Men’s Hostile Sexism and Biased Perceptions of Intimate Partners

Abstract

Hostile sexism (HS) expresses attitudes that characterize women who challenge men’s power as manipulative and subversive. Does endorsing HS negatively bias perceptions of women’s behavior and, in turn, create animosity within intimate relationships? Committed heterosexual couples reported on their own behavior and perceptions of their partner’s behavior five times across a year (Study 1) and daily for 3 weeks (Study 2). Men who more strongly endorsed HS perceived their partner’s behavior as more negative than was justified by their partner’s reports. Furthermore, more negative perceptions of the partner’s behavior mediated the links between men’s HS and feeling more manipulated by their partners, behaving more negatively toward their partners, and lower relationship quality. This indicates that men who endorse HS behave more negatively toward intimate partners and experience lower relationship satisfaction because their antagonistic attitudes toward women in general permeate the way they perceive those partners.

Keywords

hostile sexism bias close relationships relationship perceptions relationship satisfaction relationship behavior

Hostile sexism (HS) is an ideology which describes women as competing for men’s power and willing to use underhanded tactics to do so. For example, HS is indexed by agreement that women pursue power “by getting control over men” and use the “guise of equality” to get ahead of men (Ambivalent Sexism Inventory; Glick & Fiske, 1996). Thus, at the heart of HS are aggressive and threatening attitudes toward women who hold the potential to challenge men’s power. Accordingly, HS tends to target women who challenge men’s societal dominance. For example, men who endorse HS evaluate feminists and career women more negatively, but do not view homemakers more negatively (Glick, Diebold, Bailey-Werner, & Zhu, 1997). However, research has also shown that men who endorse HS are relatively more accepting of aggression within close relationships (e.g., Forbes, Adams-Curtis, & White, 2004; Glick, Sakalli-Ugurlu, Ferreira, & Aguiar de Souza, 2002) and exhibit more hostile behavior toward intimate partners (e.g., Overall, Sibley, & Tan, 2011). These findings indicate that men who endorse HS also view intimate relationships as a context of competition for power and control, consistent with fears that women will exploit men’s relational dependence to subvert men’s power (Glick & Fiske, 1996).

In the current research, we tested whether endorsing HS biases perceptions of women’s behavior within intimate relationships. We predicted that men who endorsed HS would perceive their (female) partner’s behavior more negatively than was justified. We also tested whether such biased perceptions are a key reason why men’s endorsement of HS is associated with more hostile behavior toward intimate partners as well as feeling more manipulated and dissatisfied within relationships. We outline the theoretical basis and novelty of our predictions below, and then present two studies which make several methodological advances by assessing how HS influences perceptions of actual (rather than hypothetical or described) women, testing the veracity of those perceptions by comparing perceptions with the behavior reported by the partner, and examining how HS shapes cognition and behavior as relationships progress across 1 year (Study 1) and daily over a 3-week period (Study 2).

Ambivalent Sexism and Intimate Relationships

The intergroup relations between men and women differ markedly from the relationship between men and women within intimate contexts. Men typically have more direct access to status and resources than women, affording them more societal-level power. In contrast, within intimate (heterosexual) contexts men and women experience more equal power because men and women depend on each other for intimacy, support and reproduction. According to Ambivalent Sexism Theory, the tension between men’s societal dominance and the interdependence required for intimate relationships produces two forms of sexist ideologies (Glick & Fiske, 1996). The first, HS, comprises attitudes commonly identified as “sexist.” HS asserts men’s societal advantages by expressing hostile and aggressive attitudes toward women who challenge men’s power, such as feminists or career women. However, hostile attitudes toward women impede men’s ability to fulfill their relational needs and do little to promote women’s adoption of traditional roles. The second set of attitudes, Benevolent Sexism (BS), addresses these needs by expressing subjectively positive attitudes toward women who fulfill conventional roles in the home. BS reveres women’s interpersonal strengths but also casts women as needing protection, perpetuating assumptions of women’s inferiority outside of domestic domains.

HS and BS work in tandem to maintain gender inequality by shaping the structure of heterosexual relationships and limiting the degree to which women can gain societal power (Glick & Fiske, 1996). The role of HS is perhaps the most obvious and straightforward: punishing women who challenge men’s dominance limits women’s social power, such as through derogating and discriminating against career women (Glick et al., 1997; Masser & Abrams, 2004). The effect of BS is more subtle. Revering women as relationship partners helps men achieve satisfying intimate relationships but also preserves men’s dominance by rewarding women for adopting traditional roles. For example, men who endorse BS behave in more caring ways in relationships (Overall et al., 2011) and these benefits foster women’s acceptance of BS attitudes. In turn, however, endorsement and expression of BS reduces women’s career aspirations, work-related performance and confidence to adopt independent roles outside the home (e.g., Dardenne, Dumont, & Bollier, 2007; Rudman & Heppen, 2003).

Thus, HS and BS operate as a system: HS threatens power-challenging women outside the relationship domain whereas BS is directed toward supportive women within intimate relationships. Glick et al. (1997; Study 2) found that men who more strongly endorsed HS expressed more negative evaluations of career women, but HS was not associated with evaluations of homemakers. In contrast, men who endorsed BS expressed more positive evaluations of homemakers, but BS was not associated with evaluations of career women. Other research also supports that nontraditional subtypes (“sexual temptresses”) elicit hostile attitudes whereas traditional subtypes (homemakers) trigger benevolent attitudes (e.g., Sibley & Wilson, 2004). According to Glick et al. (1997), categorizing women as competitive “women out there” versus supportive “relationship partners” helps men avoid feeling conflicted about women and fulfill the drive to preserve societal dominance while maintaining the well-being of intimate relationships.

Yet, research indicates that the endorsement of HS also influences how men evaluate and react to romantic partners. For example, men who more strongly endorse HS are more accepting of violent behavior and verbal aggression toward intimate partners (Forbes, Adams-Curtis, et al., 2004; Forbes, Jobe, White, Bloesch, & Adams-Curtis, 2005; Glick et al., 2002; Yamawaki, Ostenson, & Brown, 2009), particularly when those partners are seen to be challenging men’s authority (e.g., partners who “do not behave well should be treated severely,” Chen, Fiske, & Lee, 2009, p. 771). Moreover, men who strongly endorse HS exhibit greater hostility during conflict with their partner (Overall et al., 2011), experience heightened dissatisfaction when facing problems (Hammond & Overall, 2013), and are more afraid of intimacy (Yakushko, 2005). These findings indicate that the power concerns associated with HS are not restricted to maintaining dominance outside the home. Indeed, the power which women hold within relationships should be particularly threatening for men who endorse HS. For example, men’s endorsement of HS encompasses beliefs that women want to control men, that relationship dependence leaves men exposed to being manipulated (e.g., being put “on a tight leash”; Ambivalent Sexism Inventory; Glick & Fiske, 1996), and that men are shamed if they lose dominance in the relationship (Chen et al., 2009). Thus, the attitudes which HS expresses toward women in general should be readily applied to romantic partners. We tested this proposition by investigating whether HS is associated with negative perceptual biases in close relationships and, in turn, more negative relationship evaluations and behavior.

HS and Biased Perceptions

What is it about men’s endorsement of HS that should bias their perception of women? HS encompasses beliefs that men must compete for power over women, which stem from goals for group-centered dominance and the belief that outgroups are competitive (Sibley, Wilson, & Duckitt, 2007). However, because HS also recognizes the power women hold at the interpersonal level via women’s responsibility for managing the domestic domain and ability to act as “gatekeepers” of sex, HS attitudes also encompass concerns that women will use their interpersonal capability “unfairly” to undermine men’s power. For example, HS warns that women will use their sexuality to manipulate men and exploit intimate relationships to undermine men’s power (Glick & Fiske, 1996). Thus, echoing the broader competitive conceptualization of gender relations, men who endorse HS likely see relationships as a competition for control in which men’s power will be usurped unless aggressively safeguarded.

