Sex Roles

, Volume 57, Issue 11, pp 787–799

The Interpersonal Power of Feminism: Is Feminism Good for Romantic Relationships?

Authors

    • Department of PsychologyRutgers University
  • Julie E. Phelan
    • Department of PsychologyRutgers University
Original Article

DOI: 10.1007/s11199-007-9319-9

Cite this article as:
Rudman, L.A. & Phelan, J.E. Sex Roles (2007) 57: 787. doi:10.1007/s11199-007-9319-9

Abstract

Past research suggests that women and men alike perceive feminism and romance to be in conflict (Rudman and Fairchild, Psychol Women Q, 31:125–136, 2007). A survey of US undergraduates (N = 242) and an online survey of older US adults (N = 289) examined the accuracy of this perception. Using self-reported feminism and perceived partners’ feminism as predictors of relationship health, results revealed that having a feminist partner was linked to healthier relationships for women. Additionally, men with feminist partners reported greater relationship stability and sexual satisfaction in the online survey. Finally, there was no support for negative feminist stereotypes (i.e., that feminists are single, lesbians, or unattractive). In concert, the findings reveal that beliefs regarding the incompatibility of feminism and romance are inaccurate.

Keywords

FeminismClose relationshipsFeminist stereotypesIntergroup relationsGender attitudes

Introduction

Although feminists deserve the lion’s share of credit for advancing women’s rights, college-aged adults do not identify with feminists (e.g., Aronson 2003; Buschman and Lenart 1996; Renzetti 1987; Williams and Wittig 1997), and attitudes toward them are surprisingly negative (Haddock and Zanna 1994; Renzetti 1987; Rudman and Fairchild 2007). Feminist stereotypes are also unflattering; feminists tend to be stigmatized as unattractive, sexually unappealing, and likely to be lesbians (Goldberg et al. 1975; Rudman and Fairchild 2007; Swim et al. 1999; Unger et al. 1982). The fact that women are just as prone to these views as men is particularly disturbing. It is difficult to imagine any other group stigmatizing the pioneers who struggled for their equality; for example, if African Americans disdained Rosa Parks, Martin Luther King, Jr., and other civil rights workers, it would be inconceivable and cause for alarm.

In an effort to determine why the seeds of feminism have fallen on hard ground, some authors have suggested that the “post-feminism” era has replaced women’s interest in collective power with interest in self-empowerment (Levy 2005; Riger 1993; Zucker 2004). For example, Rich (2005) described several women in her sample as hostile toward feminism because they viewed it as a movement for victims, or women who could not achieve success based on their own merit. Others have suggested that feminism is now subsumed in the language of choice, such that women can be either vanguards or traditionalists; as long as they choose their life’s path, it counts as feminist (Taylor 1992)—a view that negates the goals of the Women’s Movement. Finally, it should be noted that feminism has always been subject to swings of the cultural pendulum, with advances toward gender equity being met with backlash designed to return women to their historically low stratum in the social hierarchy (Faludi 1991; Valian 1999). What is particularly disturbing is that, by eschewing feminism, women themselves may be participating in backlash. Thus, it is important to understand the reasons why women today tend not to embrace feminism.

The goal of the present research was to follow up on evidence suggesting that both women and men avoid identifying with feminists, and supporting feminist causes, to the extent that feminism is perceived to be incompatible with heterosexual romance (Rudman and Fairchild 2007). Because these beliefs may undermine the ability to collectively advance gender equality, it was important to test their accuracy. To do so, we employed a laboratory survey of undergraduates and an online survey that included older adults who are likely to have had longer relationships, as well as greater life experience. The main objective was to examine whether heterosexual feminists (or men paired with feminists) have troubled romantic relationships, as is popularly perceived. A secondary goal was to investigate the accuracy of negative feminist stereotypes (i.e., that they are likely to be single, lesbians, or unattractive).

Feminism and Gender Relations

For intergroup relations researchers, gender is unique because men and women are intimately interdependent (Fiske and Stevens 1993; Glick and Fiske 1996). They depend on each other for sexual and emotional gratification, as well as sexual reproduction. Traditionally, they have been socialized to occupy different family roles (with men as breadwinners and women as caretakers; Eagly 1987). As a result, women have historically been consigned to dyadic more than economic power (Johnson 1976), and their principal route to status and influence has involved attracting the best possible marital partner.

The Women’s Movement sought to change that through the use of collective power, but women continue to be socialized in ways that hinder aspirations to the highest echelons of status and influence, in part by educating them in romance (Holland and Eisenhart 1990; Rudman and Heppen 2003). Cultural romantic scripts idealize women (i.e., place them on a pedestal), but they also emphasize male initiative and female passivity (Holland 1992; Impett and Peplau 2003; Sanchez et al. 2005). Indeed, women who automatically associate male partners with chivalry and heroism (e.g., Prince Charming, White Knight) also show less interest in financial independence and leadership roles, suggesting that implicit romantic fantasies can curtail women’s ambitions (Rudman and Heppen 2003). Further, the media often portrays feminists as radical man-haters, which could lead to the perception that they are lesbians who resent men (Bell and Klein 1996; Misciagno 1997). This misperception may stem from the fact that feminists have courageously challenged cultural romantic scripts (e.g., De Beauvoir 1952; Firestone 1970; Millet 1970). In her interviews with women of all ages, Sigel (1996) found ambivalence toward feminism; although women appreciated the benefits derived from the Women’s Movement, they worried it had gone too far and negatively affected relations with men. For all of these reasons, it seemed likely that feminism might be viewed as incompatible with romance and if so, it might help to account for feminism’s current lack of popularity.

