Despite evidence that attitudes are shifting towards greater support for gender equality in the public sphere, gender relations appear stubbornly inegalitarian in the private sphere of intimate relationships between women and men (Donnelly et al., 2016; Overall & Hammond, 2018; Pepin & Cotter, 2018; Scarborough et al., 2019). Gender stereotypes and norms about traditional gender roles persist (Eagly et al., 2020; Haines et al., 2016), and are especially stable in the arena of courtship and romance between women and men (Cameron & Curry, 2020). To illustrate, heteronormative dating scripts express that men should ask women out on dates, pay for dates, take the lead in initiating romantic and sexual activities, buy an engagement ring, make the marriage proposal, and that women should take their husband’s surname after marriage (Lever et al., 2015; Paynter & Leaper, 2016; Robnett & Leaper, 2013; Sprecher et al., 2021; Wu et al., 2022).

The current research considered a range of predictors of women’s endorsement of heteronormative dating scripts. First, we investigated known predictors of these scripts: sexism and feminist identity (reviewed below). Extending prior research, we investigated women’s preferences for partners and/or relationships as simultaneous predictors of their endorsement of heteronormative dating scripts. Specifically, we examined their preference for dominant men, their preference for long-term relationships, and preference against short-term relationships. We conducted a novel test of the extent to which women’s sexism and feminist identity predicted their endorsement of heteronormative dating scripts over and above indices of their personal romantic preferences. Thus, in the following sections we considered whether women’s endorsement of heteronormative dating scripts is better explained by their acceptance of gender inequality, or whether other priorities related to partner and relationship preferences matter regardless of their attitudes towards gender roles.

Beliefs About Gender Inequality Predict Heteronormative Dating Scripts

An individual’s personal norms, expectations, and behaviour in their romantic lives is often difficult for people to conceptualise as a form of societal gender inequality (e.g., Lamont, 2014; 2015; Robnett & Leaper, 2013). However, researchers consider conventional attitudes about gender roles in romantic relationships to be problematic if they perpetuate gender inequality and reinforce men’s higher status and control over women (Eaton & Rose, 2011; Glick & Fiske, 2001; Hammond & Overall, 2015, 2017; Lee et al., 2010; Sassler & Miller, 2011). To illustrate, one study has found that women’s implicit chivalric romance fantasies—such as associating male partners with “Prince Charming”—were associated with a lower personal interest in power and leadership (Rudman & Heppen, 2003). Similarly, researchers have investigated people’s enactment of heteronormative dating scripts. Men who ask and pay for dates with women are more likely to expect sex from their dating partners (Emmers-Sommer et al., 2010), and correspondingly, women report feeling greater pressure to engage in sexual activity when men pay for dates (Lever et al., 2015). Indeed, holding more traditional beliefs about gendered roles in dating contexts is linked with a host of negative outcomes for women. For instance, people who hold sexist beliefs assign greater blame to the victims of acquaintance rape to the extent they are judged as being “unladylike” and “improper” (Abrams et al., 2003). Accordingly, several researchers argue that heteronormative dating scripts are based in attitudes that uphold gender inequality.

One set of attitudes underlying the endorsement of heteronormative dating scripts is ambivalent sexism (Paynter & Leaper, 2016). Ambivalent sexism theory (Glick & Fiske, 1996, 2001) holds that sexist attitudes have a hostile form and a benevolent form. Hostile sexism is overt antipathy and antagonism towards women, particularly when they do not fulfil conventional gender roles. Benevolent sexism is superficially positive and holds that men should put women on a pedestal, and chivalrously protect and provide for them. Both benevolent sexism and hostile sexism are closely linked to norms around how men and women should behave in romantic contexts with one another, including placing expectations on men to take an active, leading role, while women are expected to be passive, receptive, and nurturing (Hammond & Overall, 2017). Indeed, studies have shown that both hostile and benevolent sexism are associated with greater endorsement of a range of gendered scripts for dating and marriage between women and men (Bermúdez et al., 2015; Cameron & Curry, 2020; Paynter & Leaper, 2016; Robnett & Leaper, 2013; Viki et al., 2003). In addition, people who strongly endorse benevolent and hostile sexism respond more negatively to counter-stereotypic dating behaviour (McCarty & Kelly, 2015), and tend to prefer more traditional partners (Chen et al., 2009; Lee et al., 2010; Travaglia et al., 2009). Therefore, if sexism is embedded in these heteronormative dating scripts, then the implications of people endorsing these scripts might not be limited to romantic relationships, but instead might extend beyond the dyad to reinforce broader gender inequalities.

