Benevolent and hostile sexism in a shifting global context

Abstract

The theory of and research on ambivalent sexism — which encompasses both attitudes that are overtly negative (hostile sexism) and those that seem subjectively positive but are actually harmful (benevolent sexism) — have made substantial contributions to understanding how sexism operates and the consequences it has for women. It is now clear that sexism takes different forms, some of which can be disguised as protection and flattery. However, all forms of sexism have negative effects on how women are perceived and treated by others as well as on women themselves. Some of these findings have implications for understanding other social inequalities, such as ableism, ageism, racism and classism. In this Review, we summarize what is known about the predictors of ambivalent sexism and its effects. Although we focus on women, we also consider some effects on men, in particular those that indirectly influence women. Throughout the Review we point to societal shifts that are likely to influence how sexism is manifested, experienced and understood. We conclude by discussing the broader implications of these changes and specifying areas of enquiry that need to be addressed to continue making progress in understanding the mechanisms that underlie social inequalities.

Introduction

Addressing the substantial gender inequalities that exist across a range of life domains¹ requires an understanding of the effects of sexism. According to ambivalent sexism theory², which was developed to account for the relationship between (cisgender and heterosexual) men and women, sexism includes a hostile component (overtly negative attitudes about men and women) and a benevolent component (attitudes towards men and women that seem subjectively positive but are actually harmful). These components differ in tone but are positively correlated and work together to perpetuate gender inequalities².

Research suggests that children^3,4, young people^5,6 and adult men and women around the world⁷ endorse ambivalent sexism (that is, agree with items that measure both benevolent sexism, such as “women should be protected by men,” and hostile sexism, such as “women seek to gain power by getting control over men.”). Indeed, according to one study, half of the British population holds these attitudes⁸. Ambivalent sexism is therefore a critical factor in shaping girls’ and women’s lives in a variety of social contexts.

Although there has been substantial progress in this area of research⁹, theoretical insights are often assumed to hold across time, cultures and social groups. Consequently, theoretical advances do not account for societal shifts in gender relationships over time, or consider the socio-political and cultural contexts in which they operate. For example, binary views of gender are more widely challenged than before¹⁰ (at least in some places), which influences ideas about what it means to be a man or a woman, as well as what relationships between individuals of different gender groups should look like. In addition, legal and policy developments change the background against which relationships between men and women play out. For example, the number of countries offering paid paternity leave has increased, and so has its uptake¹¹, which has led to greater labour participation of both mothers and fathers¹². Although the negative effects of the COVID-19 pandemic on workload and household work burdens disproportionately influenced women^13,14, changes in women’s participation in the workforce provided a normative climate against which couples could evaluate, and be evaluated by others as a function of, their decisions in this area. In addition, because divorce and same-sex parenting and single parenting are increasing^15,16, men and women now often have both traditionally male and traditionally female roles within families.

More broadly, the spread of neoliberalism as a prevailing socio-political ideology has influenced ideas of equality and how best to achieve progress (for example, by changing individuals rather than social structures)¹⁷. For women, this shift has been associated with greater agency in terms of workplace involvement and contribution to the global marketplace¹⁸, but often without adequate policy and structural support (such as adequate parental leave or strong employment non-discrimination laws). Instead, women are simultaneously tasked with traditional gender chores, such as childcare and housekeeping, while also being told to ‘lean into’ their careers when they inevitably experience obstacles not faced by heterosexual men. Neoliberalism both empowers women to strive for, and blames women for failing to achieve, outcomes that are often beyond their individual control, masking subtler and more blatant ways in which sexism shapes and constricts lives. Although the full extent of the consequences of this global shift is not straightforward, these changes might influence how sexism is expressed and experienced.

Researchers have begun to recognize such societal shifts in ideas about gender and romantic relationships beyond heterosexual couples¹⁰, but research in this area is still scarce. In addition, the geographical contexts of research on ambivalent sexism have diversified¹⁹, but the majority of research is still carried out in a restricted number of countries (including New Zealand, Spain, Turkey, the UK and the USA), so comparative work and reflections on cultural specificities are still largely missing.

In this Review, we take stock of the current understanding of ambivalent sexism to facilitate further research that addresses relevant societal shifts and their global contexts. First, we describe benevolent and hostile sexism and their predictors. We then review what is currently known about how both benevolent sexism and hostile sexism influence how women are perceived and treated (by both men and women). Next, we discuss how these types of sexism influence how women feel and behave, as well as romantic relationships between men and women. Although the applicability of findings to present socio-political contexts will be flagged throughout the paper, the final section more thoroughly considers shifts in global context and how these open up avenues for future research. We focus on research published within the past five years, but key older studies are also mentioned where they exemplify core theoretical aspects. We also focus primarily on sexism towards women. Ambivalent attitudes towards men also encompass hostile and benevolent components²⁰, but these attitudes are less well understood. Importantly, they are strongly related to ambivalent sexism towards women and have been proposed to serve the same function of supporting male dominance over women²¹. Some examples of effects of ambivalent sexism on men are mentioned, especially where their effects on women are most direct.