If the antagonistic views that HS expresses toward women in general also apply within romantic relationships, then endorsing HS should color men’s perceptions of their intimate partners. Consistent with several models of social perception, individuals interpret the behavior of others in line with their existing beliefs about the world (for reviews, see Bradbury & Fincham, 1990; Noller & Ruzzene, 1991). For example, individuals who are more sensitive to rejection interpret others’ behavior as more rejecting and negative than is warranted (e.g., Downey & Feldman, 1996). Similarly, extensive literature illustrates that the stereotypes individuals hold shapes perceptions of others’ behavior. For example, people imbue the ambiguous behavior of an African American target with more hostility and threat when holding the stereotype that African Americans are dangerous (Devine, 1989; Lepore & Brown, 1997). Thus, the concerns expressed by HS about women in general should bias men’s perceptions of their intimate partner’s behavior. In particular, men who endorse HS are likely to be more cautious and untrusting of the intentions underlying their partner’s behavior. Seen in the light that women seek to exploit men’s relationship dependence, common relationship behaviors, such as criticism and affection, are likely to be perceived more negatively (i.e., even more critical and insulting or less affectionate and supportive than they actually are).

We did not expect men’s endorsement of BS would show the same pattern of bias. Although BS works to maintain the same gender hierarchies as HS, BS supports and reveres women’s relationship power and capacity for intimacy, praising women for their loyalty and morality rather than warning about the dangers of manipulative and power-challenging women (Glick & Fiske, 1996, 2001). Accordingly, men who endorse BS behave with more openness and care toward their partners (Overall et al., 2011) and report greater relationship satisfaction (Sibley & Becker, 2012). BS is also related to romanticized views of relationships (Hart, Hung, Glick, & Dinero, 2012), which may promote perceiving partners in an idealized and positive biased light (e.g., Murray et al., 2011). On the other hand, BS is linked with negative evaluations of women who are portrayed as disloyal (Abrams, Viki, Masser, & Bohner, 2003), so such idealization may be attenuated when judging behaviors which have important implications for their relationship (see Fletcher & Kerr, 2010).

To test the links between sexist attitudes and biased perceptions, in the present research, we asked both members of committed heterosexual couples to report on (a) their own and (b) their perceptions of their partner’s behavior multiple times across a year (Study 1) and daily over a 3-week period (Study 2). We assessed common behaviors that affect the functioning of intimate relationships, including negative (e.g., being critical or insulting) and positive (e.g., being support or affectionate) behaviors. To assess bias, we compared the perceptions of the partner’s behavior with the behavior actually reported by the partner. We predicted that men who more strongly endorsed HS would perceive the behavior of their intimate partners as more negative than warranted based on their partner’s reports. We also predicted that these negatively biased perceptions would have important consequences for men’s relationship behavior and evaluations.

Consequences of Biased Perceptions

A bulk of research has shown negative biases are damaging for relationships. The more individuals possess negatively biased perceptions of their partner, the less able they are to maintain satisfaction, commitment and regard for their partner (Fletcher & Kerr, 2010; Miller & Rempel, 2004; Murray et al., 2011). In particular, perceiving negative or hurtful partner behavior undermines trust and satisfaction, and elicits retaliatory hostility and defensiveness (Gottman, 1998; Wieselquist, Rusbult, Foster, & Agnew, 1999). Accordingly, the negatively biased perceptions we expected to be associated with men’s HS should produce relationship dissatisfaction and more negative behavior toward intimate partners.

These potential outcomes are particularly important given the existing links between men’s HS and aggressive behavior toward intimate partners (e.g., Forbes, Adams-Curtis, et al., 2004; Overall et al., 2011) and lower relationship satisfaction (e.g., Sibley & Becker, 2012). The current research extends prior research by testing whether men’s HS is associated with more hostility and dissatisfaction within relationships because they hold more negatively biased perceptions of their partner’s behavior. Figure 1 outlines our predictions. We propose that the beliefs within HS that women are competitive and vying for control are applied to intimate partners. These expectations should color men’s perceptions such that they overestimate the negativity of their partner’s behavior. In turn, these negatively biased perceptions should be associated with more negative relationship evaluations and more hostile behavior toward partners. Thus, we predicted that negatively biased perceptions of the partner’s behavior would mediate the links between men’s HS and relationship evaluations and behavior.

Figure 1.

The hypothesized links between men’s hostile sexism, biased perceptions of the partner’s behavior, and relationship outcomes.

We also predicted that a central outcome of negatively biased perceptions of the partner would be feeling manipulated. Men’s endorsement of HS should be associated with feeling more manipulated by their partner because of beliefs that men are vulnerable to being undermined or humiliated by women who are seen as “manipulative temptresses” (Glick & Fiske, 1996, p. 494). Importantly, the extent to which men who endorse HS feel manipulated by their partners should be a function of their more negative perceptions of their partner’s behavior. This test captures a key element of Ambivalent Sexism Theory that men who endorse HS resent more critical and unsupportive behavior because they see this behavior as partners using an “unfair” relationship position to purposefully undermine their power.

Current Research

The current research represents the first test of whether HS influences actual evaluations of women and whether these evaluations are biased. Prior research has focused on the links between HS and ratings of described categories of women (e.g., career women vs. homemaker; Glick et al., 1997), vignettes of hypothetical women (e.g., Masser & Abrams, 2004; Sibley & Wilson, 2004), or attitudes toward potential relationship partners (e.g., Chen et al., 2009). In contrast, we assessed perceptions of current romantic partners multiple times across a year (Study 1) and daily across a 3-week period (Study 2). Moreover, by gathering data from both couple members, we assessed the degree to which perceptions of the partner’s behavior were biased by comparing perceptions with the partner’s reports of that behavior.

As summarized in Figure 1, we expected that (a) men’s HS would be associated with more negatively biased perceptions of the partner’s behavior; (b) such negatively biased perceptions would predict feeling more manipulated by the partner, reduced relationship evaluations, and increased negative behavior toward the partner; and (c) more negative perceptions of the partner’s behavior would mediate the links between men’s HS and feeling more manipulated, lower relationship evaluations, and more hostile behavior toward the partner. Thus, in addition to representing the first examination of how HS shapes cognition and behavior as relationships develop over time, the current studies are the first to test a central reason underlying the link between HS and aggressive behavior toward women: men who endorse HS perceive their female partner’s behavior more negatively than it actually is.

Study 1

Method

Participants

Eighty-six heterosexual couples replied to recruitment advertisements posted around a New Zealand university. Participants’ ages ranged from 18 to 45 (M = 21.86, SD = 4.08) and they were involved in long-term, committed relationships with a mean length of 2.58 years (SD = 1.70). Participants reported their relationship status as married (11.6%), cohabitating (43%), serious (39.5%), or steady (5.8%).

Procedure

Couples completed five questionnaires across a 1-year period. The first questionnaire was completed as part of an initial laboratory session. Participants were then mailed a separate set of questionnaires four times over the following year at 3-month intervals. Fifteen couples from the initial sample did not complete at least one follow-up questionnaire (required for the current analyses) leaving the sample described above. Over the year, some couples did not complete all questionnaires either because they chose not to or because the relationship dissolved (subsequent Ns across each 3-month phase: 82, 68, 66, 57). On average, individuals who remained across the study perceived their partner’s behavior to be relatively less negative at the initial session (M_Diff = −0.37, SE = 0.19, df = 169, t = 1.96, p = .05), highlighting the importance of perceptions of the partner’s behavior. Importantly, our analytic approach accounts for sample attrition by weighting the estimates according to how many measurements exist for each couple.

