Introduction

Ambivalent Sexism Theory states that attitudes toward women are characterised by two fundamental dimensions: one reflects Benevolent Sexism (BS), the other Hostile Sexism (HS; Glick and Fiske 1996). HS is aggressive in tone and expresses coercive and overtly negative evaluations of women who are seen as competing with men or seen as stepping outside of traditional (patriarchal) gender roles. In contrast, BS is an ideology which has a deceptively positive appearance. BS expresses attitudes which prescribe that men cherish women and protect and care for them (Glick and Fiske 1996). The catch in this archetype is that women are only deemed to deserve men’s protection and adoration if they conform to traditional gender roles.

One factor contributing to the ubiquity of BS ideology is that it seems to increase life satisfaction, despite positioning women as weaker than men. Recent research by Napier et al. (2010) identified a positive association between endorsement of BS and life satisfaction in a sample of men and women from 32 countries. This effect is particularly intriguing because it was most pronounced in more egalitarian nations. It remains an open question, however, as to whether or not the positive relationship between BS and life satisfaction occurs through identical mechanisms for men and women. The present research seeks to answer this question utilising a large and representative national sample of New Zealanders. We argue that for women, the association between BS and life satisfaction should be accounted for by the palliative effects of rationalising the status quo. That is, for women, but not necessarily for men, the link between BS and life satisfaction should occur only because BS justifies existing gender inequality.

A cornerstone of Ambivalent Sexism theory is that BS and HS function together as a joint system which maintains gender inequality. This system offers positive evaluations of women as a reward for conforming to traditional gender roles, while at the same time offering negative evaluations of women who do not conform. Protection, security and resources are given to women at the individual level, while women’s resistance to inequality is disarmed at the societal level (Glick and Fiske 1996; Jackman 1994). For instance, longitudinal research conducted in New Zealand indicates that women who endorse BS tend to find HS more palatable over time (Sibley et al. 2007). Napier et al.’s (2010) study provides a novel line of evidence suggesting that one reason behind the pervasiveness of BS is that people who believe women are warm, wonderful and in need of protection from men are simply happier. BS predicts positive outcomes for individuals (Napier et al. 2010), whereas accumulated across society, BS predicts endorsement of more aggressive ideologies that control and constrain women’s social roles as a group (Glick and Fiske 2001b; Glick et al. 2000; Glick et al. 2004).

Here, we present a Differential Process Model, in which we argue that there should be key gender differences in the rationalization that mediates, or act as the mechanism, through which BS predicts life satisfaction. As outlined in Fig. 1 we propose that, for women, BS should predict an increase in gender-specific system justification. That is, endorsing BS should cause women to view gender relations as more fair and balanced. We argue that this effect occurs for women more so than men. This should be the case because benevolently sexist women should be more motivated to reduce the dissonance associated with viewing the position of their gender ingroup as in need of protection from men (i.e., an imbalance in gender relations). As detailed later, we expect system justification to mediate the relationship between BS and life satisfaction for women, as indicated by the processes in Napier and Jost’s (2008) model. Endorsing system justifying beliefs which frame gender inequality as fair and deserved should decrease women’s ideological dissonance and thus increase women’s life satisfaction. This process should be most apparent in highly egalitarian nations, like New Zealand where our data were collected, because it is in such nations that BS becomes more necessary as a method of ideological control (Glick et al. 2004; Napier et al. 2010).

Fig. 1
figure 1

A Differential Process Model of Benevolent Sexism and Life Satisfaction for Men and Women. (Note. The model depicts the hypothesised mechanisms by which BS increases life satisfaction. For women, this positive effect is theorised to occur indirectly as a result of justifying gender inequality. For men, BS is theorised to directly increase life satisfaction as it frames their higher status as a deserved responsibility. HS is not depicted in this model. Following Napier et al. 2010, HS is hypothesised to magnify the direct association between BS and life satisfaction when controlling for gender.)

Full size image

As also shown in Fig. 1 we argue that, for men, endorsing BS positions the gender outgroup, women, as weaker and needing men’s protection. As opposed to the effect for women, this should not elicit dissonance as it does not have negative implications for evaluations of the ingroup’s (men’s) efficacy and competence. Rather, we argue that maintaining the attitude that women are warm but fragile should complement men’s perceptions of their social role as protectors and providers. This is an ideology that suits men directly because it legitimizes the gender imbalance in access to status, resources and power. Thus, for men, we expect that BS should correlate directly with increased life satisfaction without the intermediary dissonance-reducing requirement of viewing the system as fair and stable. In sum, we argue that both men and women find ‘joy’ in endorsing BS but for women this increase in life satisfaction occurs through the intermediary mechanism of gender-specific system-justification. We will test these predictions using a series of regression models assessing the links between gender, BS, HS, the resulting gender by sexism factorial interactions, system justification and life satisfaction. The following sections outline previous findings and the rationale for our proposed Differential Process Model.