Two forms of sexism

Prejudice is traditionally conceptualized as a negative attitude that explains and shapes antagonistic relationships between dominant and subordinate groups²². Sexism is a form of prejudice that specifically subordinates women to men. Although sexism can take very clearly negative (and even violent) forms, attitudes towards women are not necessarily negative in obvious ways; in fact, people often describe women more positively than they describe men — the ‘women are wonderful’ effect²³. However, positive descriptions of women tend to be restricted to traits related to warmth (women are sociable and nice), whereas men are more positively described in domains such as agency and competence that determine status and power in society (men are bright and capable)^23,24. In addition, relationships between men and women are not necessarily characterized by antagonism; instead, they often involve the coexistence of male dominance with cooperation and even intimacy. Ambivalent sexism theory^2,25 was developed to account for these specific circumstances and proposes that sexism combines antipathy (hostile sexism) with subjective benevolence (benevolent sexism) towards women, which together maintain men’s dominance over women.

Hostile sexism is similar to the traditional conceptualization of prejudice as antipathy: it is negative in tone and disparages women who challenge traditional gender roles and ideologies (for example, professionally successful women). It communicates a view of gender relationships as competitive, with women wanting to dominate men and threatening men’s higher status in society. By contrast, benevolent sexism has a more positive tone: it idealizes and flatters women who embody traditional ideals (such as stay-at-home mothers), and portrays women as morally pure and uniquely caring, but also as weak and unable to take care of themselves. Benevolent sexism portrays gender relationships as cooperative and complementary, with men in charge of protection and security and women dedicated to nurture and reproduction.

Both hostile and benevolent sexism encompass three components, which are assessed using the Ambivalent Sexism Inventory²⁵: paternalism, gender differentiation and heterosexual intimacy (Table 1). Paternalism refers to men’s superiority over women, either aggressively (in hostile sexism) or protectively (in benevolent sexism). Gender differentiation draws a line between men and women, distributing roles associated with power to men (in hostile sexism) and nurturing roles to women (in benevolent sexism). Heterosexual intimacy accommodates heterosexual men’s dependency on women for sexual satisfaction; hostile sexism aims to restrict women’s use of sex to manipulate men and benevolent sexism idealizes women as necessary to complete men.

Table 1 Components of hostile and benevolent sexism and example items from the ambivalent sexism inventory²⁵

Although hostile and benevolent sexism are opposite in tone, they both draw on gender stereotypes and therefore tend to be positively associated²⁵ across nations⁷. The more hostile sexism there is in a given society, the more individuals in that society also endorse benevolent sexism⁷. Correspondingly, women who report more daily experiences with hostile sexism also report more daily experiences with benevolent sexism²⁶. However, because hostile and benevolent sexism express gender stereotypes in distinct ways, there are important differences in how these two forms of sexism are perceived: hostile sexism is regarded as more objectionable than benevolent sexism²⁷, in part because it is perceived as more sexist²⁸. Benevolent sexism is perceived as harmless²⁹ and even romantic³⁰, and this makes men who endorse benevolent sexism seem likeable^19,28,31. Hostile sexism is less frequently endorsed⁷ and expressed, and indeed women report more lifetime experiences with benevolent than hostile sexism²⁶. However, in part because of the warmth it transmits, benevolent sexism can make hostile sexism seem more acceptable when expressed by the same person³².

Benevolent sexism is also seen as less objectionable than hostile sexism because it offers women benefits. For example, because benevolent sexism offers protection to women³³, men who express benevolent sexism are seen as caring³⁴. In addition, women who endorse benevolent sexism see the social system as fair³⁵ and consequently report greater life satisfaction³⁶.

In sum, both benevolent and hostile sexism express the belief that women are and should be submissive to men. However, benevolent sexism is considered more acceptable, and at times even flattering. This positive perception is a key property of ambivalent sexism that contributes to the perpetuation and pervasiveness of gender inequalities.

Predictors of ambivalent sexism

Understanding how sexism operates requires consideration of why people might endorse sexist views. Whereas some factors predict endorsement of both benevolent and hostile sexism, others appear to uniquely predict one type of sexism (Fig. 1).

Fig. 1: Predictors of benevolent and hostile sexism.

Unique and shared predictors of benevolent and hostile sexism. Results can differ across studies; the predictors represented here reflect the most consistent associations reported in the literature.

Demographic factors

Existing comparative evidence using the Ambivalent Sexism Inventory relies upon non-representative samples²⁵, so it is not possible to establish precisely how benevolent and hostile sexism vary across countries. However, the evidence suggests that hostile sexism is strongest in countries characterized by lower gender equality and less wealth, health and education, as measured by United Nations indicators³⁷. These findings suggest that sexism is not only detrimental to women’s own advancement, but might also be detrimental to society as a whole, reducing overall educational achievement and impairing social prosperity.

Because benevolent and hostile sexism serve to justify and perpetuate male privilege, it is not surprising that men endorse benevolent and hostile sexism to a greater extent than women²⁵ across nations⁷, with gender differences typically being larger for hostile than benevolent sexism. Research comparing sexism scores between cisgender (those who identify with the gender they were assigned at birth), transgender (those who identify with a gender different from that assigned at birth) and gender-diverse individuals (those who identity as non-binary, genderfluid or genderqueer, for example) has produced mixed results. One study revealed higher hostile sexism scores among cisgender men, and lower benevolent sexism scores among cisgender women and gender-diverse individuals assigned female at birth, than other gender groups³⁸. However, another study revealed higher scores on both components among transgender than cisgender individuals³⁹. These discrepancies highlight the need for more research in this area.

Regarding age, men’s hostile sexism and women’s hostile and benevolent sexism are higher in adolescence and young adulthood, lower in middle adulthood, and again higher in older age. By contrast, men’s benevolent sexism increases with age^6,40. This finding is argued to reflect age-normative changes in the importance of goals related to power, identity and relationships that underlie ambivalent sexism, such as the fact that middle-aged individuals have greater relational and role stability as well as greater independence than young and older adults. It remains to be seen whether these age and gender patterns hold across time and cultures with different views on power, identity and relationships.

