, or message strength when comparing women to men or Republicans to Democrats on average, we do see differences when considering the interaction between party and gender. In terms of the credibility of the speaker, Republican women were viewed as the most credible, although the difference was only statistically significant when compared to Republican men. In terms of believability, Republican women were significantly more believable than all others. Finally, in terms of strength of message, we again see Republican women at the top, although similar to credibility, the difference is only significant when comparing to Republican men. This finding is particularly interesting as Republican women, though underrepresented in the Republican party, are being deployed as the messengers of a woman’s issue with gendered messaging. By doing this, Republicans are making claims about gendered representation: who is authentically representing women and what does that look like? This compels gender and politics scholars to continue to confront the intersection of substantive and descriptive representation, and how that has been used by Republican women. We take these results to suggest a relationship that deserves more attention and research in future work on electoral politics, especially given the increased representation of Republican women as the face of pro-life legislation.
Theoretically, future work should also focus on the way in which these frames are deployed in the post-Dobbs landscape. There is evidence of this pro-woman rhetoric emerging in the debate around trans rights, so future work should be attentive to the deployment of this frame outside the abortion discussion and, in turn, evaluate and compare its effectiveness in those places as well. In February 2025, Iowa Governor Kim Reynolds cited the protection of women and girls when signing a law that would strip gender identity from the Iowa Civil Rights Act.Footnote 27 In March 2025, North Carolina State Senators Vickie Sawyer and Brad Overcash co-sponsored a bill entitled the “Women’s Safety and Protection Act,” which bans the use of restrooms, changing rooms, and other accommodations that comport with a trans person’s gender identity under the reasoning that this would protect women from harassment, sexual assault, and other forms of violence.Footnote 28 Furthermore, in Congress, House Bill 28 would disallow transwomen athletes from competing in women’s sports, under Title IX rules, noting also that the Government Accountability Office must make a report on all “… the benefits for women or girls in single-sex sports that would be lost as a result of male participation … [as well as] negative psychological, developmental, participatory, and sociological effects of male participation on girls.”Footnote 29 As it is clear from just a few out of many examples, the pro-woman rhetoric has potential persuasiveness in areas beyond abortion.
Though our study focused solely on the US, the pro-woman rhetoric has also been found beyond US borders. Various studies have identified the existence of pro-woman framing in pro-life movements in Germany (Ferree Reference Ferree2003), the UK (Lowe Reference Lowe2019; Lowe and Page Reference Lowe and Page2022), and Canada (Saurette and Gordon Reference Saurette and Gordon2016), to name a few. This indicates the prospective resonance of this frame in governments and societies that have historical and contemporary legal rights for women, in which this feminist-adjacent rhetoric can be successfully employed. As the US continues to backslide on abortion, the potential ripple effect in pro-life tactics across other democratic countries is a worthwhile area of study.
Ultimately, as recent elections and ballot initiatives have shown us, the broader electorate does not support extreme pro-life policies. If Republicans want to avoid the public backlash at the ballot box, they have two choices. One is legislative: moderate abortion policies, or refrain from taking new legislative action. This, however, has proven not to be the path taken by legislators, as according to the Guttmacher Institute, state lawmakers introduced 675 regulatory abortion laws in 2023 (Forouzan and Guarnieri Reference Forouzan and Guarnieri2023) and 508 in 2024 (Forouzan, Guarnieri, and Fairbanks 2024). The other is rhetorical: win the narrative on abortion by deploying a persuasive frame that sounds less extreme, or reasonable, and maybe even feminist.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/S1743923X25100561.
Acknowledgments
The authors would like to thank the editors and anonymous reviewers for their comments on this paper, as well as San Francisco State University, Francis Marion University, and Rutgers, The State University of New Jersey for their support of this research.
Open access