Ambivalent Sexism as Ambivalent Justification

At their core, the ideologies which constitute ambivalent sexism serve to legitimise gender differences. BS justifies the power advantage of men by promoting the message that women are in need of men’s affection and protection (Glick and Fiske 1996; Jackman 1994). That is, gender inequality is framed instead as men’s rightful responsibility to care for women. This process of rationalising societal inequality is known as system justification. System justification is motivated by the desire to attain social stability and maintain the status quo (Jost and Banaji 1994). One hypothesis made by System Justification Theory is that disadvantaged groups will legitimise the power of high-status groups by internalising beliefs that justify inequality (Jost and Hunyady 2002).

Our Differential Process Model integrates Ambivalent Sexism Theory with System Justification Theory and thus takes the perspective that BS and HS facilitate the dissonance-driven motivation to rationalise men’s power. In disadvantaged groups, such as women, the motivation to legitimise societal systems sometimes takes priority when the needs of the self and the needs of the ingroup are not salient (Jost et al. 2003). Despite the detrimental effects inherent in maintaining an unequal status quo, disadvantaged groups often justify the system significantly more than advantaged groups (Jost et al. 2003; for a review see Jost et al. 2004). Thus System Justification Theory makes the seemingly counterintuitive prediction that the primary means by which people reduce ideological dissonance in the face of an unequal system will be to further endorse that system.

The prediction made by System Justification Theory is reflected in cross-cultural research on sexism. Women tend to endorse BS and HS more as societal gender inequality increases (Glick et al. 2000). Glick et al.’s (2000) comparative study of 19 nations reported that levels of sexism predicted gender inequality at a national level and, moreover, in each country the average scores of men’s ambivalent sexism predicted women’s level of sexism. In countries with the highest gender inequality, BS was endorsed more by women than by men. Glick et al. (2000) argued that women accede to BS in these nations because it provides protection from the threats expressed by HS. That is, women endorse benevolent prejudice toward their ingroup as a defence against the prevalence of HS in society.

System Justification Theory predicts that dominant patriarchal ideologies should influence the views of women even in highly egalitarian nations. In nations with high gender equality, however, women should be less likely to endorse BS as “self-defence” against the threat of HS (Glick et al. 2000, p. 773). Recent research suggests that the adoption of BS in these nations is prompted by perceptions of societal norms rather than a protection-racket motivation. Kay et al.’s (2009, Study 3) results from a Canadian sample indicated that increasing women’s need for system justification increased their beliefs in the legitimacy of generally accepted norms: those who were told there were more women in parliament deemed women to be more desirable and ideal politicians, while those who were told that men were the majority believed that men were more ideal politicians. Furthermore, Sibley et al. (2009) studied the longitudinal influence of perceived normative sexism on New Zealand undergraduates. They reported that, for both men and women, the perception of men’s BS at a societal level was the principal referent for determining their own BS. That is, women in more egalitarian nations ‘bought into’ the dominant ideology of BS (which predicted increases in their HS over time). Thus BS maintains the gender imbalance by enticing women with a positive justification for inequality which is seen to be a generally accepted societal norm.

One such subjectively positive and rationalising quality of BS is its perpetuation of complementary role stereotypes, that is, stereotypes which emphasise warmth and morality as traits which are exclusive to women and thereby compensate for their lower status in society. Jost and Kay (2005) exposed undergraduates in the United States to these complementary statements. Statements which emphasised the superiority of women’s warmth and morality when compared with men increased women’s support for the current system of gender relations, and moreover, increased their support for society as a whole. The increase in gender-specific system justification did not however occur for men (Jost and Kay 2005). The system justification of men was consistently high regardless of exposure to complementary stereotypes. These findings highlight the crucial role which BS plays in persuading women to endorse the established order of gender relations.

Sexism with Benefits

As argued in the preceding section, those who take a benevolently sexist perspective on gender relations tend to perceive gender inequality in economic domains to be fair and legitimate (Jost and Kay 2005). Our model predicts that this increase in gender-specific system justification should in turn increase life satisfaction. In accordance with System Justification Theory, the reduction of ideological dissonance (i.e., increased belief in the status quo as fair and stable) leads to a decrease in negative affect and an increase in positive affect (Jost and Hunyady 2002). Thus we expected that women will cognitively buffer the stress of ideological dissonance (i.e. existence of gender inequality in a supposedly egalitarian society) through instilling a sense of stability and fairness into the social system. One such way that this is performed is through endorsing sexist justifications of existing gender inequality, beliefs which are positively associated with life satisfaction (Napier et al. 2010).