Studies are beginning to show the importance of taking race into account when attempting to understand the drivers of sexism (Box 1). One study showed that Black American women endorse benevolent sexism to a greater extent than white American women⁴¹. Crucially, benevolent and hostile sexism are not significantly correlated among Black American participants⁴², and there is also no gender difference in the endorsement of these two types of sexism among these participants^41,42. However, the benevolent sexism subscale of the Ambivalent Sexism Inventory has poor measurement properties for Latinx and African American participants, suggesting that it is not appropriate for assessing this construct in all racial or ethnic groups⁴². This measurement issue also highlights the need to expand understanding to other cultural contexts and intersections between multiple identities.

Even though ambivalent sexism is endorsed across sexual orientations^43,44, individuals who are, or desire to be, in heterosexual romantic relationships report stronger benevolent and hostile sexist attitudes than sexual-minority respondents^43,44,45. However, as mentioned above, existing measurement tools are not appropriate for comparing heterosexual and sexual-minority samples, creating doubt about how these differences in scores should be interpreted⁴⁶.

Box 1 How benevolent sexism justifies racist attitudes

Research on ambivalent sexism can contribute to understanding how sexism can exacerbate race inequalities. For example, stereotypes of white American women tend to infantilize them as dependent and helpless (in line with benevolent sexism), whereas stereotypes of Black American women portray them as hypersexualized (in line with hostile sexism)¹⁷³ and Asian American women are seen as both hypersexual and submissive¹⁸². Indeed, women from racial and ethnic minorities often experience a combination of racism and sexism¹⁸³. This racialized sexual harassment can, in turn, justify their sexual exploitation, increase the extent to which they are blamed for sexual violence¹⁸⁴, and reduce the extent to which they are willing to complain about sexual harassment for fear of not being believed or of attracting attention to themselves as targets of sexual attention¹⁸³. Moreover, American students associate whiteness with femininity and blackness with masculinity, which can justify harsher treatment of Black men and women more generally. In addition, stereotypes of strength and aggressiveness contribute to why police officers intervene less frequently in domestic abuse incidents involving Black American women¹⁸⁴.

Muslim women are essentialized by particular understandings of Islam that render them responsible for the family’s honour by being modest and chaste. Yet, at the same time, these views of Muslim women are at the core of Western Islamophobic perceptions of Muslim people as oppressive and inferior, with the use of the hijab seen to symbolize women’s forceful subjugation¹⁸⁵. This simultaneously racist and (benevolently) sexist discourse is well encapsulated by the term ‘hijabophobia’¹⁸⁶, which reflects the view of a “submissive and voiceless Muslim woman who needs to be saved from her barbaric and misogynistic religion.”¹⁸⁵

Situational factors

The more an individual’s circumstances reflect traditional gender roles, the higher their benevolent sexism scores. For example, having more children predicts stronger endorsement of benevolent sexism two years later — and not the other way around⁴⁷. That is, people might endorse benevolent sexism to justify the traditional gender roles they have adopted in their life, rather than adopting these roles because they endorse benevolent sexism. If this is the case, then changes in gender roles — for example, through increases in same-sex parenting, or men’s increased participation in childcare — might lead to reductions in endorsement of benevolent sexism.

Ideological factors

Religiosity is another form of traditionalism that drives sexism. Both forms of sexism, but benevolent sexism in particular, have been positively associated with religiosity across affiliations such as Christianity and Islam^{48,49,50,51,52}. Simple reminders of religion can be sufficient to increase endorsement of benevolent sexism⁵³. Some have argued (but not yet demonstrated) that reductions in religiosity worldwide coincide with scientific and technological advances that increase fertility and reduce child mortality. These advances thereby reduce the need to control women’s reproduction and sphere of activity, which was historically facilitated by religious norms⁵⁴. Thus, one prediction is that declines in religiosity might translate into a reduction in sexist attitudes.

Ideological variables related to political conservatism also predict sexism. In fact, political conservatism has been found to explain more variance in ambivalent sexism than gender⁸. Moreover, in both men and women, hostile sexism is predicted most strongly and consistently by social dominance orientation (a view of the world in which groups of people compete for dominance and superiority), whereas benevolent sexism is most strongly and consistently predicted by right-wing authoritarianism (which stems from perceptions of the world as a dangerous place and reflects a desire for security)^55,56. These findings support the idea that hostile sexism is primarily driven by the idea that men’s dominance over women is both appropriate and desirable, a belief that can be shared by men and women. By contrast, benevolent sexism is driven by a need for security (implied in right-wing authoritarianism). These findings lead to the prediction that political rhetoric associated with the rise in right-wing populism and world events that promote the idea that the world is an unsafe place (such as the COVID-19 pandemic) might increase endorsement of these forms of sexism.