Measures

HS and BS

At the initial session, participants completed a short-form version of the Ambivalent Sexism Inventory (Glick & Fiske, 1996). HS was indexed by the average of six items, such as “Women seek to gain power by getting control over men” and “Once a woman gets a man to commit to her, she usually tries to put him on a tight leash” (−3 = strongly disagree, 3 = strongly agree; α = .68). Six items also assessed BS (e.g., “Women should be cherished and protected by men”; α = .63). These short-form scales demonstrate strong correlations (rs > .90) with the full scales, good across-time reliability (Sibley & Perry, 2010), and predict observed relationship behavior (Overall et al., 2011).

Own and perceptions of partner’s relationship behavior

At all five measurement phases, participants reported on their own and their partner’s conflict behavior over the prior 3 months (i.e., the time since the prior measurement phase). Participants were asked to think about times in the prior 3 months when their partner behaved in ways that caused relationship problems or conflict, and rated items assessing the communication strategies they used to change their partner’s behavior. Six items assessed negative strategies (e.g., “I insulted or put down my partner to make him/her change”) and six items assessed positive strategies (e.g., “I encouraged my partner to express their thoughts and feelings about what I wanted changed”). Positive items were reverse-scored and all items were averaged so that higher scores indicated more negative behavior (average α across the five measurement phases = .83). Participants completed analogue items which assessed perceptions of the partner’s behavior over the prior 3 months, such as “My partner insulted or put me down to make me change,” “My partner encouraged me to express my thoughts and feelings about what he/she wanted changed” (average α = .83).

Relationship quality

At each phase, participants also completed the 7-item Perceived Relationship Quality Components (PRQC) inventory (Fletcher, Simpson, & Thomas, 2000), which assesses satisfaction, commitment, closeness, trust, passion, love, and romance (e.g., “How close is your relationship?” 1 = not at all, 7 = extremely; average α = .88).

Results

HS and Biased Perceptions

Descriptive statistics and correlations for measures at the initial session are presented in Table 1. As predicted, men who endorsed HS possessed more negative perceptions of their partner’s behavior. To assess whether these perceptions were biased, however, requires comparing the perceptions of the partner’s negative behavior with the partner’s actual reported behavior, which is the standard benchmark for assessing bias within intimate relationships (Fletcher & Kerr, 2010; Gagné & Lydon, 2004; West & Kenny, 2011). To do this, we used the most up-to-date analytic strategy for assessing bias developed by West and Kenny (2011) using the measures across all five time points. The basic multilevel model is as follows:

P_{i j} = b_{0 j} + b_{1 j} (partner’s reported negative behavior at time i) + e_{i j}

Table 1.

Descriptive Statistics and Correlations of Measures at the Initial Session (Study1).

	Men	Women	Gender diff.
	M (SD)	M (SD)	t	1	2	3	4	5
1. Hostile sexism	0.13 (1.16)	−0.49 (1.05)	3.68*	.23*	.28*	−.01	−.11	−.00
2. Benevolent sexism	0.51 (1.02)	0.30 (1.09)	1.29	.30*	.12	−.03	.12	.02
3. Negative behavior	3.01 (0.80)	3.16 (1.04)	−1.06	.32*	.10	.43*	.59*	−.48*
4. Perceptions of partner’s negative behavior	3.39 (1.06)	3.41 (1.30)	−0.83	.28*	.16	.54*	.44*	−.54*
5. Relationship quality	6.07 (0.57)	6.04 (0.69)	0.08	.06	.15	−.26*	−.38*	.46*

Note. Possible scores range from −3 to 3 for Hostile Sexism and Benevolent Sexism, and 1 to 7 for all other scales. Bold correlations on the diagonal represent correlations across partners. Correlations above the diagonal are for women; correlations below the diagonal are for men.

p < .05.

In this equation, perceptions of the partner’s negative behavior (P) by person j at a particular time point (i) is a function of an intercept (b₀ for person j), the partner’s self-reported negative behavior (b₁) for that time point (i for person j), and an error term (e_ij). As specified by West and Kenny (2011), perceptions of the partner’s behavior (the outcome variable) were centered on the partner’s reports of his or her behavior by subtracting the grand mean of partner-reported behavior from individuals’ perceptions of their partner’s behavior. This centering strategy means that the intercept represents the difference between the partner’s reported behavior and perceptions of that behavior or bias. A positive intercept indicates that perceivers generally overestimated the negativity of their partner’s behavior.

To test whether bias was greater for men who more strongly endorsed HS, we entered HS (grand mean centered) as a predictor of the level 1 intercept (which modeled bias). Consistent with prior research, and because HS and BS are positively associated (see Table 1), we also entered BS as a simultaneous predictor of bias (see Equation 2).

b_{0 j} = B_{00} + B_{01} (HS) + B_{02} (BS) + u_{0 j}

Our key research question concerning whether men’s HS produced more biased perceptions is tested by Equation 2. However, in addition to modeling bias, the effect of the partner’s reported negative behavior (b₁ in Equation 1) assesses tracking accuracy. This represents the degree to which perceptions of the partner’s negative behavior correspond to changes in the partner’s self-reported behavior across time points (see West & Kenny, 2011). We expected that all participants would demonstrate high levels of tracking accuracy, revealing that perceptions were capturing how behavior varied across time in their relationships. In addition, although we hypothesized that men who endorsed HS would perceive their partner’s behavior to be more negative, we expected this bias to be applied within the context of accurate recognition of high versus low levels of negative behavior enacted by the partner. Thus, we did not expect that HS would be associated with differences in tracking accuracy. Nonetheless, to be complete we also added HS and BS as simultaneous predictors of tracking accuracy (see Equation 3).

b_{1 j} = B_{00} + B_{11} (HS) + B_{12} (BS) + u_{0 j}

All analyses were conducted using the MIXED procedure in SPSS 20. Following multilevel modeling procedures for repeated measures data within dyads (Kenny, Kashy & Cook, 2006), we used a no-intercept model to simultaneously estimate model parameters for men and women separately. Models allowed the error variances to differ for men and women and allowed errors for a given time to be correlated. Both models also allowed directional bias (b_0j) and tracking accuracy (b_1j) to vary by male and female perceivers for each dyad (i.e., be random variables) and these effects to covary within and across dyad members. See Overall, Fletcher, and Kenny (2012) for associated SPSS syntax.

The results are shown in Table 2. The intercept assessing bias was positive and significant indicating that both men and women generally overestimated the negativity of their partner’s behavior across the year, but also accurately tracked changes in their partner’s behavior across the year. Examining the effects of sexist attitudes, as predicted (and shown in bold), men who more strongly endorsed HS were more biased—They perceived more negative behavior than was reported by their partners. In contrast, men’s BS and women’s sexism were not associated with bias (see Table 2). There were also no significant associations between sexist attitudes and tracking accuracy. Thus, men who more strongly endorsed HS recognized when their partner was behaving more versus less negatively, but consistently perceived their partner’s behavior more negatively than their partner’s reports indicated was justified.

Table 2.

The Effects of Hostile Sexism and Benevolent Sexism on Directional Bias and Tracking Accuracy of Perceptions of the Partner’s Negative Behavior (Study 1).

	Men’s Perceptions of partner’s negative behavior			Women’s perceptions of partner’s negative behavior			Gender differences in bias and accuracy
	B	SE	t	B	SE	t	B	SE	t
Directional bias	.29	.08	3.46**	.32	.12	2.78**	−.02	.07	−0.24
Tracking accuracy	.27	.07	3.86**	.17	.08	1.97*	.05	.05	0.99
Effect of hostile sexism
Directional bias	.20	.07	2.63*	−.10	.11	−0.92	.15	.07	2.24*
Tracking accuracy	−.08	.05	−1.41	.09	.09	0.98	−.08	.05	−1.60
Effect of benevolent sexism
Directional bias	−.05	.08	−0.64	.08	.11	0.80	−.07	.07	−1.02
Tracking accuracy	−.00	.06	−0.04	−.14	.07	−1.84	.07	.05	1.38

Note. The first two columns present results from models simultaneously calculating all effects for men and women accounting for the dependence across couple members. Predicted effects are presented in bold. The final column presents tests of whether the effects differed across gender.

p ≤ .05. **p < .01.