In integrating Jost and Kay’s (2005) research with Napier et al.’s (2010) research, our Differential Process Model posits that gender-specific system justification should mediate the relationship between BS and life satisfaction. Previous research conducted at both the intergroup level and the interpersonal level supports this hypothesis. Firstly, Napier and Jost’s (2008) model explains the higher subjective well-being of conservatives in comparison to that of liberals. The happiness of those who follow a right-wing orientation was due to their higher belief in meritocracy (Napier and Jost 2008). That is, beliefs such as ‘hard work brings people what they deserve’ justified the presence of economic inequality. Napier and Jost’s (2008, Study 1) model showed that system justification mediated the relationship between conservatism and subjective well-being. As economic inequality increased in the United States from 1974 to 2004, system justification moderated the effect of inequality on well-being (Napier and Jost 2008, Study 2). They reported that over this period, the well-being of conservatives decreased to a lesser extent than that of liberals. That is, the system-justifying beliefs of conservatism provided a buffer against the detrimental effects of ideological dissonance.

System justification also functions as a buffer against inequality in other domains. Within interpersonal relationships, women’s endorsement of a traditionalist ideology cushions the detrimental effects of gender inequality and thus predicts increased marital satisfaction. In an innovative study pertinent to this area, Greenstein (1996) reported that traditionalist housewives (i.e., those who believed that men should be the ‘breadwinner’ of the household) perceived unequal divisions of domestic labour to be fairer than more egalitarian housewives. Moreover, the fairness of this division was less related to marital satisfaction for traditionalist housewives than for more egalitarian housewives (Greenstein 1996). Extending this line of research, Lavee and Katz (2002) sampled four Israeli communities and reported that the relationship between the division of labour and marital satisfaction was mediated by perceptions of fairness. This mediation only occurred for women; men’s perceptions of fairness was unrelated to the segregation of domestic labour. Although Lavee and Katz (2002) explained these results in terms of constructed gender roles and the cultural expectations of women, this study is highly consistent with the Differential Process Model we formulate here. Our model details the specific cognitive processes attached to the “traditional” ideology and “perceptions of fairness” reported in Lavee and Katz (2002): BS should facilitate the rationalisation of gender inequality and thus characterise gender relations as fair, a belief which should predict increased life satisfaction. We also argue that the satisfaction of women is contingent on this rationalisation providing a buffer against the negative effects of gender inequality. In comparison, men’s life satisfaction should increase without the need to justify gender inequality.

Other Important Factors

We aim to model the gender-distinct paths which link BS with life satisfaction. This section examines Hostile Sexism (HS) and socio-economic deprivation, two critical factors implicated in both system justification and life satisfaction. These variables are included as controls in our models in order to rule out their possible confounding effects. Firstly, HS typifies outwardly negative attitudes toward women who resist traditional gender roles. Previous research has identified two opposing effects of endorsing HS: one reflects that the attitudes it expresses are in discordance with egalitarian values, the other reflects its tandem role with BS as a buffer against inequality. Napier et al. (2010) reported that, for men, the endorsement of hostile justifications for gender inequality had a negative main effect on life satisfaction. Moreover, in egalitarian nations, the endorsement of purely hostile justifications was related to lower life satisfaction.

However, HS appeared to also function as buffer. In more egalitarian nations, the life satisfaction of those who endorsed both benevolent and hostile justifications of gender inequality was no different to the life satisfaction of those who did not endorse a sexist justification (Napier et al. 2010). In sum, we expect there to be a negative relationship between the endorsement of HS and life satisfaction. However, we also expect that HS should act as a moderator and generally increase the positive relationship between BS and life satisfaction. This should occur in addition to the separate indirect effects of system justification for men and women proposed in our model. Given our aim to model the paths for each gender, we control for the differing effects of HS in men and women. Thus our regression models include the gender x HS interaction term as well as the three-way interaction between gender, BS and HS.

The second critical factor which we expect to affect both system justification and life satisfaction is socioeconomic deprivation. According to System Justification Theory, groups with low socioeconomic status have a stronger motive to legitimise the system and reduce ideological dissonance (Jost et al. 2004). Thus, impoverished groups sometimes endorse system-justifying ideologies more strongly than high socioeconomic groups (Jost et al. 2004). Furthermore, socioeconomic deprivation should predict lower satisfaction with life. Factors such as income, education and employment are positively associated with well-being and life satisfaction (Diener et al. 1999). Our model includes a measure taken at the neighbourhood level. The 2006 New Zealand Index of Deprivation accounts for features of regional deprivation such as income, qualifications and employment status (Salmond et al. 2007).