Further evidence that benevolent sexism is driven by a need for security is that women’s endorsement of benevolent (but not hostile) sexism increases when they believe that men have more hostile attitudes towards women⁷. Women also endorse benevolent sexism to a greater extent when their fear of crime is enhanced⁵⁷. This finding leads to the prediction that actions that highlight women’s vulnerability to sexual violence (for example the #MeToo movement) might ironically increase women’s feelings of insecurity and their endorsement of benevolent sexism in an attempt to secure protection. Similarly, particularly high exposure to discrimination among Black American women (which raises the need for safety) might explain why they endorse benevolent sexism to a greater extent than white American women⁴¹, but this has not been directly tested. Furthermore, men and women who are more afraid of disease and contagion endorse benevolent sexism to a greater extent, presumably because the restrictions benevolent sexism imposes on women’s behaviour can protect against disease⁵⁸. This finding is particularly interesting in light of the COVID-19 pandemic — fear of disease during the pandemic might have led to increases in benevolent sexism. Finally, men’s benevolent sexism increases when they feel anxious about their sense of manhood⁵⁹ or their romantic relationship⁶⁰. Interestingly, men who do not have such security needs (men with a tendency to avoid attachment) report low benevolent and high hostile sexism⁶⁰.

In sum, a range of factors increase benevolent and hostile sexism, some of which are unique to each form of sexism. Importantly, changes within a given society in these various predictors (for example, general decreases in religiosity, or temporary fluctuations in insecurity, particularly for women) might have implications for the manifestation of ambivalent sexism. The direct links between these societal changes and endorsement of ambivalent sexism requires further evidence.

Effects of ambivalent sexism

It is important to understand the different ways in which sexism can be expressed because they can have different consequences. In this section we summarize and compare the effects of benevolent and hostile sexism. Although the review is not exhaustive, it includes those effects that are most crucial for understanding the impact of ambivalent sexism across a range of domains (Table 2).

Table 2 Summary of key effects of benevolent and hostile sexism

Gender roles

Hostile and benevolent sexism contribute to maintaining the status quo by regulating how women (and men) behave. Hostile sexism is correlated with negative stereotypes or disparaging views about women who challenge the status quo by behaving non-traditionally, such as career women⁶¹, women in stereotypical male employment positions (such as managers)⁶² or feminists⁶¹. By contrast, benevolent sexism is associated with positive stereotypes about or support for women who reinforce gender inequalities by behaving in line with traditional gender roles, such as housewives^7,25 or women who do not confront sexism⁶³. In addition, hostile sexism punishes women who deviate from traditional gender roles and benevolent sexism encourages women to abide by them in exchange for protection and financial security. For example, women’s endorsement of hostile sexism is associated with the derogation of women who breastfeed in public⁶⁴ and women who are highly sexually active⁶⁵; men’s endorsement of benevolent sexism is associated with favourable views of women who breastfeed their children in private⁶⁶, and predicts unfavourable attitudes towards women who engage in pre-marital sex⁶⁷. Men who endorse benevolent sexism often engage in protective behaviours towards women (the ‘white knight’ effect)³³, and the idea that women need protection is often used as an argument in favour of restricting transgender women’s access to the bathroom of their affirmed gender⁶⁸.

Men do not necessarily benefit from these restrictive attitudes. Indeed, both men and women who do not conform to the rigid gender role prescriptions that underlie ambivalent sexism — such as LGB individuals^69,70, men who perform stereotypically feminine behaviours (such as styling someone’s hair)⁷¹, men who express gender-egalitarian beliefs⁷² and transgender individuals^73,74 — are the target of negative attitudes, particularly by those high in hostile sexism⁷⁰. This lack of conformity is perceived to threaten the gender hierarchy in which men dominate, so it is not surprising that these negative attitudes tend to be stronger among men than women^73,74. These rigid notions of gender contribute to regulating men’s behaviour, and directly or indirectly influence women’s social standing. It is unclear whether these gender role prescriptions (and their effects on how men and women are perceived) are retained as men and women are seen to successfully take on more counter-stereotypical roles, such as women being successful at work or men successfully parenting.

Self-views

Sexism influences how women feel and think about themselves and their bodies. Benevolent sexism is particularly problematic in this regard because its flattering and less obviously sexist tone discourages women from rejecting the stereotypes it makes salient. Consequently, women exposed to benevolent (but not hostile) sexism describe themselves more in line with gender stereotypes and remember more gender-stereotypical information about themselves^75,76.

Beauty ideals are important for the subjugation of women because they often reduce women to sex objects, draw attention away from their competence, and undermine their self-confidence, thereby facilitating men’s dominance. Both benevolent and hostile sexism are associated with the endorsement of beauty ideals (such as thin bodies)⁷⁷, self-objectification⁷⁸ and body dissatisfaction⁷⁹. These, in turn, make women vulnerable to psychological ill health, for example, by decreasing adherence to physical medical exams and exacerbating eating disorders⁸⁰. Interestingly, benevolent sexism has been associated with both thin^77,81 and large⁷⁹ body ideals, the former presumably because they render women fragile and dependent, and the latter presumably because large bodies signal fertility. In addition, benevolent sexism has also been associated with women’s increased use of cosmetics, which can improve satisfaction with appearance⁸¹ and reverse the relationship between benevolent sexism and body image^79,82. In sum, both forms of sexism lead to attitudes that seek to control, and draw attention to, women’s appearance, but the effects of benevolent sexism are slightly more complex.

Sexism also influences men’s views of themselves and their bodies. Although sexism can enhance the value of being a man, such narrow notions of masculinity can lead those who do not (always) fit this notion to experience low self-esteem and body dissatisfaction⁸³. The role of ambivalent sexism in beauty ideals for transgender and gender-diverse people has not been directly researched and is an important focus for future research.