Alternative explanations

We ran a series of additional analyses to rule out several alternative explanations, including whether the bias associated with men’s HS was due to individual’s own negative behavior being projected onto partners (i.e., assumed similarity; see Gagné & Lydon, 2004; Kenny, & Acitelli, 2001) or simply the result of more global negativity within the individual or the relationship, such as general aggressiveness or dissatisfaction. Although own negative behavior (B = .41, t = 6.33, p < .001) and more negative relationship evaluations (B = −.29, t = −3.88, p < .001) predicted more negative perceptions of the partner’s behavior, the bias associated with men’s HS was not reduced when controlling for these variables (B = .15, t = 2.34, p = .02 and B = .21, t = 2.91, p = .01, respectively) and HS was not moderated by these variables (B = −.01, t = −0.17, p = .87 and B = .06, t = 0.98, p = .33). These additional analyses provide good evidence that the bias associated with men’s HS is specific to the beliefs HS encompasses rather than the result of more global negativity or general aggressiveness toward partners.¹

Consequences of Biased Perceptions

We predicted that biased perceptions would lead to lower relationship quality and greater negative behavior toward the partner (see Figure 1). We ran two sets of models to examine these consequences, simultaneously calculating model parameters separately for men and women using procedures for modeling repeated measures dyadic data (Kenny et al., 2006). The first set of models regressed relationship quality and negative behavior toward the partner on HS and BS to assess whether men’s HS was directly associated with relationship quality and negative behavior toward the partner. The second set of models tested whether HS, BS, and perceptions of the partner’s behavior predicted residual change in (a) relationship quality and (b) negative behavior toward the partner (described further below) to assess whether more negative perceptions of the partner’s behavior (controlling for HS and BS) were associated with the predicted negative consequences. Third, we calculated indirect effects to test whether men’s endorsement of HS was associated with lower relationship quality and more negative behavior via more negative perceptions of the partner’s negative behavior. Predictor variables were grand mean centered.

Direct effects of men’s HS

The direct effects of men’s endorsement of HS and BS are presented in the upper section of Table 3. Men’s endorsement of HS predicted greater negative behavior toward partners over the year but, unexpectedly, was not directly related to relationship quality. However, a nonsignificant direct effect does not rule out the possibility that HS is associated with relationship quality via perceptions of the partner’s behavior (see Rucker, Preacher, Tormala, & Petty, 2011), which is tested in the following models. There were no significant effects for women’s endorsement of HS or BS (ts = −0.84 to 1.86).

Table 3.

Analyses Examining the Effects of Hostile Sexism, Benevolent Sexism, and Perceptions of the Partner’s Negative Behavior on Relationship Quality and Own Negative Behavior Directed Toward the Partner (Study1).

	Relationship quality			Own negative behavior toward the partner
	B	SE	t	B	SE	t
Model 1
Men’s hostile sexism	−.02	.03	−0.78	.15	.06	2.46*
Men’s benevolent sexism	.08	.03	2.42*	.01	.07	0.12
Model 2
Men’s hostile sexism	−.03	.03	−0.88	.03	.03	0.94
Men’s benevolent sexism	.03	.04	0.91	−.01	.04	0.45
Perceptions of partner’s negative behavior	−.12	.04	3.01*	.14	.05	3.18**
Partner’s self-reported negative behavior	−.08	.04	−1.82	.00	.04	0.05

Note. These effects were calculated using dyadic models that simultaneously calculated effects for women and men accounting for the dependence across couple members. The table presents the results for men only. There were no significant effects of women’s HS or BS.

p < .05. **p < .01.

Effects of perceiving more negative behavior

Next, we examined whether more negative perceptions of the partner’s behavior predicted (a) lower relationship quality and (b) more negative behavior toward partners (controlling for HS and BS). To illustrate, relationship quality of individual j at time i was modeled as a function of (a) relationship quality measured at time i − 1 so any significant effects represent prediction of residual change in relationship quality over the 3-month phase, (b) perceptions of the partner’s negative behavior over the prior 3 months, (c) the partner’s reported negative behavior so that any effects of perceptions were over and above actual levels of the partner’s behavior (i.e., represented bias), (d) HS, and (e) BS. Similar models were run to assess own negative behavior as the outcome, but because participants reported on their negative behavior over the previous 3 months (compared with current relationship quality), these models predicted own negative behavior at time i + 1 from perceptions of the partner’s behavior at time i, controlling for own and partner-reported negative behavior at phase i. The focal results from both models are presented in the lower section of Table 3. As predicted, more negative perceptions of the partner’s behavior predicted reductions in relationship quality and subsequent increases in own negative behavior toward the partner.

Indirect effects of men’s HS

The indirect effects and associated confidence intervals testing whether men’s HS was associated with (a) lower relationship quality and (b) greater negative behavior toward the partner via more negative perceptions of that partner’s behavior are shown in the upper section of Table 4. The confidence intervals did not overlap zero, indicating that men who strongly endorse HS experienced poorer relationship quality and behaved more negatively toward partners because they possessed more negative perceptions of their partner’s behavior.

Table 4.

Indirect Effects Between Men’s Hostile Sexism and Relationship Outcomes Mediated by (More Negative) Perceptions of the Partner’s Behavior as Shown in Figure 1 (Studies 1 and 2).

		95% Confidence interval
Indirect effect tested	Indirect effect	Lower limit	Upper limit
Study 1
Men’s hostile sexism → perceptions of negativity → relationship quality	−.023	−.051	−.005
Men’s hostile sexism → perceptions of negativity → negative behavior	.028	.005	.061
Study 2
Men’s hostile sexism → perceptions of negativity → daily felt manipulated	.087	.044	.144
Men’s hostile sexism → perceptions of negativity → daily relationship satisfaction	−.117	−.189	−.044
Men’s hostile sexism → perceptions of negativity → daily negative behavior	.114	.043	.182

Note. Estimates for paths between variables (indicated by→) are displayed in Tables 2 and 3 (Study 1) and Tables 6 and 7 (Study 2). Asymmetric Confidence intervals were calculated following Mackinnon, Fritz, Williams, and Lockwood (2007). Confidence intervals which do not overlap “0” can be considered significant.

Discussion

In Study 1, both members of heterosexual couples reported on the frequency of their own, and their partner’s, negative versus positive behavior five times across a year. Individuals tended to overestimate the extent to which their partner had behaved negatively when compared with the level of negative behavior their partner reported. As predicted, this negative bias was stronger the more men endorsed HS and, in turn, these more negative perceptions predicted lower relationship quality and more negative behavior (see Figure 1).

Study 2

The longitudinal design of Study 1 allowed us to examine perceptions over 1 year in ongoing relationships. However, the measures relied on participants to recall their own and their partner’s behavior over the prior 3 months. This introduces error because (a) people’s recollections tend to attribute positive events to themselves and emphasize the role of others in negative events (Mezulis, Abramson, Hyde, & Hankin, 2004) and (b) people are more likely to attend and react to their partners’ negativity relative to positive behavior, and so negative behavior may have had more influence on perceptions (Baumeister, Bratslavsky, Finkenauer, & Vohs, 2001). Thus, in Study 2, we reduced recall bias by asking couples to report on their own and their perceptions of their partner’s behavior at the end of every day for 3 weeks. Using the analytic method presented in Study 1, we tested whether men who more strongly endorsed HS overestimated the negativity of their partner’s behavior by comparing perceptions of the partner’s behavior to the partner’s self-reported behavior. We also measured daily outcomes to test whether more negatively biased perceptions were associated with feeling more manipulated by the partner, experiencing lower relationship satisfaction, and behaving more negatively toward the partner.