Overview of Hypotheses

Consistent with the general palliative function of system justification proposed by Napier and Jost (2008), our Differential Process Model states that gender-specific system justification will mediate the increase in life satisfaction associated with endorsing BS. Secondly, we expect that gender will moderate both the direct effect and the indirect effect of BS on life satisfaction. Jost and Kay (2005) reported that BS increases gender-specific system justification in women. Our model thus proposes that the increase in benevolently sexist women’s life satisfaction should be explained by gender-specific system justification, while for men there should be a direct relationship between BS and life satisfaction (Hypothesis 1). Finally, Napier et al. (2010) reported that those in highly egalitarian nations who endorsed both BS and HS were as satisfied as those who did not endorse any form of sexism. Our Differential Process Model is consistent with this feature of Napier et al.’s (2010) model and proposes that HS should moderate the satisfaction-enhancing effect of BS when controlling for the differential effects of gender. We therefore expect the complementary ideology of HS to magnify the direct association between BS and life satisfaction (Hypothesis 2).

Method

Sampling Procedure

This study analyzed data from the 2009 New Zealand Attitudes and Values Study (NZAVS-2009). The NZAVS-2009 questionnaire was posted to 40,500 participants from the 2009 NZ electoral roll (see Reid and Sibley 2009, for technical details). The publicly available version of the 2009 electoral roll contained 2,986,546 registered voters in NZ. This represented all citizens over 18 years of age who were eligible to vote regardless of whether or not they chose to vote, barring people who had their contact details removed due to specific case-by-case concerns about privacy. In sum, roughly 1.36% of all people registered to vote in New Zealand were contacted and invited to participate. The overall estimated response rate (adjusting for address accuracy of the electoral roll and including anonymous responses) was 16.6%.

Participant Details

The NZAVS-2009 sampled a total of 6,507 participants (3864 women, 2640 men, 3 unreported). Complete data for the measures analysed in this study were available for 6,100 participants (94%; 3684 women, 2416 men). The large majority of missing data was due to unreported addresses, which meant we could not match these people to area-unit information about regional deprivation. Of those for which complete data were available, 71.6% were New Zealand European (n = 4,370), 17.7% of the sample were Māori (n = 1,077), 3.5% were of Pacific Nations ancestry (n = 211), 4.5% were of Asian ancestry (n = 272), and 2.8% were coded as ‘other’ (n = 170). The average age of participants was 47.74 (SD = 15.73).

Materials

BS and HS were measured using shortened five-item scales from Glick and Fiske’s (1996) Ambivalent Sexism Inventory (ASI). BS was measured using items 8, 9, 12, 19, and 22 as listed in the ASI (e.g. “Every man ought to have a woman whom he adores.”) HS was assessed using items 5, 11, 14, 15 and 16 as listed in the ASI (e.g. “Women are too easily offended.”) Items were rated on a scale ranging from 1 (strongly disagree) to 7 (strongly agree) and were averaged to create scale scores. Higher scale scores indicated greater levels of BS (α = .72) or HS (α = .81).

Gender-specific system justification was measured using a single item taken from Jost and Kay’s (2005) gender-specific system justification items. The item referred specifically to perceptions of fairness in the relations between men and women in New Zealand: “Men and women both have a fair shot at wealth and happiness in NZ.” The item was rated on a scale ranging from 1 (strongly disagree) to 7 (strongly agree). A higher score indicated higher gender-specific system justification.

Life satisfaction was measured using two of the five items from the Satisfaction with Life Scale (SWLS; Diener et al. 1985). These items were “I am satisfied with my life” and “In most ways my life is close to ideal.” Both items were rated on a scale ranging from 1 (strongly disagree) to 7 (strongly agree). Responses were positively correlated (r (6346) = .62, p < .01) and were averaged such that a higher scale score indicated higher life satisfaction.

The NZAVS-2009 assessed several other variables. Age and level of deprivation were explicitly controlled for in the analyses. To create area-unit levels of deprivation, participants’ addresses were matched to scores catalogued in the New Zealand Index of Deprivation 2006 (NZDep2006; Salmond et al. 2007). The NZDep2006 assigns relative deprivation scores to area-units (with a median population of 87). Scores range from 1 to 10, with higher values indicating more deprivation. The index comprises nine items which assess several aspects of socioeconomic deprivation (receipt of government benefits, income below an income threshold, ownership of a home, single parent families, employment, qualifications, number of occupants per bedroom, access to a telephone and access to a car) at a neighbourhood level.

The NZAVS survey took approximately 45–60 min to complete. A $500 grocery voucher prize draw was offered as an incentive to complete the study.

Results

Overview of Model

Descriptive statistics and correlations between all variables are presented in Table 1. Interaction terms were created by calculating the products of (centered) BS x HS, gender x BS, gender x HS and the three-way interaction term of gender x BS x HS.