As well as sexism, women’s self-identification as a feminist may also be relevant to the endorsement of heteronormative dating scripts. Feminist identity is distinct from sexist attitudes in so far as it reflects the willingness to self-identify as a feminist both privately and publicly, to support the feminist movement, and to see feminism as important to one’s identity (Szymanski, 2004). Despite being correlated, many women endorse beliefs in gender equality but do not self-identify as a feminist (Fitz et al., 2012; Redford et al., 2018; Robnett et al., 2012; Weis et al., 2018). Therefore, feminist identity should be considered in addition to sexism, particularly since feminism often critiques conventional gender roles. Indeed, past work shows that women who identify as feminists endorse more egalitarian expectations for dating and for marriage (Wu et al., 2022; Yoder et al., 2007), reject sexual double-standards more strongly (Bay-Cheng & Zucker, 2007), and report enacting fewer gendered romance behaviours, including being more likely to ask and pay for a date with men (Rickard, 1989). In contrast, those who rate low on feminist identity believe that feminism is in conflict with romance (Rudman & Fairchild, 2007). However, one study that examined the link between feminist identity and the endorsement of heteronormative dating scripts (Paynter & Leaper, 2016) found no evidence for an association among women undergraduates (although there was a negative correlation among men). Therefore, in the current research we included a measure of self-identification as a feminist in addition to sexist attitudes. We expected that women who identified as feminists would be more likely to reject heteronormative dating scripts.

In sum, this research highlights that sexism (both hostile and benevolent) and women’s self-identification as a feminist should predict their endorsement of heteronormative dating scripts. Linking attitudes that reinforce gender inequalities with heteronormative dating scripts would suggest that these scripts may be expressions of beliefs that support men’s higher status position in relationships and society. However, there is a lack of research on women’s personal preferences for partners and relationships that could more directly explain the attraction of these scripts. As we argue below, heteronormative dating scripts may be appealing to women because they signal characteristics about the men who enact them that are aligned with their personal preferences for partners and relationships aside from an adherence to conventional gender roles.

Partner and Relationship Preferences as Predictors of Heteronormative Dating Scripts

Beyond adherence to conventional gender roles, women’s endorsement of heteronormative dating scripts may partly be based in their individual preferences for male partners to be high on dominance. In heteronormative dating scripts, men take the lead and play a more dominant and assertive role while women take a more passive and reactive role (e.g., Cameron and Curry, 2020; Lamont, 2014, 2015; McCarty and Kelly, 2015). If heteronormative dating scripts signal men’s dominance (e.g., Paynter and Leaper, 2016), then women’s personal preference for these characteristics should predict a greater endorsement of these scripts, regardless of whether they endorse sexist attitudes or reject a feminist identity.

Women’s endorsement of heteronormative dating scripts may also be driven by their preference for a committed relationship with an invested partner. Specifically, people vary in desired relationship commitment, with evolutionary perspectives stating that people are flexible in the extent to which they are motivated towards seeking long-term and/or short-term relationships and seek cues in dating contexts about the compatibility of potential partners’ motives for long- or short-term relationships (Jackson & Kirkpatrick, 2007). Research suggests that women view heteronormative dating scripts, such as men paying on dates and buying an engagement ring, as signals of men’s willingness to invest in their partners (Lamont, 2014; Lever et al., 2015). Thus, if heteronormative dating scripts signal men’s potential investment in a relationship, they should appeal to women who desire a committed relationship with an invested male partner, independently of their sexist attitudes and feminist identity. Accordingly, the endorsement of heteronormative dating scripts should be stronger among women with a greater preference for a long-term relationship and a preference against short-term relationships, perhaps over and above their sexist attitudes and feminist identity.

Critically, even the perspective that women’s endorsement of heteronormative dating scripts is partly driven by personal preferences acknowledges that those preferences occur in the context of societal gender inequality. Indeed, women’s preferences for traditional relationship roles emerge in response to the existing contexts of inequality in their society (see Eagly et al., 2009). For example, when societal inequalities limit women’s ability to obtain resources for themselves, male partners who will be assertive, career-driven, and devoted providers become relatively more appealing (e.g., Eastwick et al., 2006; Sibley and Overall, 2011), showcasing the interplay between romantic “preferences” and societal level inequalities. This pattern is particularly apparent when examining the preferences of women who endorse sexism (see Hammond and Overall, 2017). Women who more strongly endorse sexist attitudes have stronger preferences for older partners with good financial prospects (Eastwick et al., 2006; Lee et al., 2010; Sibley & Overall, 2011; Travaglia et al., 2009), who are assertive and dominant (Chen et al., 2009; Hall & Canterberry, 2011; Lee et al., 2010), and who are warm, devoted, and reliable providers (Cross et al., 2016; Hopkins-Doyle et al., 2019). Thus, women’s preferences for dominant and invested partners may help to explain the expected links between women’s sexist attitudes and heteronormative dating scripts. Accordingly, in the current research we test whether the expected links between women’s sexist attitudes and endorsement of heteronormative dating scripts would be statistically mediated by women’s greater preferences for dominant male partners or greater desires for invested male partners.

Current Research

Across two studies and three samples, we investigated the extent to which women’s (1) sexist attitudes, (2) feminist identity, (3) preference for a dominant partner, and (4) the preference for short- and long-term relationships simultaneously predicted their endorsement of heteronormative dating scripts. Given that we were looking at short- and long-term relationship preferences, we only examined women who were not currently in a relationship. In line with previous research, we hypothesised that women higher in hostile sexism, higher in benevolent sexism, and lower in feminist identity would have a stronger endorsement of heteronormative dating scripts. However, we extended prior research by testing these potential associations simultaneously alongside women’s greater preference for a dominant partner, preference for a long-term relationship, and preference against short-term relationships. Study 2 was a pre-registered replication that also tested the extent to which women’s relationship preferences mediated the link between women’s endorsement of sexism and their endorsement of heteronormative dating scripts.