Method

Participants

Seventy-eight heterosexual couples replied to recruitment advertisements posted around a NZ university. Participants’ ages ranged from 17 to 48 years (M = 22.44, SD = 4.82), and they were involved in long-term committed relationships, with a mean length of 2.58 years (SD = 1.99). Participants’ relationships were characterized as married (10.3%), cohabitating (34.6%), serious (48.7%), or steady (6.4%).

Procedure

Demographic information and assessment of sexist attitudes were gathered during an initial session. Participants received detailed instructions regarding the web-based diary they were asked to complete at the end of each day for the following 3 weeks. Participants were reimbursed 90 NZD for completing all study components.

Materials

HS and BS

Participants completed the full version of the Ambivalent Sexism Inventory (Glick & Fiske, 1996), which comprised 11 items for each scale (−3 = strongly disagree; 3 = strongly agree). Items were averaged to construct overall scores for HS (α = .85) and BS (α = .79).

Relationship Quality: Participants Completed the 7-Item PRQC Described in Study 1 (α = .81) Daily questionnaires

Participants completed an average of 19.3 entries (92%). Items asked participants to report on their own and their partner’s behavior each day, feelings of being manipulated by the partner, relationship satisfaction and conflict. All items were presented on a scale ranging from 1 (not at all) to 7 (very much).

Relationship behavior

Participants rated five items that captured similar behaviors assessed in Study 1 and have been previously validated to assess important daily behaviors which shape relationship outcomes (see Overall & Sibley, 2010): “I acted in a way that could be hurtful to my partner,” “I was critical or unpleasant toward my partner,” “I shared and discussed my feelings and opinions with my partner,” “I was affectionate and loving toward my partner,” and “I was supportive to my partner.” Positively-valenced items were reverse-scored and then items were averaged so that higher scores indicated more negative behavior that day (α = .76). The same items were reworded to assess perceptions of the partner’s behavior (e.g., “My partner was critical or unpleasant toward me”; α = .83).²

Feeling Manipulated, Relationship Satisfaction, and Conflict were assessed each day by single items: “I felt manipulated by my partner,” “I was satisfied with our relationship,” and “I experienced conflict or disagreement with my partner.”

Results

HS and Biased Perceptions

Table 5 presents the descriptive statistics for both the initial questionnaire and daily measures. The structure of our data and analyses were the same as in Study 1. Measures of perceived behavior over the 21 days were nested within dyad and so we used the multilevel modeling procedures outlined by West and Kenny (2011) to test whether men’s endorsement of HS predicted more biased perceptions of their partner’s negative behavior. The resulting model was equivalent to Equations 1 to 3 (Study 1), with time points (i) representing the daily measures across the 3-week diary period. As before, we simultaneously estimated model parameters for men and women controlling for the dependence across couple members.

Table 5.

Descriptive Statistics for Questionnaire and Daily Diary Measures (Study 2).

	Men	Women	Gender differences	Correlations between questionnaire measures
	M (SD)	M (SD)	t	1	2	3
Questionnaire measures
Hostile sexism	.09 (1.14)	−.06 (1.19)	0.76	.44*	.14	.12
Benevolent sexism	.32 (0.99)	−.19 (1.01)	3.19*	.39*	.25*	.13
Relationship quality	6.12 (0.63)	6.12 (0.61)	−0.02	−.07	.15	.24*
Daily measures
Negative behavior	2.27 (1.05)	2.37 (1.10)
Perceptions of partner’s negative behavior	2.29 (1.16)	2.32 (1.23)
Relationship satisfaction	5.87 (1.44)	5.88 (1.50)
Relationship conflict	2.47 (1.84)	2.54 (1.90)

Note. Possible scores range from −3 to 3 for Hostile Sexism and Benevolent Sexism and 1 to 7 for all other measures. Bold correlations on the diagonal represent correlations across partners. Correlations above the diagonal are for women; correlations below the diagonal are for men.

p < .05.

The results from this analysis are presented in Table 6. The intercept assessing bias did not significantly differ from zero, indicating that on average participants did not under- or overestimate the degree to which their partner behaved negatively. Both men and women were also very accurate at tracking changes in their partner’s negative behavior across days. Examining the effects of sexist attitudes, as predicted (and shown in bold), men who more strongly endorsed HS overestimated the degree to which their partner behaved negatively on a daily basis. Also consistent with Study 1, HS was not significantly associated with tracking accuracy, indicating that men who endorsed HS demonstrated consistent negative bias across both high and low levels of their partner’s actual negative behavior. Unlike Study 1, men who more strongly endorsed BS underestimated the degree to which their partners behaved negatively. There were no significant effects of women’s HS or BS.

Table 6.

The Effects of Hostile Sexism and Benevolent Sexism on Directional Bias and Tracking Accuracy of Perceptions of the Partner’s Daily Negative Behavior (Study 2).

	Men’s perceptions of partner’s negative behavior			Women’s perceptions of partner’s negative behavior			Gender differences in bias and accuracy
	B	SE	t	B	SE	t	B	SE	t
Directional bias	−.02	.08	−0.32	.00	.08	0.03	−.01	.02	−0.63
Tracking accuracy	.63	.04	15.55**	.74	.05	16.55**	−.02	.02	−0.97
Effect of hostile sexism
Directional bias	.19	.06	3.21**	−.05	.05	−0.93	.12	.02	5.29*
Tracking accuracy	.04	.04	1.07	−.00	.04	−0.04	.03	.02	1.43
Effect of benevolent sexism
Directional bias	−.20	.06	−3.05**	.10	.06	1.59	−.07	.03	−2.30*
Tracking accuracy	−.05	.04	−1.18	.05	.04	1.20	−.02	.03	−0.79

p < .05. **p < .01.

Alternative explanations

As in Study 1, we tested whether the negative bias associated with men’s HS was simply due to assumed similarity or more global negativity by controlling for men’s own negative behavior and relationship evaluations. The more men behaved negatively, the more they exhibited greater bias (B = .63, t = 17.49, p < .001), but men’s own negative behavior did not account for (B = .09, t = 2.53, p = .01) or moderate (B = .00, t = 0.07, p = .95) the link between men’s HS and bias. The bias linked with HS also remained when controlling for daily relationship satisfaction, feeling manipulated, or initial relationship quality (Bs > .16, ts > 2.66, ps < .01). Similarly, although greater daily conflict was associated with greater bias (B = .20, t = 15.14, p <.001), conflict did not moderate (B = .02, t = 1.43, p = .15) or account for (B = .23, t = 3.50, p < .001) the bias linked with HS. The positive bias associated with BS was not as robust, and was eliminated when controlling for men’s own behavior (B = −.06, t = −1.56, p = .12) suggesting this may be due to projecting one’s own positive behavior onto the partner (Kenny, & Acitelli, 2001).³

Consequences of Biased Perceptions

We next tested whether the negatively biased perceptions of men who endorsed HS were associated with feeling more manipulated, less satisfied, and behaving more negatively toward the partner. As in Study 1, we examined this in three steps by (a) running a first set of models examining whether HS (and BS) were directly associated with each of these daily outcomes, (b) running a second set of models testing whether perceptions of the partner’s negative behavior predicted these daily outcomes controlling for HS (and BS), and then (c) calculating indirect effects to test whether men’s endorsement of HS was linked with these negative daily outcomes via more negatively biased perceptions of their partner’s behavior.

Direct effects of men’s HS

The direct effects of men’s endorsement of HS and BS are presented in the upper section of Table 7. As expected, men who more strongly endorsed HS felt more manipulated by their partners, were less satisfied in their relationships, and behaved more negatively toward their partners on a daily basis. There were no significant effects for women’s endorsement of HS or BS (ts = −0.89 to 1.35).

Table 7.

Analyses examining the effects of Hostile Sexism, Benevolent Sexism and Daily Perceptions of the Partner’s Negative Behavior on Feeling Manipulated by the Partner, Relationship Satisfaction, and Own Negative Behavior Directed Toward the Partner (Study 2).