Table 1 Descriptive statistics and correlations between BS, HS, system justification, life satisfaction, age and level of deprivation in a postal sample of people randomly selected from the New Zealand electoral roll
Full size table

We tested a moderated mediation model (or a conditional indirect effect model) using the procedures outlined by Preacher et al. (2007). This model tested the conditional indirect effect of BS on life satisfaction mediated by gender-specific system justification and moderated by gender. Our model proposed that BS should predict an increase in gender-specific system justification, which in turn should predict an increase in life satisfaction. However, gender should simultaneously moderate the extent to which BS predicts the mediator. In other words, the extent to which BS indirectly predicts life satisfaction should be dependent on both an individual’s gender and their endorsement of HS.

We first entered the gender x BS interaction term into a regression predicting system justification, controlling for the effects of Hostile Sexism and other interaction effects. The analyses also controlled for age and level of regional deprivation. Regression coefficients are displayed in Table 2 (Model 1). The model met the prerequisites for mediated moderation through the significance of the gender x BS interaction term (β = .05, t = 2.03, p < . 05). This indicated that gender was a moderator of the relationship between BS and system justification. That is, BS had different effects on gender-specific system justification for men and women. Diagnostic tests indicated the model was not unduly biased by mulicollinearity, as tolerance statistics for all variables in the model were well above the minimum value of .20 as recommended by Menard (1995).

Table 2 Regression coefficients testing the effects of Benevolent Sexism, Hostile Sexism, gender (men were dummy coded as 0; women were dummy coded as 1), age, level of deprivation and all interaction terms (BS x HS, Gender x BS, Gender x HS, Gender x BS x HS) on the hypothesised mediator of system justification (Model 1). The second model tested the effects of BS on life satisfaction including the variables from the first model with the addition of the hypothesised mediator, system justification, as a predictor (Model 2)
Full size table

We then conducted a second regression model in which all variables predicted life satisfaction. This step included the hypothesised mediator, system justification, as a predictor. The coefficients from Model 2 are presented in Table 2. System justification was a strong predictor of increased life satisfaction (β = .21, t = 16.88, p < .001). This indicated that rationalising the current state of gender relations provided a palliative effect on well-being. Again, diagnostic tests indicated the model was not unduly biased by mulicollinearity, as tolerance statistics for all variables in the model were well above the minimum value of .20.

In Model 2, the gender x BS interaction was a significant predictor of life satisfaction (β = −.13, t = −4.27, p < .001), indicating that the direct association between BS and life satisfaction differed for men and women. Moreover, the BS x HS interaction was significant (β = .03, t = 2.45, p < .05). This suggested that HS moderated the extent to which BS had a direct influence on life satisfaction. The three-way interaction of Gender x BS x HS was not significant. That is, there was no difference between men and women in terms of how endorsing complementary sexist ideologies affected life satisfaction. This finding replicated the results of Napier et al. (2010). This three-way interaction was nevertheless retained in the model to fully control for its effects.

Hypothesis 1

The simple slopes obtained from the gender moderation are shown in the first part of Table 3. The indirect effects of both moderators were calculated simultaneously, thus the effects displayed are those that occurred controlling for the moderating effect of HS. For men, the conditional indirect effect of BS on system justification was significant (slope = .015, se = .005, z = 2.93, p < .001). The simple slope of the conditional indirect effect for women also significant and stronger than that observed for men (slope = .028, se = .005, z = 5.81, p < .001). This suggested that both men and women experienced increased life satisfaction due to BS increasing perceptions of fairness in the relations between men and women in NZ.

Table 3 Conditional indirect effects and conditional direct effects of gender as a moderator of the positive relationship between Benevolent Sexism and life satisfaction mediated by system justification
Full size table

The simple slopes for the conditional direct effect of BS on system justification are displayed in Table 3. There was a notable difference between men and women in terms of the direct effect. As BS increased in men, so did their life satisfaction (slope = .152, se = .023, z = 6.72, p < .001). Conversely, the direct effect for women did not significantly differ from zero (slope = .028, se = .019, z = .1.48, p = .14). For women, the relationship between BS and life satisfaction was instead completely mediated by system justification. In other words, endorsing BS was positively related with women’s life satisfaction only because of the palliative effect of rationalising gender inequality. The positive relationship between BS and life satisfaction for men, in contrast, was both direct and indirect, thus the partial mediation reported in Table 2 occurred only for men. Figure 2 displays our model of the theorized effect of BS on life satisfaction.