Study 1

Method

Participants and Procedure

Data for Study 1 were composed of two independent samples. Sample 1a was collected through snowball sampling on social media (n = 112), and Sample 1b (n = 157) was collected through paid Facebook advertising. Participants who consented to the study completed an online questionnaire hosted on Qualtrics. Participants in Sample 1a were aged 19–62 years (M = 29.88, SD = 11.25) and participants in Sample 1b were aged 18–81 years (M = 50.33, SD = 16.13; see Table 1 for full sociodemographic information). Participants were recruited as part of a larger study open to women living in Australia aged 18 years and over, but only those who were not currently in a relationship completed all the measures relevant to this study. We excluded 11 lesbian women (five from Sample 1a and six from Sample 1b) because our research questions concerned relationships between women and men. We also excluded eight people from Sample 1a: One participant did not complete the required scales and a further seven cases were straight-line responses. All participants were eligible to enter a prize draw for one of five AU$50 retail vouchers. Funding for the study was provided by the School of Psychology, Deakin University.  Ethical approval for the study was obtained from the Human Ethics Advisory Group – Health, Deakin University (approval number HEAG-H 108_2021).

Table 1 Sociodemographic characteristics of Sample 1a (N = 112) and Sample 1b (N = 157)
Full size table

Materials

Participants completed the following measures in the context of a larger questionnaire that also included measures of other constructs, such as well-being. We report measures relevant to this study here.

Endorsement of Heteronormative Dating Scripts

Participants completed eight items that were averaged to assess higher endorsement of heteronormative dating scripts (1 = strongly disagree to 7 = strongly agree). Five items were adapted from the Heterosexual Dating Double-Standards Scale (Paynter & Leaper, 2016), re-worded from items about male and female undergraduates to refer to men and women in relationships more generally (e.g., “Men should be the ones who ask women out for a first date”). We added three items related to marriage, guided by the research on marriage conventions (Monger, 2013; Robnett & Leaper, 2013): “When a couple get engaged, the man should buy the ring for his partner”, “Men should ask a woman’s family for permission before proposing marriage”, “When women get married, they should be walked down the aisle by their father”.

Hostile and Benevolent Sexism

Participants completed the 12-item short-form of the Ambivalent Sexism Inventory (Glick & Fiske, 1996). Six items were averaged to index endorsement of hostile sexism (e.g., “Women seek to gain power by getting control over men”) and six items were averaged to index endorsement of benevolent sexism (e.g., “Women should be cherished and protected by men”; 0 = disagree strongly to 5 = agree strongly).

Feminist Identity

The Self-Identification as a Feminist scale (Szymanski, 2004) measured feminist identity. We averaged participants’ agreement with four items (e.g., “I consider myself a feminist”; 1 = strongly disagree to 5 = strongly agree).

Preference for a Dominant Partner

Following Snyder et al. (2011), we developed a list of 12 adjectives relating to dominance. We asked participants “How attractive do you find the following characteristics in a partner?” on a five-point scale (1 = not attractive at all to 5 = extremely attractive). The 12 characteristics averaged together were: dominant, formidable, commanding, protective, alpha male, authoritative, masculine, confident, assertive, strong, powerful, and leader.

Long-Term and Short-Term Relationship Orientations

The Sociosexual Orientation Inventory (Jackson & Kirkpatrick, 2007) measured participants’ preference for long-term and short-term relationships. Seven items were averaged to measure long-term orientation (e.g., “I hope to have a romantic relationship that lasts the rest of my life”) and ten items measured short-term orientation (e.g., “I can easily imagine myself being comfortable and enjoying “casual” sex with different partners”; 1 = strongly disagree to 7 = strongly agree).

Sociodemographic Variables

Participants provided a range of sociodemographic information including age, gender, relationship status, location, income, education, employment, country of birth, and sexual orientation.

Results

Analyses were conducted in IBM SPSS Statistics Version 27. A small number of missing values that were missing completely at random were replaced using expectation maximisation (Sample 1a: two values, Sample 1b: six values). Descriptive statistics, reliabilities, and zero-order correlations for both samples are displayed in Table 2. Correlations suggested that women’s endorsement of heteronormative dating scripts was related to greater endorsement of both hostile sexism and benevolent sexism, lower feminist identity, greater preference for a dominant partner, as well as with a lower preference for short-term relationships. Accordingly, we utilised hierarchical regressions to test our hypotheses concerning the unique predictors of women’s endorsement of heteronormative dating scripts after accounting for the variables that have previously been investigated and adjusting for sociodemographic variables. Data screening for both samples revealed minimal presence of outliers, and no concerns around multicollinearity, normality, linearity, nor heteroscedasticity of residuals.