Affect and physiology

Automatic responses to both types of sexism are evident in changes in physiology and affect, which might place women at increased risk of physiological ‘wear and tear’, including cardiovascular disease, over the life course⁸⁴. Cardiovascular disease is the leading cause of mortality among women around the world, but it remains under-recognized, underdiagnosed and undertreated⁸⁵. Thus, research into the specific contributions of ambivalent sexism to this condition is critical to health equity. For example, being a target of either benevolent^86,87 or hostile⁸⁷ sexism leads to cardiovascular signatures indicative of threat. However, being a target of hostile sexism leads to a greater initial spike in cardiovascular reactivity, whereas benevolent sexism leads to a lower initial spike but slower recovery to baseline⁸⁷ (Fig. 2). These findings might be consistent with evidence that exposure to benevolent sexism increases activation of the dorsolateral prefrontal cortex, a brain region involved in cognitive control and thought suppression, suggesting that women ruminate about benevolent sexism for some time after experiencing it⁸⁸. Sexism can also be a substantial physical stressor for men when they feel their adherence to strict notions of masculinity is questioned⁸⁹.

Fig. 2: Cardiovascular responses to ambivalent sexism.

Women’s cardiovascular systems are relatively more reactive to experiences of hostile sexism, but relatively slower to recover and return to baseline after experiences of benevolent sexism. This illustration is based on data from ref.⁸⁷, where cardiovascular responses were measured by systolic and diastolic blood pressure, heart rate, cardiac output, pre-ejection period and total peripheral resistance.

Consistent with the portrayal of sexism as a stressor, it can elicit anxiety in men²⁹. For women, experiences with both benevolent and hostile sexism are associated with increased self-reported anxiety^90,91 and anger^28,92. However, these associations are relatively stronger for hostile than benevolent sexism^28,91, perhaps because women do not always identify benevolent sexism as overtly (or uniquely) negative. Men and women tend to overestimate and underestimate how women’s affect will be influenced by exposure to hostile and benevolent sexism, respectively, potentially because they have only a naive understanding of the difference between them²⁹. Furthermore, some evidence suggests that the affective impact of benevolent sexism varies depending on the specific component of benevolent sexism experienced; specifically, one study showed that experiences with protective paternalism are associated with more self-doubt, lower self-esteem and poorer psychological wellbeing, whereas experiencing complementary gender differentiation was associated with less self-doubt, more self-esteem and better wellbeing⁹³. Future work must continue to disentangle the overlapping and unique affective and physiological sequelae of exposure to various forms of ambivalent sexism among women.

Violence towards women

Restrictive gender role prescriptions can encourage men who feel their masculinity is threatened to behave in ways that they believe demonstrate their manhood, such as displaying aggression⁹⁴. Only hostile sexism has been shown to predict men’s self-reported likelihood to sexually harass women⁹⁵ and tolerance of sexual harassment⁹⁶. However, both hostile and benevolent sexism predict men’s inclination to commit acquaintance rape and blame victims of sexual assault^97,98. For hostile sexism, this is because it is associated with the idea that women actually want and control sex even when they claim not to. For benevolent sexism, this is restricted to cases of acquaintance rape and attributed to the idea that women who enter a relationship with a man invite sexual attention⁹⁷. Because of these perceptions of victims’ culpability, those high (versus low) in benevolent sexism recommend more lenient sentences for perpetrators of acquaintance rape⁹⁹. In addition, because those high in hostile sexism believe that victims actually want sex, hostile sexism predicts less support for measures that reduce male violence towards women and more support for measures that encourage women to avoid male violence¹⁰⁰; benevolent sexism is positively associated with support for both types of measure owing to its focus on women’s protection¹⁰⁰.

However, the protection against violence offered by benevolent sexism does not necessarily extend to Black women. In situations where police shoot suspects of armed robberies, benevolent sexism leads to perceptions of white (versus Black) female suspects as more feminine, which in turn leads to more blame on the officer than the suspect when the suspect is white, but not when she is Black¹⁰¹. This underlines the need for more research into the intersection of race and gender to examine the limits of ambivalent sexism theory, or expand it to diverse racial groups.

Hostile sexism is also linked to sexual aggression towards women by increasing objectification^102,103 and denying women uniquely human emotions¹⁰⁴. Benevolent sexism has no such effect. In fact, one study showed that, for both men and women, benevolent sexism increases the association of women with positive and uniquely human emotions¹⁰⁴. The fact that benevolent sexism can promote this positive image of women might be another reason why women feel flattered by it, despite the fact that it can nevertheless be associated with negative outcomes, including gender violence.

Careers

Sexism influences how women are perceived and treated in the work domain. For example, hostile sexism is associated with the idea that gender income inequality is legitimate because it arises from women’s choice of work arrangements that are associated with lower salaries¹⁰⁵. In addition, hostile sexism leads to fewer recommendations to hire women as managers⁶² and predicts negative attitudes towards women managers¹⁰⁶. Once at work, female employees are often treated in benevolently sexist ways by receiving ample praise but little concrete recognition for their work, such as career-enhancing opportunities¹⁰⁷, promotions or salary raises^108,109. Benevolent sexism is associated with lower competency standards for female (versus male) employees, resulting in positive evaluations of women when they are compared to other women, but not when they are compared to men (to whom they are deemed inferior)¹¹⁰. Benevolent sexism also results in more dependency-oriented (versus autonomy-oriented) help offered to female employees¹¹¹, which leads others to perceive women as less competent¹¹², irrespective of whether or not they have requested the help offered¹¹³. Merely observing a female job candidate being treated in a benevolently sexist manner leads observers to infer that she is less competent or hireable¹¹⁴. Finally, benevolent sexism has been related to more support for employment equity policies, but only for stereotypically feminine, not masculine, positions¹¹⁵. Taken together, this evidence suggests that benevolent sexism encourages behaviours towards female employees that seem positive, but in fact undermine women’s careers. Thus benevolent sexism might partially explain why women remain under-represented in higher-status and more-powerful roles. It remains to be examined whether these relationships become weaker when and where sexist individuals are a minority in the workplace and their attitudes towards female employees have less power.