Fig. 2
figure 2

A Differential Process Model of Benevolent Sexism and life satisfaction for men and women including simple slope coefficients. (Note. For women, gender-specific system justification completely mediated the relationship between BS and life satisfaction. For men, BS was positively related to life satisfaction both directly and through the mediator of system-justification. * p < .05, non-significant paths shown by greyed lines.)

Full size image

Hypothesis 2

We calculated the simple slopes for HS as a moderator of the direct association between BS and life satisfaction, controlling for the gender moderation. Simple slopes for the conditional direct effect were calculated at low (−1 SD) and high (+1 SD) levels of HS. The direct effect was significant at low levels of HS (slope = .11, se = .03, z = 3.48, p < .001). This moderating effect indicated that HS magnified association between BS and life satisfaction (slope = .20, se = .03, z = 7.39, p < .001). In sum, endorsing the ‘complementary’ ideology of HS had the hypothesized effect of enhancing the positive relationship between BS and life satisfaction. Consistent with Napier et al. (2010), the endorsement of HS directly predicted a decrease in life satisfaction (β = −.15, t = −7.21, p < .001).

In sum, regression analyses supported our hypothesis that the relationship between BS and life satisfaction would be mediated by gender-specific system justification. The extent to which system justification acted as a mediator was moderated by gender. For women, gender-specific system justification fully mediated life satisfaction. As depicted in Fig. 2, women’s endorsement of BS exclusively predicted an increase in gender-specific system justification, which in turn predicted an increase in life satisfaction. For men, endorsement of BS also predicted a direct increase in life satisfaction regardless of system justification.

Discussion

We proposed and tested a Differential Process Model explaining why benevolently sexist men and women were happier than non-sexists. Consistent with the earlier findings reported by Napier et al. (2010), endorsement of BS was associated with a concurrent increase in life satisfaction. Extending the innovative work of Napier et al. (2010), our model indicated that the links between BS and life satisfaction differed for men and women. BS predicted a concurrent increase in life satisfaction for women via the intermediary mechanism of gender-specific system justification. Our model thus indicated that endorsing BS led women to be more satisfied with life because benevolently sexist attitudes rationalise existing gender inequality. In comparison, BS increased men’s life satisfaction without the intermediary requirement to justify gender inequality. Furthermore, controlling for gender-specific system justification, our results also supported the earlier finding by Napier et al. (2010) that the positive association between BS and life satisfaction was enhanced by the simultaneous endorsement of HS. That is, the positive relationship between BS and life satisfaction was enhanced for ambivalently sexist men and women compared to those who exclusively endorsed BS. These effects were subtle. The effect sizes we observed were fairly small, but they were reliable, and as such help to provide a more complete picture of the different mechanisms through with sexism operates to legitimize gender inequality by making BS palatable to women at the individual level.

Sexism, System Justification and Life Satisfaction

Men who endorsed BS generally experienced greater life satisfaction than those who rejected BS. Our results showed that the relationship between men’s BS and their life satisfaction was partially due to the palliative effects of system justification. This evidence is consistent with the proposal of Wakslak et al. (2007) that members of advantaged groups should experience increased life satisfaction through rationalising the status quo. Justifying the system in this way reduces the guilt of occupying a higher status (Wakslak et al. 2007). This notion reaffirms Glick and Fiske’s (2001a, p. 111) argument that men who view their power as a “burden gallantly assumed” justify the gender imbalance in the availability of that power, and thus retain positive self-images.

The principal focus of the current research was on modelling how women endorse systems which sustain gender inequality, however. Our model states that one consequence of adopting BS is that it provides a rationalisation of the status quo and thus reduces ideological dissonance. That is, BS provides women a buffer against the detrimental effects of gender inequality. Our model integrates two lines of research to generate this premise. The first of which, such as the experimental research by Jost and Kay (2005) and Kay et al. (2009), has indicated that a specific operation of BS is to legitimise the status quo of gender relations. The second line of research is that of the positive effect of system justification on well-being, the “palliative effect” noted by Jost and Hunyady (2002) and Napier and Jost (2008). In sum, the belief that men’s greater power and access to resources should be utilised to protect women provides a justification for gender inequality in society. Those who subscribe to this belief are thus more likely to believe that men and women have equal opportunities to gain wealth and happiness, and in turn report greater satisfaction with their lives.