Table 2 Descriptive statistics, Cronbach’s alphas, and correlations for key study variables for Samples 1a and 1b (Study 1)
Full size table

Women’s endorsement of heteronormative dating scripts was regressed on hostile sexism and benevolent sexism (Step 1), feminist identity (Step 2), preference for a dominant male partner (Step 3), and long-term and short-term orientation (Step 4), alongside sociodemographic covariates (i.e., age, location, education, employment status, income, country of birth, and sexual orientation). Results from the Sample 1a model and the Sample 1b model are presented in Table 3. In the first step, greater hostile sexism and greater benevolent sexism were strong predictors of greater endorsement of heteronormative dating scripts. In the second step, greater feminist identity predicted lower endorsement of heteronormative dating scripts, and the effect of hostile sexism dropped to a moderate effect size that was not statistically significant in Sample 1a but was significant in Sample 1b. In the third step, greater preference for a dominant partner predicted greater endorsement of heteronormative dating scripts in both samples, and the effect of other predictors remained unchanged. In the final step, a greater preference for a dominant partner (in both samples, Sample 1a: β = 0.18, p = .014, 95% CI [0.07, 0.57]; Sample 1b: β = 0.24, p = .001, 95% CI [0.18, 0.72]) and greater long-term orientation (in Sample 1a only; β = 0.24, p = .001, 95% CI [0.10, 0.41]) predicted greater endorsement of heteronormative dating scripts. However, women’s endorsement of benevolent sexism remained a strong and significant predictor of endorsement of heteronormative dating scripts (Sample 1a: β = 0.26, p = .004, 95% CI [0.10, 0.52]; Sample 1b: β = 0.37, p < .001, 95% CI [0.26, 0.63]), while hostile sexism remained significant in Sample 1b only (β = 0.17, p = .024, 95% CI [0.03, 0.41]). Feminist identity also remained significant in both samples (Sample 1a: β = − 0.32, p = .003, 95% CI [-0.61, − 0.03]; Sample 1b: β = − 0.15, p = .031, 95% CI [-0.35, − 0.02]). The only sociodemographic variable that was significant in the final model was education: women who were university-educated reported a greater endorsement of heteronormative dating scripts in Sample 1a only (β = 0.14, p = .043, 95% CI [0.01, 0.63]).

Table 3 Standardised betas in hierarchical regressions predicting endorsement of heteronormative dating scripts in Samples 1a and 1b (Study 1)
Full size table

Discussion

We examined several theorised predictors of the endorsement of heteronormative dating scripts in two samples of single women. As hypothesised, women’s greater benevolent sexism and lower feminist identity simultaneously predicted greater endorsement of heteronormative dating scripts in both samples. Thus, women’s expectations for romance to follow gendered scripts (e.g., believing men should open doors for women and pay on dates) are closely related to their sexist attitudes and rejection of feminism. Nonetheless, they were also related to a greater preference for a dominant partner in both samples, over and above their sexist attitudes and rejection of feminism. In addition, some results were mixed and needed clarification: Women’s greater endorsement of heteronormative dating scripts was predicted by greater hostile sexism in Sample 1a and by the preference for a long-term relationship in Sample 1b. We sought to clarify these associations in Study 2 in a pre-registered replication and consider potential explanations for the link between sexist attitudes and the endorsement of heteronormative dating scripts.

Study 2

Study 2 was a pre-registered replication of Study 1. We hypothesised that women’s endorsement of heteronormative dating scripts would be predicted by their hostile and benevolent sexist attitudes, lower feminist identity, preference for a dominant partner, preference for a long-term relationship, and preference against a short-term relationship. Study 2 also extended Study 1 by considering possible mechanisms for the link between women’s sexist attitudes and endorsement of heteronormative dating scripts. Women who endorse hostile sexism and benevolent sexism may prefer heteronormative dating scripts because they represent desired characteristics in male romantic partners. Specifically, past research suggests that women who endorse benevolent sexism prefer romantic partners who are highly devoted to their relationships and internalise “protector/provider” roles (see Cross et al., 2016; Gul and Kupfer, 2019; Hopkins-Doyle et al., 2019). Thus, women who more strongly endorse benevolent sexism might endorse heteronormative dating scripts because they signal male partners’ willingness to invest in traditional relationship roles. Similarly, women who endorse hostile sexism tend to prioritise male power (Glick & Fiske, 1996), and thus may prefer romantic partners who are assertive and dominant (e.g., Chen et al., 2009; Lee et al., 2010), and in turn more strongly endorse heteronormative dating scripts that prescribe male dominance in relationships. We tested our pre-registered prediction in a fully-saturated path model in which women’s hostile sexism and benevolent sexism simultaneously predicted the mediators (1) the desire for partner investment, and (2) the preference for a dominant partner, and in turn, their endorsement of heteronormative dating scripts.

Method

Participants and Procedure

Study 2 was part of a larger study that recruited people using paid Facebook advertising. People of any gender living in Australia aged 18 years and over were eligible to participate. However, only women who were not currently in a relationship completed the measures described here. The initial sample included 201 single women, of which nine lesbian women and three participants who did not complete the relevant scales were excluded. This left a sample of 189 women aged 18–77 years (M = 45.29, SD = 15.38; see Table 4 for full sociodemographic information). The procedure for this study was similar to Study 1, Sample 1b. The study was advertised as being about social behaviour, relationships, and gender roles, and the prize draw offered one of two AU$50 retail vouchers. Funding for the study was provided by the School of Psychology, Deakin University, and ethical approval for the study was obtained from the Human Ethics Advisory Group – Health, Deakin University (approval number HEAG-H 113_2022). The method, hypotheses, and analytic plan was pre-registered on the Open Science Framework (https://osf.io/sh8xe/?view_only=5283fbea26c1404db2a71684083e2fd3).