	Feeling manipulated by the partner			Relationship satisfaction			Own negative behavior toward the partner
	B	SE	t	B	SE	t	B	SE	t
Model 1
Men’s hostile sexism	.24	.08	3.03**	−.18	.08	−2.14*	.19	.07	2.69**
Men’s benevolent sexism	−.22	.09	−2.49*	.03	.09	0.35	−.15	.08	−1.82
Model 2
Men’s hostile sexism	.10	.06	1.62	.01	.05	0.12	.04	.03	1.26
Men’s benevolent sexism	−.15	.07	−2.17*	−.06	.06	−1.06	−.04	.04	−0.97
Perceptions of partner’s negative behavior	.46	.03	13.64**	−.61	.03	−19.31**	.59	.02	31.54**
Partner’s self-reported negative behavior	.09	.03	2.63**	−.13	.03	−4.36**	.13	.02	6.97**

p < .05. **p < .01.

Effects of perceiving greater negative behavior

Using the same analytic strategy reported in Study 1, we next examined whether more negative perceptions of the partner’s behavior predicted residual increases in daily feelings of being manipulated, reductions in daily satisfaction, and increased negative behavior toward the partner (controlling for HS, BS and the partner’s level of self-reported negative behavior). The focal results are presented in the lower section of Table 7. As predicted, greater perceptions of a partner’s negative behavior on a given day was associated with significant increases in feeling manipulated, decreases in satisfaction, and greater negative behavior toward that partner that day.

Indirect effects of men’s HS

The lower section of Table 4 presents the indirect effects between men’s HS and feeling manipulated, relationship satisfaction and own negative behavior via perceptions of the partner’s negative behavior. The confidence intervals did not overlap zero indicating that men who endorsed HS felt more manipulated, experienced lower satisfaction, and behaved more negatively because they perceived more negative behavior from the partner. Moreover, the effect of men’s HS on these daily outcomes was no longer significant when controlling for perceptions of the partner’s behavior (see lower section of Table 7), providing evidence that negatively biased perceptions of the partner’s behavior fully mediated these links.

Discussion

Study 2 replicated and extended the findings from Study 1 by showing men who more strongly endorsed HS overestimated the negativity of their partner’s daily behavior. In turn, more negatively biased perceptions were associated with feeling more manipulated by the partner, lower relationship satisfaction, and greater negative behavior toward the partner (see Figure 1). These effects were generally stronger than in Study 1 probably because the daily diary method reduced error in assessing both the partners’ behavior and perceptions of that behavior, and also tested more immediate daily outcomes of perceived behavior. Finally, unlike Study 1, men’s BS predicted more positive perceptions of the partner’s behavior, although this bias was eliminated when controlling for men’s own relationship behavior indicating this effect might be due to projection or assumed similarity rather than being specific to more benign interpretations of the partner’s behavior.

General Discussion

HS expresses animosity toward women who challenge men’s power and encompasses expectations that women will try to control men by exploiting their relational dependence (Glick & Fiske, 1996). The current studies tested whether these antagonistic beliefs about women bias men’s perceptions of their intimate (female) partner’s behavior, and whether such negatively biased perceptions account, at least in part, for the dissatisfaction and aggressive behavior linked with men’s endorsement of HS in close relationships. Both members of committed heterosexual couples reported on their own and their partner’s behavior five times across a 1-year period (Study 1) and every day for a 3-week period (Study 2). As hypothesized, men who more strongly endorsed HS perceived their partner’s behavior as more negative than was justified by their partner’s reports. In turn, these more negative perceptions meant that men who endorsed HS felt more manipulated by their partner, behaved more negatively toward their partner, and experienced lower relationship satisfaction (over and above the negativity reported by those partners). These results highlight that the negative characterization of women contained within HS is restricted not only to women outside the relationship domain who challenge men’s societal power but also to color perceptions of intimate partners. The findings also suggest that more negatively biased perceptions of intimate partner’s behavior foster negative behavior and dissatisfaction within close relationships, limiting the degree to which men who endorse HS will experience fulfilling, rewarding relationships.

HS and Biased Perceptions of Intimate Partners

The current research is the first to show that endorsement of HS biases men’s perceptions of women’s behavior. Prior research has shown that men who endorse HS evaluate descriptions of feminists and career women more negatively (e.g., Glick et al., 1997; Sibley & Wilson, 2004). In the current studies, we repeatedly assessed perceptions of actual women who men repeatedly interacted with across time. Moreover, by comparing men’s perceptions to the partner’s reports, we assessed not just whether perceptions were more negative but more biased. As predicted, men who endorsed HS consistently perceived women’s behavior more negatively than was justified. Moreover, these negatively biased perceptions occurred within the context of committed heterosexual relationships revealing that the attitudes expressed by HS are not restricted to women who subvert men’s societal dominance. Although prior theory and research has focused on how men’s HS asserts their societal power (Glick & Fiske, 1996, 2001), the power concerns associated with men’s HS should also arise within intimate contexts. In no other context are men’s desires and goals so dependent on women (and vice versa). Furthermore, such relational dependence leaves intimate partners vulnerable to exploitation, and men who endorse HS believe women will exploit their dependence to undermine their power. As our data reveal, such concerns mean men who endorse HS will be more likely to see common relationship behaviors, such as being critical or supportive, as more negative and underhanded than their partners intended.

Our results also demonstrate that the negatively biased perceptions of men who endorse HS will limit the degree to which they (and their partners) can experience fulfilling and rewarding relationships. A wealth of research has shown that relationships thrive when partners perceive one another more positively (e.g., Fletcher & Kerr, 2010; Murray et al., 2011). In contrast, perceiving the partner more negatively damages trust and feelings of security and tends to trigger retaliatory behavior toward the partner (Gottman, 1994; Murray, Bellavia, Rose, & Griffin 2003). We also expected that, given their fears regarding women’s motivation to use relationships to control men, men who endorse HS would feel more manipulated when perceiving negativity. Accordingly, in the current studies, the negatively biased perceptions associated with men’s HS were associated with feeling more manipulated, reduced relationship satisfaction, and increased negative behavior toward the partner.

These outcomes are consistent with prior research showing that endorsing HS attitudes is associated with more accepting attitudes of aggression toward intimate partners, greater hostile behavior when encountering relationship conflict, and lower relationship satisfaction (e.g., Forbes, Adams-Curtis, et al., 2004; Hammond & Overall, 2013; Overall et al., 2011). The current studies extend this prior work by showing that (a) men’s endorsement of HS predicts more negative behavior and evaluations as relationships progress across time and that (b) these aggressive reactions arise, at least in part, because men possess more negatively biased perceptions of their partner’s behavior. Prior research has suggested that men’s aggressive attitudes and behavior in relationship contexts represent coercive strategies focused on retaining power and personal autonomy (Forbes, Adams-Curtis, et al., 2004; Forbes, Jobe, et al., 2005; Hart et al., 2012; Overall et al., 2012). However, no research has demonstrated that men intentionally engage in hostility to dominate women. Regardless, the perceptual biases shown in the current research should play an important role in understanding the use of power-maintaining strategies because biased perceptions accentuate the degree to which relationship interactions are negative, competitive, and challenging.

Finally, our research adds to the distinction between hostile versus benevolent sexist attitudes. A central tenet of Ambivalent Sexism Theory is that men’s endorsement of BS should promote men’s relationship well-being. Accordingly, men’s BS predicted greater relationship quality (Study 1) and feeling less manipulated by partners (Study 2), and was associated with underestimating the negativity of the partners’ behavior (Study 2). This positive bias may have occurred because men who endorse BS possess romanticized views of relationships, such as believing in “true love” (Hart et al., 2012) and that they are “completed,” and “fulfilled” by intimate partners (Glick & Fiske, 1996, 2001). However, the effects of BS were inconsistent across studies, and the positive bias only emerged in Study 2. Perhaps the behaviors assessed in Study 1 circumvented idealization because these behaviors concerned changing problematic partner behavior—a context that likely disrupts positive biases (Fletcher & Kerr, 2010; Overall et al., 2012) and directly opposes the image of a warm and loyal caretaker (e.g., Abrams et al., 2003). Isolating the relationship contexts in which BS fosters positivity versus negative reactance is an important direction for future research.