The critical difference between men and women was that, for men, the association between BS and life satisfaction occurred at least partially without the intermediary requirement to view gender relations as fair. Our results therefore suggest that men who endorse sexist ideologies benefit over and above the positive effects of rationalising inequality. The exact nature of this relationship remains unclear, however. We suggest that the positive direct effect of BS on men’s life satisfaction may be attributed to markers of status, resources and power. It is also important to note that the measure of life satisfaction utilised in our model reflects a global evaluation of one’s standard of living, rather than constructs such as emotion and esteem (Diener et al. 1985). Recent research by Kahneman and Deaton (2010) tested the distinction between life satisfaction and emotional well-being in a national sample of the United States. Life satisfaction is uniquely and strongly associated with socioeconomic status, particularly income and education (Kahneman and Deaton 2010). As argued previously, the function of sexist ideologies is to reinforce men’s access to powerful ‘executive’ positions in society, men’s role as breadwinners in the familial domain, and men’s leadership in political systems. It follows that men who believe they have an advantage in gaining such markers of status, resources and power will be more satisfied with their lives.

The argument above is consistent with an intergroup dominance perspective. The material gains of endorsing sexism explain men’s greater life satisfaction. This explanation is consistent with the two most prominent theories of individual differences in prejudice: Right-Wing Authoritarianism (Altemeyer 1981) and Social Dominance Orientation (Pratto et al. 1994). Respectively, these reflect the attitudinal motivation toward attaining ingroup-cohesion and the attitudinal motivation toward attaining intergroup-dominance. The relationship between cohesion and dominance is complicated in gender relations because these motivations apply to subtypes of the same group (Sibley and Wilson 2004). Men should desire to maintain their high status, yet should also require women to meet interpersonal needs, such as those of love and affection, but also in some cases status-marker needs signalling men’s status to other men through the presentation of an attractive female partner (Sibley and Overall 2011).

Sexist ideologies reflect the dynamic between social cohesion and intergroup dominance (Sibley et al. 2007; Sibley et al. 2007). In tandem, BS and HS prescribe that women should be praised for adopting traditional gender roles (i.e., domestic, nurturing and warm) roles, and derogated for rejecting patriarchal gender roles (i.e., challenging dominance). Longitudinal research conducted in New Zealand has reported that Right-Wing Authoritarianism predicted increases in BS, while Social Dominance Orientation predicted increases in HS (Sibley et al. 2007). From this perspective, men who endorse BS, and particularly those who additionally endorse HS, maintain and legitimise their greater access to resources and high status positions. Our results support this argument. BS was directly associated with concurrent life satisfaction in men. The simultaneous endorsement of HS enhanced the association between BS and life satisfaction. Future research should examine the role of men’s endorsement of BS and HS on their ambition to acquire power, status and resources.

The relationship between HS and life satisfaction follows two opposing paths. The total effect reflects the combination of both positive and negative effects. HS predicted increased life satisfaction through magnifying the direct association between BS and life satisfaction, which occurred only for men. That is, benevolently sexist men who held antagonistic attitudes toward deviant women experienced a stronger direct increase in life satisfaction compared to men who exclusively endorsed BS. However, the endorsement of HS also predicted decreased life satisfaction. Our regression model, replicating the results of Napier et al. (2010), reported a negative main effect of HS on life satisfaction, which remained significant when controlling for the effects of BS and relevant interactions. It follows that, in more egalitarian nations, aggressive attitudes towards women are not tenable and cause ideological dissonance. Antagonistic views of non-conformist women are not compatible in societies which praise egalitarianism (Glick et al. 2000; Schwartz and Rubel-Lifschitz 2009). Thus the enhancing function of HS on the positive association of BS with life satisfaction was mitigated by the simultaneous negative direct association between HS and life satisfaction.

The Dark Side of Gender-Specific System Justification

One possible implication of our model is that endorsing BS is an adaptive response to the presence of gender inequality in an egalitarian society. Women justify their lower status by believing that they need to be provided for by a male breadwinner, and are thereby more satisfied with life. However, the endorsement of BS has several negative effects on women. Exposure to statements which subtly express BS have been shown to increase women’s perceptions of themselves as incompetent (Barreto and Ellemers 2005). Exposure to such statements also increases women’s memories of themselves as incompetent (Dumont et al. 2010). Furthermore, women high in BS are more open to accepting restrictions (justified through positive ‘protection’ terms) prescribed by their male partners (Moya et al. 2007). BS is thus associated with women’s levels of life satisfaction, while at the same time operating in various ways to disarm resistance to inequality.

Our findings should be taken in light of previous research which has identified characteristics which mask the malevolent nature of BS (Overall et al. in press; Sibley and Perry 2010). The increase in life satisfaction is one such quality which makes the ideology particularly palatable. BS is not typically categorised as a ‘sexist’ or ‘prejudicial’ set of attitudes (Barreto and Ellemers 2005). In fact BS presents ideals which are often viewed as appealing. On average, women rate descriptions of benevolently sexist men as more attractive than men described as hostilely sexist, and in some cases as more attractive than non-sexists (Bohner et al. 2010; Killianski and Rudman 1998). The tempting qualities of BS are however bundled with harmful consequences, which seem to be experienced exclusively for women. There are constraints on acceptable gender roles for women who conform to the patriarchal ideal. These women must fit the stereotype of high warmth and low competence in order to gain praise and protection.