Table 4 Sociodemographic characteristics of participants in study 2 (N = 189)
Full size table

Materials

Study 2 included all the measures described in Study 1 with two modifications. First, the heteronormative dating scripts scale was measured on a five-point scale (1 = strongly disagree to 5 = strongly agree). Second, we added a measure of participants’ desire for invested partners.

Desire for Invested Partner

We adapted three items from the perceived partner reliability index (Cross et al., 2016). Participants answered the extent to which they agreed or disagreed that the following qualities were important in a future partner: “My partner is the kind of person who will stick by me through good times and bad”, “My partner is the kind of person who will always put our relationship first”, and “My partner is the kind of person who will make sacrifices to take care of me and our relationship” (1 = strongly disagree to 7 = strongly agree).

Results

Data screening revealed no outliers, and no concerns around multicollinearity, normality, linearity, nor heteroscedasticity of residuals. One participant missed one item on the benevolent sexism scale, which was replaced using expectation maximisation. Descriptive statistics, reliabilities, and zero-order correlations for the measures in Study 2 are displayed in Table 5. As shown in Table 5, correlations mirrored those found in Study 1; women’s endorsement of heteronormative dating scripts was related to greater endorsement of hostile sexism and benevolent sexism, lower feminist identity, a greater preference for a dominant partner and long-term relationships, and a lower preference for short-term relationships. Further, greater benevolent sexism and greater endorsement of heteronormative dating scripts was associated with women’s greater desire for an invested partner.

Table 5 Descriptive statistics, Cronbach’s alphas, and correlations for key study variables in study 2
Full size table

We first conducted a hierarchical regression that replicated the model in Study 1, with the addition of desire for an invested partner in Step 5. Results are presented in Table 6. Replicating the results from Study 1, women’s hostile sexism and benevolent sexism simultaneously predicted a greater endorsement of heteronormative dating scripts in the first step. In subsequent steps, only benevolent sexism, and not hostile sexism, remained a significant and strong predictor of heteronormative dating scripts. With all variables entered into the model (Step 5), greater benevolent sexism (β = 0.35, p < .001, 95% CI [0.19, 0.41]), lower feminist identity (β = − 0.22, p = .002, 95% CI [-0.27, − 0.06]), greater preference for a dominant partner (β = 0.21, p = .001, 95% CI [0.11, 0.43]), and lower short-term orientation (β = − 0.18, p = .006, 95% CI [-0.15, − 0.03]) significantly predicted greater endorsement of heteronormative dating scripts. There was no evidence that sociodemographic variables were significant in the final model except that women who were non-heterosexual had a relatively lower endorsement of heteronormative dating scripts (β = − 0.17, p = .004, 95% CI [-0.50, − 0.10]). In sum, the majority of variance in heteronormative dating scripts was accounted for by women’s endorsement of benevolent sexism.

Table 6 Standardised betas in a hierarchical regression predicting endorsement of heteronormative dating scripts in study 2
Full size table

We next conducted the pre-registered path model in SPSS AMOS Version 27 (see OSF). Women’s hostile sexism and benevolent sexism were modelled as predictors of the endorsement of heteronormative dating scripts, mediated by the preference for a dominant partner and the desire for an invested partner. Results are displayed in Fig. 1. First, there was no evidence that women’s hostile sexism predicted a preference for a dominant partner (see Path a, Fig. 1), but greater preference for partner dominance predicted greater endorsement of heteronormative dating scripts (Path g), and women’s greater hostile sexism was directly associated with greater endorsement of heteronormative dating scripts (Path e). Second, as hypothesised, women’s benevolent sexism predicted a greater desire for an invested partner (Path d). However, there was no evidence that desire for partner investment predicted a greater preference for heteronormative dating scripts (Path h). Instead, women’s benevolent sexism was directly associated with heteronormative dating scripts (Path f). Finally, an unexpected pattern of effects emerged: Benevolent sexism predicted a greater preference for a dominant partner (Path b), and in turn, preference for partner’s dominance predicted greater endorsement of heteronormative dating scripts (Path g). The indirect associations are displayed in Table 7: there was a significant indirect effect between benevolent sexism and the endorsement of heteronormative dating scripts via the preference for a dominant partner. Thus, results were consistent with partial mediation: the link between women’s benevolent sexism and heteronormative dating scripts was partly accounted for by a preference for partners to be assertive and dominant.

Fig. 1
figure 1

Path model examining predictors of the endorsement of heteronormative dating scripts. Note: Standardised coefficients are presented, with significant values in bold

Full size image
Table 7 Specific indirect effects, total indirect effects, and total effects of the path model
Full size table

Discussion

Consistent with Study 1, women’s endorsement of heteronormative dating scripts was predicted by greater benevolent sexism, lower feminist identity, and a greater preference for a dominant partner, with benevolent sexism emerging as the strongest predictor. However, Study 2 revealed that the endorsement of heteronormative dating scripts was also lower among those with a preference for short-term relationships. There was no evidence that the endorsement of heteronormative dating scripts was predicted by hostile sexism, a preference for a long-term relationship or a desire for an invested partner when all variables were considered together. Overall, while benevolent sexism and a rejection of feminism predicted the endorsement of heteronormative dating scripts, women’s personal preference for a dominant partner also emerged as a unique predictor over and above these other variables. Path analysis did not support our hypothesised mediators of the links between women’s sexism and their endorsement of heteronormative dating scripts. We did not find evidence that a greater preference for a dominant partner explained the links between women’s hostile sexism and greater endorsement of heteronormative dating scripts. Moreover, although women’s benevolent sexism was associated with a heightened desire for an invested partner, as expected (e.g., Cross et al., 2016; Gul and Kupfer, 2019; Hopkins-Doyle et al., 2019), there was no evidence that this link explained the greater endorsement of heteronormative dating scripts. Finally, an unexpected pattern of effects emerged: Benevolent sexism predicted a greater preference for a dominant partner, and this preference for a dominant partner in turn predicted greater endorsement of heteronormative dating scripts.