Strengths, Caveats, and Future Directions

The negative bias and associated interpersonal consequences associated with men’s HS was replicated across two studies that repeatedly assessed both couple members’ perceptions and behavior across a 1-year (Study 1) and 3-week (Study 2) period. Assessing the reports of both couple members provides the means to test bias by contrasting perceptions with the reported behavior of the partner. Using partner-reports as the benchmark is the typical way to assess bias within relationships (Fletcher & Kerr, 2010; Gagné & Lydon, 2004), and this measure of bias predicts important outcomes, such as declines in relationship quality and increases in conflict (Fletcher & Kerr, 2010; Miller & Rempel, 2004; Murray et al., 2011). In the current research, we also demonstrated that biased perceptions assessed in this way predict increased negative behavior and dissatisfaction.

Nonetheless, using the partner’s reports of their own behavior as the benchmark to test the veracity of perceptions of that behavior has limitations. For example, perhaps the more aggressive behavior associated with men’s HS results in partners protectively underreporting or engaging in less negative behavior and this produces the greater bias associated with men’s HS. Additional analyses showed this was not the case: There were no significant links between men’s HS and their partner’s self-reported negative behavior. Nonetheless, self-serving biases could mean that all partners, regardless of levels of HS, generally understated their negative behavior. Accordingly, individuals generally overestimated their partner’s negative behavior in Study 1. This could be due to partner’s recalling more positive self-relevant behavior, perceiver’s recalling more negative other-relevant information, or both (Baumeister et al., 2001; Mezulis et al., 2004). When retrospective biases were minimized in Study 2, however, there was no bias evident at the sample level. Critically, across both studies, men’s HS was associated with greater overestimation of the partner’s negative behavior regardless of the average sample-level bias. Thus, even if self-serving biases influenced levels of bias, men’s HS predicted meaningful increases in negative bias which were associated with important relationship outcomes over and above the negative effects of the partner’s reported behavior.

Future investigations could extend these findings using alternative benchmarks, such as gathering ratings of behavior from third parties (e.g., friends or family) or trained observers. These methods have similar challenges because other raters may protect their own relationship by devaluing others, misconstrue the meaning of behavior based on their own experiences, and lack knowledge of the wider relationship context, which influences the meaning of behavior (Gagné & Lydon, 2004). Moreover, utilizing reports from within the relationship means we captured participants’ relationship experiences within naturally-occurring interactions and as relationship progressed across time. Experimental tasks that assess biased perceptions are also unable to test the veracity of people’s perceptions as they occur in real life or reveal how these shape ongoing relationships over time. Nonetheless, we expect that utilizing alternative methods will provide converging evidence—a good endeavor for future research.

By examining perceptions of common behaviors that frequently occur across couples’ relationships and daily life, our results also offer a powerful demonstration of the pervasive impact of men’s HS on relationships. In particular, men perceived their partner’s behavior in a more negative light regardless of how positive or negative the partner actually behaved (shown by the null interaction between HS and tracking accuracy; see Overall et al., 2012; Overall & Hammond, 2013) or how much general negativity was occurring in the relationship, as indexed by men’s own negative behavior and levels of conflict. These latter tests illustrate that men’s HS produces bias across relationship contexts and is therefore not simply driven by a heightened reactivity to conflict or a negative interpersonal orientation. As argued above, we think this bias occurs because the interdependence inherent in all relationship interactions clashes with HS-related expectations that women seek to exploit such dependence, fostering chronic biases of the partner’s behavior. Nonetheless, it is likely that such negative biases are further amplified in very threatening contexts, such as dealing with suspected infidelity or when it seems their partner is no longer committed to the relationship. The low frequency of these behaviors in the committed couples we sampled mean it is unlikely that such events were captured by our studies, but investigation of these critical events is an important direction for future research.

Despite the significant strength of assessing perceptions across the course of couples’ relationships, the correlational nature of the current studies prevents causal conclusions. There is little known about whether relationship perceptions or experiences shape the extent to which people agree with sexist ideologies. It is possible, for example, that men who perceive their romantic partners more negatively and/or have hurtful relationship experiences develop more hostile attitudes toward women (see Hart et al., 2012). However, relationship satisfaction should be a primary marker of such experiences, but relationship satisfaction (or feeling manipulated and own negative behavior) did not account for the biases associated with HS. Nonetheless, these links are likely reciprocal. For example, feeling manipulated by partners will likely strengthen beliefs that women exploit intimate contexts to control men. Identifying whether negatively biased perceptions, and its associated outcomes, bolster agreement with HS over time is an important goal for future investigations.

Finally, our samples represented a fairly conservative test of the effect of sexist attitudes because they were gathered in New Zealand, a country with relatively high levels of gender equality and relatively low endorsement of sexist attitudes (Brandt, 2011). Although endorsement of sexist attitudes differs across nations, the effects of BS and HS are typically similar, including their contribution to maintaining gender inequality (Brandt, 2011) and justification of domestic violence (Glick et al., 2002; Yamawaki et al., 2009). Thus, HS should be associated with more negatively biased perceptions of intimate partner’s behavior across nations. However, the effects of HS are likely to be more pronounced when negative stereotypes of women are more salient and when women’s relatively lower opportunity to gain power in career, legal, or governmental domains may prompt fears that women will instead seek power within relationships.

Conclusion

The current research is the first to test whether men’s endorsement of HS predicts biased perceptions of women’s behavior. The results from two longitudinal studies of couples in ongoing, committed relationships illustrated that men who more strongly endorsed HS perceived their partner’s behavior as more critical and less supportive than was merited based on their partner’s reports of that behavior. In turn, these more negatively biased perceptions were detrimental for relationships, including feeling more manipulated by partners, experiencing lower relationship satisfaction, and behaving more negatively toward partners. These findings show that the antagonistic attitudes contained within HS extend to intimate contexts and shape men’s interpretation of their female partners’ behavior. Moreover, the results highlight that the negativity they read into their relationships is one key reason why men who endorse HS are dissatisfied and behave more negatively toward intimate partners.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Royal Society of New Zealand Marsden Fund Grant (UOA0811) and University of Auckland Grants (UOA3607021; UOA3626244) awarded to Nickola C. Overall.

Notes

References

Abrams

Viki

G. T.

Masser

Bohner

(2003). Perception of a stranger and acquaintance rape: The role of benevolent and hostile sexism in victim blame and rape proclivity. Journal of Personality and Social Psychology, 84, 111-125.

Baumeister

R. F.

Bratslavsky

Finkenauer

Vohs

K. D.

(2001). Bad is stronger than good. Review of General Psychology, 5, 323-370.

Bradbury

Fincham

(1990). Attributions in marriage: A review and critique. Psychological Bulletin, 107, 3-33.

Brandt

M. J.

(2011). Sexism and gender inequality across 57 societies. Psychological Science, 22, 1413-1418.

Chen

Fiske

S. T.

Lee

T. L.

(2009). Ambivalent Sexism and power-related gender-role ideology in marriage. Sex Roles, 60, 765-778.

Dardenne

Dumont

Bollier

(2007). Insidious dangers of benevolent sexism: Consequences for women’s performance. Journal of Personality and Social Psychology, 93, 764-779.

Devine

P. G.

(1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5-18.

Downey

Feldman

(1996). Implications of rejection sensitivity for intimate relationships. Journal of Personality and Social Psychology, 70, 1327-1343.