In terms of career choices, benevolent sexism directs boys to stereotypically male domains, such as business- and maths-related fields, and girls to stereotypically female domains, such as the arts^116,117. These choices are often influenced by mothers’ benevolent sexist attitudes¹¹⁸. In addition to shaping career choices, benevolent sexism can impair how women actually perform at work, especially if the task is stereotypically masculine¹¹⁹, by decreasing self-efficacy¹²⁰ and increasing thought intrusions^88,121. At the same time, benevolent sexism restricts women’s access to career-enhancing support¹²². Women high (versus low) in benevolent sexism are more likely to accept patronizing behaviour from men, which they might perceive as supportive, but which can perpetuate their dependence on men and undermine their career prospects^34,111. Irrespective of their benevolent sexist attitudes, women might refuse support when they believe that accepting such support would confirm the sexist belief that they are dependent upon men¹²³.

Together, these findings show that although hostile sexism has more immediate and negative emotional effects than benevolent sexism, both negatively influence women’s views of themselves, and benevolent sexism in particular shapes women’s career choices and performance. The fact that benevolent sexism is often not identified as problematic means that an important deterrent of women’s careers frequently remains unaddressed. However, research on this topic might need updating, particularly because some effects of benevolent sexism rely on its subtlety and perceived flattery, which might wane when and where its sexist nature is more visible.

Healthcare

Both men’s and women’s healthcare is compromised by sexist views of women as emotional and men as brave¹²⁴. However, only support for addressing women’s (but not men’s) pain is negatively related to benevolent and hostile sexism¹²⁵. Moreover, patronizing attitudes characteristic of benevolent sexism are associated with discouraging women to undergo mammography to avoid the anxiety it might provoke, despite evidence suggesting that mammography reduces women’s anxiety about having breast cancer¹²⁶.

In addition, the idealization of women as mothers (which is fundamental to benevolent sexism) leads to controlling attitudes about pregnant women’s choices¹²⁷ and opposition to both elective and traumatic abortion¹²⁸. Men’s and women’s benevolent sexism is associated with negative attitudes towards women who have an abortion, even if it is medically motivated¹²⁹. In fact, although sexist attitudes can coerce women towards abortion when families seek to restrict the birth of female children¹³⁰, sexism can also limit access to abortion. For example, benevolently sexist language has been identified in policy-making discussions to justify restricting women’s access to abortion services¹³¹. Ironically, rather than protecting women’s health, research in the USA has shown that state-level abortion bans are tied to increased total maternal mortality¹³². Consistent with benevolent sexism, those who object to abortion often claim that they wish to protect women from the negative emotions it might elicit (such as shame, grief and regret) and portray women as incapable of making good decisions¹³³. Such arguments might take on greater importance as abortion becomes legal in more places because they provide an additional (but informal) hurdle women might need to overcome to access this care¹³⁰. Of course, benevolently sexist arguments can also be used to ensure that abortion does not become legal, as in the USA, where the Supreme Court overturned previously established abortion rights in 2022.

Legal decisions

Finally, court decisions and criminal sentencing often reflect benevolent sexism, in this case often benefiting women¹³⁴. For example, judges tend to sentence female defendants to less time in prison than male defendants for the same crime, which can be attributed to benevolent sexist ideas that women are weaker than men. Similarly, judges are more likely to allow a divorced mother to relocate with her children away from the father than when exactly the same case is presented by a father, which in turn can be attributed to the benevolently sexist belief that mothers are inherently more essential to children than fathers. The legalization of gay marriage in some countries, and associated shifts in the prevalence and visibility of same-sex parenting, might make men’s ability to provide appropriate parenting more evident and bring about change in this type of decision-making. Clearly, although these effects of benevolent sexism might bring some benefits to women, they contribute towards portraying women as weak and restricting them to the domestic sphere.

Effects on heterosexual relationships

The desire to sustain the historical norm of heterosexual relationships between cisgender men and women to raise children was originally proposed as one of the driving forces behind ambivalent sexism^25,135. Accordingly, an impressive body of research now addresses how ambivalent sexism plays out within heterosexual relationships between cisgender men and women¹³⁶.

Some women (and men) might be romantically persuaded by the chivalry inherent in benevolent sexism¹³⁷. Benevolent sexism might play a seductive part in heterosexual women’s initial attraction to men because it promises adoration and willingness to invest by potential male partners¹³⁸. Indeed, women rate benevolently sexist male strangers as more likeable and sexually attractive than hostilely sexist, or even non-sexist, male strangers³¹. This is especially true for women higher in need of security in romantic relationships (for instance, women higher in attachment anxiety)¹³⁹. Women’s benevolent sexism is also associated with preferences for male romantic partners who possess traits more consistent with traditional gender roles, such as the ability to provide status and/or resources^140,141. By contrast, men’s hostile sexism is associated with preferences for female romantic partners who possess traits more consistent with traditional gender roles, such as attractiveness or vitality¹⁴⁰. Among men and women, both benevolent and hostile sexism are associated with greater endorsement of double standards in heterosexual dating (such as the idea that men, not women, should ask for the first date and pay for the date)¹⁴². Thus, both hostile and, particularly, benevolent sexism influence heterosexual cisgender women to pursue more traditional heteronormative partners and potential relationships.