Integrative Data Analysis

Finally, we conducted an integrative data analysis (Curran & Hussong, 2009) which pooled data from all three independent samples in a hierarchical regression analysis, statistically adjusting for the sample source and the sociodemographic variables. Results from the integrative data analysis are presented in Table 8. Women’s endorsement of heteronormative dating scripts was robustly predicted by greater hostile sexism (β = 0.14, p = .001, 95% CI [0.07, 0.26]), greater benevolent sexism (β = 0.31, p < .001, 95% CI [0.28, 0.46]), a lower feminist identity (β = − 0.16, p < .001, 95% CI [-0.27, − 0.09]), a greater preference for a dominant partner (β = 0.21, p < .001, 95% CI [0.26, 0.51]), and a lower orientation to short-term relationships (β = − 0.10, p = .003, 95% CI [-0.12, − 0.03]). Importantly, however, the majority of variance in heteronormative dating scripts was accounted for by women’s endorsement of sexist attitudes. Once again, women who did not identify as heterosexual had a lower endorsement of heteronormative dating scripts (β = − 0.12, p = .006, 95% CI [-0.42, − 0.07]).

Table 8 Standardised betas in a hierarchical regression predicting endorsement of heteronormative dating scripts in an integrative data analysis of studies 1 and 2
Full size table

General Discussion

We examined known predictors of single women’s endorsement of heteronormative dating scripts, including hostile sexism, benevolent sexism, and feminist identity, as well as some novel predictors relevant to women’s partner and relationship preferences. In line with predictions, results from three samples indicated that greater hostile sexism, greater benevolent sexism, lower feminist identity, a greater preference for a dominant partner, and a lower orientation to short-term relationships predicted stronger endorsement of heteronormative dating scripts. However, contrary to expectations, there was no evidence that long-term relationship orientation predicted the endorsement of heteronormative dating scripts. Pre-registered predictions testing potential mediators of the links between women’s sexism and their endorsement of heteronormative dating scripts were not supported. Instead, one unexpected effect emerged suggesting that a greater preference for a dominant partner underlies the link between women’s benevolent sexism and heteronormative dating scripts.

While both hostile and benevolent sexism predicted the endorsement of heteronormative dating scripts, benevolent sexism was the most consistent and robust predictor, even when analysed along with partner and relationship preferences that may otherwise predict the endorsement of these scripts. The finding that benevolent sexism accounted for more of the variance in the endorsement of heteronormative dating scripts compared to all other predictors suggests that sexism underpins many of these scripts, such as that men should initiate dates and propose marriage, and that women should take their husband’s surname. These findings extend foundational research suggesting that women higher on sexist attitudes are more likely to endorse heteronormative dating scripts (e.g., Bermúdez et al., 2015; Cameron and Curry, 2020; Paynter and Leaper, 2016; Robnett and Leaper, 2013; Viki et al., 2003), by ruling out alternative explanations that these preferences are simply due to personal-level relationship preferences.

The union between romance and sexism is consistent with principles of ambivalent sexism theory that benevolent sexism prescribes differential gender roles both within and outside of relationships. This includes beliefs that men ought to be chivalrous protectors and providers who should cherish and make sacrifices for their female partners, and men taking higher-status agentic roles and women taking lower-status passive roles (see Glick and Fiske, 1996; Hammond and Overall, 2017). These prescriptive gender roles are theorised to be especially important in facilitating romantic connection between women and men by emphasising the complementarity of such roles (Glick & Fiske, 1996; Lamont, 2014). Indeed, our findings support a key tenet of ambivalent sexism theory that benevolent sexism is tightly linked to common beliefs about heterosexual romance which masks the danger of benevolent sexism (Glick & Fiske, 1996): women’s benevolent sexism, which is relatively more difficult to challenge as “sexist” (e.g., Gul and Kupfer, 2019; Hopkins-Doyle et al., 2019), was the strongest predictor of heteronormative dating scripts, providing additional evidence that the functions of benevolent sexism emerge in contexts that appear subjectively positive, including romance, dating, and courtship. Nonetheless, a recurring finding in ambivalent sexism research is that the relational benefits of benevolent sexism are typically in its appearance. In ongoing relationships, women—but not men—who endorse benevolent sexism are relatively more vulnerable to declines in relationship satisfaction (Hammond & Overall, 2013) and have relatively fewer opportunities outside their relationships due to conventional gender roles emphasising women’s strengths specifically within domestic domains (Glick & Fiske, 1996).