Fletcher

G. J. O.

Kerr

P. S. G.

(2010). Through the eyes of love: Reality and illusion in intimate relationships. Psychological Bulletin, 136, 627-658.

10.

Fletcher

G. J. O.

Simpson

J. A.

Thomas

(2000). The measurement of perceived relationship quality components: A confirmatory factor analytic approach. Personality and Social Psychology Bulletin, 26, 340-354.

11.

Forbes

G. B.

Adams-Curtis

L. E.

White

K. B.

(2004). First-and second-generation measures of sexism, rape myths, and related beliefs, and hostility toward women: Their interrelationships and association with college students’ experiences with dating aggression and sexual coercion. Violence Against Women, 10, 236-261.

12.

Forbes

G. B.

Jobe

R. L.

White

K. B.

Bloesch

Adams-Curtis

L. E.

(2005). Perceptions of dating violence following a sexual or nonsexual betrayal of trust: Effects of gender, sexism, acceptance of rape myths, and vengeance motivation. Sex Roles, 52, 165-173.

13.

Gagné

F. M.

Lydon

J. E.

(2004). Bias and accuracy in close relationships: An integrative review. Personality and Social Psychology Review, 8, 322-338.

14.

Glick

Diebold

Bailey-Werner

Zhu

(1997). The two faces of Adam: Ambivalent sexism and polarized attitudes toward women. Personality and Social Psychology Bulletin, 23, 1323-1334.

15.

Glick

Fiske

S. T.

(1996). The ambivalent sexism inventory: Differentiating hostile and benevolent sexism. Journal of Personality and Social Psychology, 70, 491-512.

16.

Glick

Fiske

S. T.

(2001). An ambivalent alliance: Hostile and benevolent sexism as complementary justifications for gender inequality. American Psychologist, 56, 109-118.

17.

Glick

Sakalli-Ugurlu

Ferreira

M. C.

Aguiar

Souza

(2002). Ambivalent sexism and attitudes toward wife abuse in Turkey and Brazil. Psychology of Women Quarterly, 26, 291-296.

18.

Gottman

J. M

, (1998). Psychology and the study of marital processes. Annual Review of Psychology, 49, 169-197.

19.

Hammond

M. D.

Overall

N. C.

(2013). When relationships do not live up to benevolent ideals: Women’s benevolent sexism and sensitivity to relationship problems. European Journal of Social Psychology, 43, 212-223.

20.

Hart

Hung

J. A.

Glick

Dinero

R. E.

(2012). He loves her, he loves her not: Attachment style as a personality antecedent to men’s ambivalent sexism. Personality and Social Psychology Bulletin, 38, 1495-1504.

21.

Kenny

D. A.

Acitelli

L. K.

(2001). Accuracy and bias in the perception of the partner in a close relationship. Journal of Personality and Social Psychology, 80, 439-448.

22.

Kenny

D. A.

Kashy

D. A.

Cook

W. L.

(2006). Dyadic data analysis. New York, NY: Guilford Press.

23.

Lepore

Brown

(1997). Category and stereotype activation: Is prejudice inevitable? Journal of Personality and Social Psychology, 72, 275-287.

24.

MacKinnon

D. P.

Fritz

M. S.

Williams

Lockwood

C. M.

(2007). Distribution of the product confidence limits for the indirect effect: Program PRODCLIN. Behavior Research Methods, 39, 384-389.

25.

Masser

B. M.

Abrams

(2004). Reinforcing the glass ceiling: The consequences of hostile sexism for female managerial candidates. Sex Roles, 51, 609-615.

26.

Mezulis

A. H.

Abramson

L. Y.

Hyde

J. S.

Hankin

B. L.

(2004). Is there a universal positivity bias in attributions? A meta-analytic review of individual, developmental, and cultural differences in the self-serving attributional bias. Psychological Bulletin, 130, 711-747.

27.

Miller

P. J. E.

Rempel

J. K.

(2004). Trust and partner-enhancing attributions in close relationships. Personality and Social Psychology Bulletin, 30, 695-705.

28.

Murray

S. L.

Bellavia

G. M.

Rose

Griffin

D. W.

(2003). Once hurt, twice hurtful: How perceived regard regulates daily marital interactions. Journal of Personality and Social Psychology, 84, 126-147.

29.

Murray

S. L.

Griffin

D. W.

Derrick

J. L.

Harris

Aloni

Leder

(2011). Tempting fate or inviting happiness? Unrealistic idealization prevents the decline of marital satisfaction. Psychological Science, 22, 619-626.

30.

Noller

Ruzzene

(1991). Communication in marriage: The influence of affect and cognition. In Fletcher

G. J.

Fincham

F. D.

(Eds.), Cognition and close relationships (pp. 203-233). Hillsdale, NJ: Erlbaum.

31.

Overall

N. C.

Fletcher

G. J. O.

Kenny

D. A.

(2012). When bias and insecurity promote accuracy: Mean-level bias and tracking accuracy in couples’ conflict discussions. Personality and Social Psychology Bulletin, 38, 642-655.

32.

Overall

N. C.

Hammond

M. D.

(2013). Biased and accurate: Depressive symptoms and daily perceptions within intimate relationships. Personality and Social Psychology Bulletin, 39, 636-650.

33.

Overall

N. C.

Sibley

C. G.

(2010). Convergent and discriminant validity of the accommodation scale: Evidence from three diary studies. Personality and Individual Differences, 48, 299-304.

34.

Overall

N. C.

Sibley

C. G.

Tan

(2011). The costs and benefits of sexism: Resistance to influence during relationship conflict. Journal of Personality and Social Psychology, 101, 271-290.

35.

Rucker

D. D.

Preacher

K. J.

Tormala

Z. L.

Petty

R. E.

(2011). Mediation analysis in social psychology: Current practices and new recommendations. Social and Personality Psychology Compass, 5, 359-371.

36.

Rudman

L. A.

Heppen

J. B.

(2003). Implicit romantic fantasies and women’s interest in personal power: A glass slipper effect? Personality and Social Psychology Bulletin, 29, 1357-1370.

37.

Rusbult

C. E.

Verette

Whitney

G. A.

Slovik

L. F.

Lipkus

(1991). Accommodation processes in close relationships: Theory and preliminary empirical evidence. Journal of Personality and Social Psychology, 60, 53-78.

38.

Sibley

C. G.

Becker

J. C.

(2012). On the nature of sexist ambivalence: Profiling ambivalent and univalent sexists. European Journal of Social Psychology, 42, 589-601.

39.

Sibley

C. G.

Perry

(2010). An opposing process model of benevolent sexism. Sex Roles, 62, 438-452.

40.

Sibley

C. G.

Wilson

M. S.

(2004). Differentiating hostile and benevolent sexist attitudes toward positive and negative sexual female subtypes. Sex Roles, 51, 687-696.

41.

Sibley

C. G.

Wilson

M. S.

Duckitt

(2007). Antecedents of men’s hostile and benevolent sexism: The dual roles of Social Dominance Orientation and Right-Wing Authoritarianism. Personality and Social Psychology Bulletin, 33,160-172.

42.

West

T. V.

Kenny

D. A.

(2011). The truth and bias model of judgment. Psychological Review, 118, 357-378.

43.

Wieselquist

Rusbult

C. E.

Foster

C. A.

Agnew

C. R.

(1999). Commitment, pro-relationship behavior, and trust in close relationships. Journal of Personality and Social Psychology, 77, 942-966.

44.

Yakushko

(2005). Ambivalent Sexism and relationship patterns among women and men in Ukraine. Sex Roles, 52,589-596.

45.

Yamawaki

Ostenson

Brown

R. C.

(2009). The functions of gender role traditionality, ambivalent sexism, injury, and frequency of assault on domestic violence perception: A study between Japanese and American college students. Violence against Women, 15, 1126-1142.