Once in established heterosexual intimate relationships, both men’s and women’s benevolent and hostile sexism can shape the ways in which romantic partners interact and how their relationships function over time. For example, benevolent sexism promotes traditional task divisions for women¹⁴³ and men in heterosexual couples¹⁴⁴. By ostensibly providing women with a sphere of influence (within rather than outside the home), ambivalent sexism tempts women to become complicit in their own subjugation. For example, hostile sexism among mothers is associated with maternal gatekeeping (behaviours that limit or exclude fathers from childcare), which leads to women performing a greater share of childcare tasks and spending more hours on these tasks than men¹⁴⁵. Furthermore, benevolent sexism among women (but not men) is related to intentions to provide dependency-oriented help to male romantic partners when completing stereotypically feminine domestic tasks (such as doing laundry), allowing men to avoid this type of labour in the long run¹⁴⁶. Thus, ambivalent sexism perpetuates broader social inequalities around gender by steering women away from education and careers in favour of a primary caregiving role in relationships and family life¹⁴⁷.

Benevolent sexism can also influence sexual functioning within relationships by focusing the couple on men’s sexual needs and women’s sexual duties^148,149. In heterosexual relationships, women’s hostile and benevolent sexism is associated with greater and lesser frequency of faking orgasms, respectively (potentially indicating that women higher in benevolent sexism place less value upon their own sexual pleasure)¹⁵⁰. Furthermore, exposure to benevolent sexism reduces condom use during sex, partially owing to women’s motivation to have sex to please a male partner rather than for their own pleasure¹⁴⁸. Such behaviours can increase the risk of sexually transmitted infections as well as pregnancy, which can have detrimental health effects and further limit women’s educational and career attainment.

Perhaps owing to differences in social acceptability, benevolent versus hostile sexism from male romantic partners is more prevalent in public versus private contexts, respectively¹⁵¹. However, women higher in benevolent sexism are more likely to accept paternalistic restrictions on their behaviours outside of the home (for example, declining a ‘risky’ educational or career opportunity) at their romantic partner’s behest (particularly when the partner offers a justification that is ostensibly about protecting the woman)³⁴. Importantly, women’s endorsement of benevolent sexism is strongly influenced by perceptions of their male partners’ benevolent sexism^93,152. Thus, being involved in a relationship with a man who holds benevolently sexist attitudes and ideals might tempt women to view benevolent sexism as a manifestation of love and protection rather than sexism and subjugation.

Ambivalent sexism probably leads to a deterioration of relationship quality in heterosexual couples. However, processes by which this might happen can differ for hostile and benevolent sexism and longitudinal research is currently lacking. In general, current evidence suggests that men’s hostile sexism decreases relationship satisfaction for men and women. Indeed, men’s hostile sexism leads to insecurities about women’s independence¹⁵³ and increases conflict¹⁵⁴ and aggression^155,156 in heterosexual relationships, which can lead some women to perceive these behaviours as normative and acceptable in intimate relationships¹⁵⁷. Women’s benevolent sexism can increase their partner’s relationship satisfaction¹⁵⁸, but is associated with shorter relationship length¹⁵⁴. The more women endorse the romanticized relationship ideals linked to benevolent sexism, the more dissatisfied they are with their relationship when the couple faces conflict^159,160. However, women with attachment insecurities can benefit from perceiving that their partner endorses benevolent sexism when there are low levels of conflict because this reassures the women of their partners’ commitment to the relationship¹⁶¹.

Hostile sexism is also associated with negative attitudes towards non-traditional family planning, such as surrogacy¹⁶². However, there is little research on how ambivalent sexism influences minority sexual relationships^163,164. Furthermore, the Ambivalent Sexism Inventory produces different means and item loadings across heterosexual individuals and sexual-minority individuals⁴⁶. Thus, this inventory might not reflect how sexism is experienced by sexual-minority individuals and should not be used to compare groups on the basis of sexual orientation. Future research on the effects of ambivalent sexism on romantic relationships should investigate how these processes might function among individuals of diverse sexual and gender identities.

Summary and future directions

Theoretical and empirical knowledge about ambivalent sexism has improved our understanding of gender inequalities by shedding light on how women are subordinated through the tandem operation of hostile and benevolent sexism. Hostile sexism has more obvious effects, but benevolent sexism is equally damaging and more insidious, largely because it wears a cloak of flattery and protection.

Over the almost three decades of research in this area, there has been little effort to consider the changing and global context in which sexism operates. Future research will need to examine whether these societal shifts have been accompanied by changes in how these forms of sexism are expressed, perceived and experienced (Table 3). For example, it is likely that an increased understanding of how sexism operates has produced reductions in both types of sexism, at least in some places, with benevolent sexism potentially showing a slower decline owing to its positive tone⁴⁰. Ambivalent sexism theory was developed to account for the specific characteristics of gender relationships as they were understood at the time. However, men’s and women’s roles have changed, even if not everywhere¹⁶⁵. For example, more women in the USA occupy high-status positions in employment, or are family breadwinners, in the 2020s than in the 1990s¹⁶⁶. These changes in gender roles can have contradictory effects. For example, they might serve to showcase women’s perceived competence in the work domain and men’s perceived suitability as carers, and increase cooperation between men and women, which could reduce sexism^167,168. However, more egalitarian gender roles might ironically increase gender competition and dominative paternalism to keep women in place and protect the gender hierarchy. The direction of these changes might be influenced by factors such as individuals’ baseline levels of sexism, ultimately leading to a more polarized society (though perhaps with a smaller minority of sexist individuals).