The current research also demonstrates the importance of women’s feminist identity in understanding why women reject heteronormative dating scripts. Our results revealed that identifying as a feminist was consistently associated with lower endorsement of heteronormative dating scripts, even after accounting for women’s sexism and their partner and relationship preferences. Contrasting women’s feminism with women’s sexism is especially important given that feminist identification and gender egalitarian beliefs (i.e., lower endorsement of sexism) have been found to be distinct factors that challenge gendered power structures and inequalities (Rickard, 1989; Wu et al., 2022; Yoder et al., 2007). The current results support that feminist identification encompasses more than egalitarian gender beliefs and the rejection of male dominance (Fitz et al., 2012; Redford et al., 2018; Robnett et al., 2012; Weis et al., 2018), and is a unique identity linked with women’s rejection of heteronormative dating scripts. The consistency of effects across samples and moderate effect sizes also helps to clarify the inconsistent links between feminist identification and heteronormative dating scripts in past work (Bay-Cheng & Zucker, 2007; Rudman & Fairchild, 2007: cf. Paynter and Leaper, 2016). Taken together, the finding that feminist identity and sexist attitudes were both consistent, unique predictors of heteronormative dating scripts further demonstrates that attitudes and identities that support gendered structures of power are closely tied to these heteronormative dating scripts.

A novel finding from our research was that women’s greater preference for a dominant male partner predicted greater endorsement of heteronormative dating scripts, over and above their sexism and feminist identity, as well as their short- and long-term relationship preferences. This association was consistent with the claim that women with a heightened personal preference for a dominant male partner will find heteronormative dating scripts more appealing because those behaviours are signals that their future romantic partner will adopt desired roles and qualities, including a more assertive and dominant relationship role (Cameron & Curry, 2020; Lamont, 2014, 2015; McCarty & Kelly, 2015; Paynter & Leaper, 2016). However, we failed to find evidence that women endorsed heteronormative dating scripts as a function of a stronger preference for a long-term relationship, under the expectation that those scripts signal men’s willingness to invest in their partners (see Lamont, 2014; Lever et al., 2015). Instead, women with a lower preference for short-term relationships endorsed heteronormative dating scripts more strongly. Thus, perhaps women who wish to avoid short-term relationships endorse heteronormative dating scripts because they hold heightened proscriptions against men who desire casual sexual encounters (see Lamont, 2014; Lever et al., 2015), rather than any particular preference for an invested relationship relative to other women. These findings extend prior research by indicating that women’s endorsement of heteronormative dating scripts extend beyond gendered attitudes and may be simultaneously driven by their prescriptions and proscriptions for romantic partners.

Exploratory analyses also indicated that women’s benevolent sexism is associated with a greater endorsement of heteronormative dating scripts partly due to greater preferences for a dominant partner. Prior work has identified that women higher in benevolent sexism desire male partners who will invest in romantic relationships, be reliable and chivalrous partners, and will protect and provide for women (see Cross et al., 2016; Gul and Kupfer, 2019; Hopkins-Doyle et al., 2019; Lamont, 2014; Lever et al., 2015), but has focused less on the extent to which those preferences for provision and protection overlap with relatively more aggressive partner preferences for dominance, authoritativeness, and being an “alpha male”. Our findings build on this research and suggest that part of the reason why benevolently sexist women prefer heteronormative dating scripts is due to their greater preference for dominant men, and that these scripts support men’s enactment of dominance. Our perspective is consistent with research showing that women higher on sexism tend to prefer higher-status partners (Eastwick et al., 2006; Lee et al., 2010; Sibley & Overall, 2011; Travaglia et al., 2009) and find assertive dating strategies more desirable (Hall & Canterberry, 2011).

Practice Implications

The current research provides direct evidence that women’s endorsement of heteronormative dating scripts does not simply reflect benign preferences for romance, as they are underpinned by attitudes that reinforce gender inequality even when accounting for other partner and relationship preferences. We reiterate the perspective that even personal partner and relationship preferences are potentially problematic when one considers their context (see Eagly et al., 2009; Eastwick et al., 2006; Sibley and Overall, 2011). That is, women’s heightened preference for a dominant, protective and “alpha male” partner who provides physical protection and material resources is a relatively adaptive response to contexts in which women experience more threats to safety, are excluded from equal participation in careers or politics, and are pressured to adopt domestic-focused roles (e.g., Eagly et al., 2009; Eastwick et al., 2006; Sibley and Overall, 2011). Further, adopting preferences and roles as a response to gender inequalities can unintentionally reproduce and reinforce the conditions that produced those inequalities in the first place (e.g., Cameron and Curry, 2020; Eagly et al., 2020; Haines et al., 2016). For example, considering men’s (rather than women’s) marriage proposal to be sweeter and more romantic (e.g., Robnett & Leaper, 2012) encourages men to behave in agentic ways and women to be passive receivers in romantic relationships between women and men. Thus, heteronormative dating scripts may be viewed as a manifestation of sexist beliefs, which have wider implications beyond intimate relationships (Barreto & Doyle, 2023).