The position of high-status groups is best maintained, not by the dominant group justifying the hierarchy, but by having subordinate groups internalise the ideologies of the dominant group (Jackman 1994). Thus the aim of men (as a dominant group) is to preserve the status quo; the function of BS is to make this process appealing to women (the subordinate group). It is the particular nature of ideology to work from the top-down, prescribing beliefs and standards at the societal level which become assimilated at the individual level. We emphasise this distinction between societal consequences and individual consequences because this balance is central to the propagation of sexism. At the individual level, women who endorse BS are more satisfied with life than those who do not. At the societal level, however, the very same beliefs rationalise the existence of gender inequality. In sum, the increase in women’s life satisfaction should be viewed as another extension of the function of BS: to constrain the role of women in society.

Caveats, Future Directions, and Conclusions

Central to Napier et al.’s (2010) findings was that the association of BS and life satisfaction was dependent on gender equality at a national level. The sample analysed in this study was taken from a large representative sample of New Zealanders registered on the electoral roll. Thus our model should apply to highly egalitarian nations. The attitudes expressed by BS prescribe women to be wonderful but fragile. It follows that these attitudes will be perceived differently in nations where women are seen as less valuable and useful than men. Egalitarian societies, in contrast, idealise the values of equality and social justice (Schwartz and Rubel-Lifschitz 2009). Napier et al. (2010) reported that in nations with high gender inequality, the endorsement of BS was still positively related to life satisfaction (provided that HS was not also endorsed). A direction for further research is to examine the conditional processes by which BS predicts life satisfaction in nations with high gender inequality.

A limitation of our model was that gender-specific system justification was indexed by a single item. It is possible that the item may not have directly assessed a social comparison between men and women, although we argue that the item reliably assessed the current situation of gender equality in New Zealand. It was not correlated with levels of deprivation, as would be expected from a general measure of system justification. Nevertheless, future studies should consider utilising the other items described in Jost and Kay (2005) and constructing scale scores indexing gender-specific system justification to give a more reliable index of this ideology.

One intriguing aspect of our results was the relationship between both benevolent and hostile forms of sexism and our measure of levels of deprivation. Table 1 shows that for both men and women, as general levels of deprivation increased, so too did levels of BS and HS. These correlations indicate that ambivalent sexist ideology is generally endorsed more strongly by those who live in areas with a lower socio-economic status. This mirrors cross-cultural research showing that countries which tend to have lower Human Development Index scores and lower Gender Equality Measure scores are those which also report higher BS and HS (see Glick et al. 2000; Glick et al. 2004; Napier et al. 2010; United Nations Development Programme 2009). This suggests that deprivation and empowerment in one’s neighbourhood may operate in the same direction as more global national-level differences in (low) human development and (low) gender empowerment to predict higher levels of sexism.

Finally, caution should be taken in regard to inferring causality based solely on our data. Our results were based on a large cross-sectional national sample and thus cannot be used to test causal effects. For instance, it is possible that those who are happier with their lives tend to adopt BS. Research has indicated that BS appears to be attractive and subjectively positive; it may therefore be a more congruous ideology to people who are more satisfied with their lives. The assumptions for our proposed model were, however, based on experimental data which has shown that exposure to benevolently sexist statements increases gender-specific system justification (Jost and Kay 2005). In addition, empirical evidence supports the notion that an increase in system justification leads to an increase in life satisfaction (Jost and Hunyady 2002; Jost et al., 2008). Moreover, previous research which has examined the processes described in our model has generally constructed the pathways from ideology to satisfaction (e.g. Lavee and Katz 2002; Napier and Jost 2008; Napier et al. 2010). Nevertheless, the most effective resolution of the question of causality is for future research to examine our proposed model utilising longitudinal data.

In conclusion, we presented a model testing the positive relationship between the endorsement of BS and life satisfaction in the egalitarian nation of New Zealand. The model extends the insightful work of Napier et al. (2010) to propose that women who endorse BS gain life satisfaction through rationalising gender inequality (and thereby reducing ideological dissonance). Our Differential Process Model theorizes that one central reason behind the persistence of BS at a societal level is that, at an individual level, it reduces motivations to change the status quo for both genders by making people happier and more satisfied with their lives. The process by which this occurs differs for men and women, however. For men, BS was directly associated with life satisfaction. For women, the palliative effect of BS was indirect and occurred because BS-ideology positioning women as deserving of men’s adoration and protection was linked to general perceptions of gender relations as fair and equitable, which in turn predicted greater levels of life satisfaction.