Table 3 Potential ramifications of the shifting global context for ambivalent sexism

The increased awareness and acceptability of non-traditional notions of gender, such as transgender and non-binary gender identities or expressions¹⁶⁹, or of non-traditional families, such as those with same-sex parents, might also influence gender-related processes. Those who endorse hostile sexism might attempt to protect the gender hierarchy by targeting gender-nonconforming individuals and non-traditional families (such as lesbian mothers) and rewarding women who abide by gender norms. However, it is also possible that these non-traditional gender identities and families could contribute to further changes in societal understanding of gender and gender norms. Future research should examine how perceptions and experiences of sexual-minority, transgender and nonbinary individuals might be influenced by the restrictive views of gender communicated and supported by hostile and benevolent sexism, and how these, in turn, might change with increased exposure to gender-nonconforming individuals.

Although research on ambivalent sexism has shed light on how attitudes towards other groups operate (Box 2), more research is needed to understand the intersection between gender and other characteristics, such as age, disability or sexual orientation. For example, little is known about how ambivalent sexism influences wellbeing and relationship functioning in same-sex couples. There is evidence that sexism contributes to intimate partner violence¹⁶³, attitudes towards same-sex parenting¹⁶⁴ and objectification¹⁷⁰ by sexual-minority individuals and within minority sexual relationships. However, these studies used a measure that is now known not to adequately capture sexism in these populations⁴⁶. Indeed, the appropriateness of existing measures of sexism beyond populations that are cisgender, heterosexual, mostly white and living in specific cultures has as yet to be confirmed^171,172. For example, efforts to validate the ambivalent sexism inventory across cultures have revealed that it might need adjustment to capture sexism in those cultures⁴². Future research needs to examine the appropriateness of measures for a range of populations and, if necessary, develop new tools to enable comparative research and better serve these groups.

Despite growing evidence that the intersectionality between gender and race shapes women’s experiences of ambivalent sexism^{41,42,101,173} the majority of research in this area has either not specified the racial composition of the samples or has described them as predominantly white. The findings of this research raise questions about the generalizability of ambivalent sexism theory. More research is needed to clarify whether the theory is less applicable to women of diverse racial groups, whether it can be adjusted and expanded to increase its generalizability, and what measures might be needed to capture sexism across racial or ethnic groups.

More generally, research examining predictors and consequences of ambivalent sexism tend to be restricted to a few cultural contexts, which cannot be regarded as a proxy for the rest of the world. The vast majority of this research fails to acknowledge the cultural context where it is carried out and so does not always reflect on how these contexts influence  the processes uncovered. Although men tend to have more power than women in most societies, the precise cultural and historical context in which gender relationships are lived cannot be ignored. Indeed, there is some evidence that predictors and consequences of sexism can vary across societies as culturally similar as the UK and the USA¹⁷⁴. At the same time, some of the research reviewed here reported similar phenomena across different cultural settings. Ultimately, what is needed is more comparative research to shed further light on the cultural contexts of sexism.

Rapid developments in societal norms and attitudes towards sex, gender and sexuality across many countries in the past few decades^175,176 reflect a global context that is shifting in response to a more intensely interconnected era. These changes are rarely welcomed by everyone and in some cases they are also not permanent. Research needs to more directly examine the effects of these changes, their trajectories across time, and how they influence and are influenced by changes in gender roles and gender-based equality. Socio-political features of this context, such as dominant neoliberal ideology, are likely to influence the ways in which sexism is manifested and entrenched^177,178. It is therefore important to understand the effects of ambivalent sexism and its components^{40,179,180,181} as manifestations and consequences of sexism morph in response to this shifting global context.

Box 2 Implications for other contexts

Sexist beliefs can be recruited to justify prejudice and discrimination towards sexual- and gender -minority individuals. By restricting men and women to particular roles and behaviours, sexism discourages other options¹⁸⁷. It is therefore not surprising that both hostile and benevolent sexism have been associated with negative attitudes towards homosexuality^188,189,190 and parenting by same-sex couples⁴⁵ in countries such as England, Turkey and the USA. In addition, the view that (specific types of) cisgender women are vulnerable and in need of protection from and by men has been core to arguments against gender-neutral public bathrooms⁶⁸. These sexist views simultaneously restrict and subjugate women, sexual minority individuals and transgender people, protecting the status quo and heteronormative male dominance.

Insights regarding ambivalent sexism have inspired research into other types of prejudice, which also have ambivalent characteristics that often combine hostility and benevolence, typically involving paternalism. For example, classism^135,191, ageism¹⁹² and ableism¹⁹³ are all rooted in stereotypes of warmth and incompetence in groups characterized by low status and cooperative relationships with more powerful groups¹⁹⁴. Just as with sexism, ambivalence leads to polarizing attitudes towards members of these groups, with some offered benevolence (for example, those with visible disabilities, who are perceived as dependent) and others being targets of hostility (such as those with less apparent disabilities, who are perceived as wanting unwarranted special privileges)¹⁹³. This conceptualization has further implications: for example, it emphasizes the importance of clarifying where the line might lie between paternalism and legitimate help. Indeed, although some individuals might legitimately require a level of care or assistance (for example, individuals with memory impairments), where help goes beyond the required level it can easily become paternalistic, which, by discouraging agency, can accelerate the individual’s deterioration.