The findings from our study offer an important point of reflection about potentially problematic aspects of the adherence to heteronormative scripts in current romantic relationships between women and men. Of particular interest to feminist activists, our results help to illustrate that even many seemingly innocuous romantic conventions may be problematic because they are based in sexist expectations about gender and are not simply accounted for by personal preferences. Further, given that expectations for heteronormative romantic scripts in dating contexts are partly grounded in sexist attitudes, educators might also consider the need to critically examine conventional assumptions about gender roles in relationships, particularly how they may be perpetuated within sex and relationships education programs.

Limitations and Future Research Directions

One strength of this research was that we extended prior research on young undergraduate participants (e.g., McCarty and Kelly, 2015; Paynter and Leaper, 2016; Robnett and Leaper, 2013) by sampling older age groups from the general population. A caveat of our approach is that we only examined single women and expectations for dating. Future research could consider women currently in a relationship with a man and index experiences of romance in their relationships, including comparisons between different relationship stages (e.g., dating vs. engaged vs. married). Heteronormative dating scripts are theorised to provide a script for earlier stages of relationship formation, and thus most research has examined them in dating contexts (e.g., pursuing relationships, proposing marriage). Exploring how heteronormative dating scripts operate in long term relationships, when a partnership or marriage is established, would offer an important extension to the current work. Similarly, future research could also investigate individual differences that would make heteronormative dating scripts relatively less appealing, such as those related to agency, power, and dominance. For instance, sense of control has been found to predict a greater willingness to initiate relationships among women (MacGregor & Cavallo, 2011), and weakens the preference for high-status romantic partners (Lammers & Imhoff, 2021), whereas a greater sense of power is associated with increased sexual assertiveness (Lammers & Stoker, 2019). If women who lack agency, power, and dominance are more likely to endorse these scripts, this supports the argument that such scripts are rooted in a sexist ideology that functions to disempower women and restrict their agency.

The goals of the current research focused on women’s endorsement of dating scripts for relationships between women and men; exploring romance scripts in other gender groups is an important next step. We expect similar results to emerge among men who date women, with a few key differences. First, we expect hostile sexism to be a stronger and more consistent predictor of heteronormative dating scripts for men compared to women, in line with Paynter and Leaper’s (2016) findings. Second, given that Paynter and Leaper (2016) found that men’s (but not women’s) feminist identity was negatively associated with endorsement of heteronormative dating scripts, we expect this association may be stronger among men than women. In addition, given that heteronormative dating scripts are theorised to be signals of distinct relationship roles, the correlates of partner preferences should be the inverse to women: men who desire more submissive partners should endorse heteronormative dating scripts more strongly (see Cameron and Curry, 2020; Lamont, 2014, 2015; McCarty and Kelly, 2015). Finally, exploring the numerous ways queer people challenge and reject heteronormative dating scripts, and enact more egalitarian dating behaviour, will offer important insight into the alternative dating norms that do not limit people to gendered ways of behaving (e.g., men proposing to women vs. marriage proposals among same-gender couples; see Jowett and Peel, 2019; Lamont, 2017). Certainly, contrasting more egalitarian versus conventional gendered dating behaviour is essential to showcase the restrictive nature of heteronormative dating scripts.

Despite the strengths of the current study, including sampling across age groups from the general population and statistically accounting for sociodemographic variables, the correlational methods used limits causal conclusions. Replicating the patterns of our results in romantic interactions in experimental and/or longitudinal studies that indexes the extent to which people’s sexism precedes their endorsement and enactment of heteronormative dating scripts would also be useful to assess the extent to which sexist attitudes and heteronormative dating scripts reciprocally influence one another. Similarly, investigating the consequences of people’s heteronormative dating scripts for gender inequalities, both within the relationship and beyond the relationship, is a key direction for future research. While recent work shows that people’s endorsement of sexism is linked with greater inequalities in their household (i.e., greater gendered divisions of labour, Gerst et al., 2021; Waddell et al., 2021), it is also valuable to explore whether heteronormative dating scripts promote greater prioritisation of men’s needs and desires beyond the relationship, such as partners’ prioritising men’s (but not women’s) career successes as justified by men’s expected roles (e.g., needing to buy an expensive engagement ring). Lastly, examining the predicted links between sexism, feminism, relationship and partner preferences and heteronormative dating scripts in less egalitarian countries is an especially important next step, given that women’s needs for protection and provision will be qualitatively different in contexts that place greater limits their roles (see Eagly et al., 2009; Eastwick et al., 2006; Sibley and Overall, 2011).

Conclusion

Three studies examined a range of predictors of single women’s endorsement of heteronormative dating scripts, such as expecting men to pay for romantic dates and take the lead in romantic contexts. Evidence showed that women’s endorsement of heteronormative dating scripts was strongly predicted by their hostile sexism, benevolent sexism, and lower feminist identity. To a lesser degree, evidence also showed that women’s preference for dominant partners and preference against short-term relationships were related to their endorsement of heteronormative dating scripts, over and above gender attitudes. Our findings provide further evidence that the endorsement of heteronormative dating scripts is neither trivial nor free of negative implications. Expectations that men should take the lead in romance with women (i.e., heteronormative dating scripts) are intertwined with beliefs and attitudes that men should also take leading roles in the workplace, politics, and higher education (i.e., sexism and lack of feminist identity), even after accounting for other personal preferences. Future research should examine whether a greater diversity in relationship roles and dating scripts advances societal change towards gender equality.