To directly test this hypothesis, Rudman and Fairchild (2007) examined feminist orientations as a function of (1) the lesbian feminist stereotype and (2) beliefs that feminism creates heterosexual relationship conflict. They found that women and men alike shied away from feminism to the extent they viewed feminists as lesbians or perceived feminism to be incompatible with romance. For example, people who endorsed beliefs that “Feminism can cause women to resent men,” “Feminism can add stress to relationships with men,” and “Most men would not want to date a feminist,” were less likely to identify with feminists, to report positive attitudes toward them, and to endorse women’s civil rights (e.g., to support the Equal Rights Amendment). Additional findings showed that unattractive women were rated as more likely to be feminists than attractive women, but that this difference was wholly explained by beliefs that unattractive women are low on sex appeal or likely to be lesbians. As a result of these unfavorable beliefs, young adults may view feminism as antithetical to romance and a hindrance to their own relationships.

Overview of the Research and Hypotheses

Our primary goal was to examine the accuracy of popular beliefs that feminism is incompatible with romance and thus, to test the credibility of a factor that causes women and men alike to shy away from feminism. Are feminist women (or men with feminist partners) likely to have troubled relationships? To our knowledge, this issue has yet to be investigated. As a first step, we conducted a laboratory survey of college students (Study 1) and an online survey designed to include older adults (Study 2). Because it takes two to form a healthy relationship, we assessed participants’ own feminism and perceptions of their partners’ feminism. For example, people who are mismatched (e.g., feminist women with non-feminist partners) may have more difficulty in their relationships, compared with people who are matched with similar partners (Smith et al. 1993, 1995). Moreover, because men are typically more sexist in their attitudes than women and invested in women’s occupation of traditional roles (Glick and Fiske 1996; Haddock and Zanna 1994), non-feminist men might be particularly resentful of a female partner’s feminism, leading to poor relationship quality for them.

As a measure of feminism, we combined participants’ identification with feminism with how much they liked feminists and career women. A comparable measure assessed partners’ perceived feminism. As the first investigation of the perceived negative link between feminism and heterosexual relationship satisfaction, we examined several different components of relationship health (i.e., overall relationship quality, agreement about gender equality, relationship stability, and sexual satisfaction) to determine precisely what (if anything) leads to more conflict in feminists’ relationships and men’s relationships with feminists. Relationship quality included questions about trust and conflict within the relationship, as well as positive and negative emotions experienced within the relationship. Relationship equality assessed whether participants agreed with their partners about gender equality and the appropriate roles in the relationship. Because feminists challenge traditional gender roles, issues of gender equality may be a significant source of conflict within their relationships. Relationship stability assessed the likelihood of terminating the relationship—an important indicator of relationship health. Finally, Study 2 (which included older adults) assessed sexual satisfaction because female autonomy and independence, explicit goals of the feminist movement, are believed to create sexual conflict for men (i.e., undermine their satisfaction; Rudman and Fairchild 2007).

Examining only heterosexuals who are currently in a romantic relationship, we had the following hypotheses:
  1. Hypothesis 1

    If feminism is incompatible with romance, women who are feminists should be more likely to experience poor relationship health, compared with women who are traditionalists. That is, feminism should negatively covary with indicators of relationship health.

     
  2. Hypothesis 2

    Women in a mismatched relationship (i.e., feminists in a relationship with a non-feminist man, or traditionalists in a relationships with a feminist man) should experience more relationship turmoil than women who report sharing a similar feminist identity with their partner. In other words, feminism and partner’s feminism should interact to predict relationship health.

     
  3. Hypothesis 3

    If feminism is incompatible with romance, men in relationships with feminists should be more likely to experience poor relationship health, compared with men who are paired with traditional women. That is, partner’s feminism should negatively covary with indicators of men’s relationship health.

     
  4. Hypothesis 4

    Men in mismatched relationships should experience more relationship turmoil than men who report sharing a similar feminist identity with their partner (as for women in Hypothesis 2).

     
Finally, considerable evidence suggests that feminists are targets of stereotypical beliefs (e.g., Bell and Klein 1996; Goldberg et al. 1975; Rudman and Fairchild 2007; Misciagno 1997; Swim et al. 1999; Unger et al. 1982), but a test of their accuracy has not yet been undertaken. We combined our two samples to examine whether feminist stereotypes contain a kernel of truth.
  1. Hypothesis 5

    If feminist stereotypes are accurate, then feminist women should be more likely to report being single, lesbian, or sexually unattractive, compared with non-feminist women.

     

Study 1

The main objective was to assess the accuracy of beliefs that feminism is troublesome for romantic relationships (Rudman and Fairchild 2007). To do so, we compared self-reported feminism and perceived partners’ feminism as predictors of relationship health for undergraduate heterosexuals currently involved in a romantic relationship.

Method

Participants

Five hundred and thirteen volunteers (298 women, 215 men) participated in exchange for partial credit toward their Introductory Psychology research participation requirement. Participants who were not in a current relationship (129 women, 126 men) or who reported not being exclusively heterosexual (21 women, 14 men) were excluded for the analyses, leaving a sample of 242 volunteers (156 women, 86 men). Of these, 136 (56%) were European American, 60 (25%) were Asian American, 16 (6%) were African American, 17 (6%) were Hispanic American, and the remainder reported another ethnic identity.

Materials

Self and Partner’s Feminism

Participants responded to two items using 6-point scales ranging from 1 (strongly disagree) to 6 (strongly agree). The items were, “I am a feminist” and “My partner is a feminist.” To measure attitudes, they also reported (on four separate scales) how warmly they and their partners felt toward feminists and career women on thermometer measures ranging from 1 (very cold) to 10 (very warm). A principal components factor analysis showed that feminist labeling and attitudes toward feminists and career women yielded one factor, so we averaged these three items to form a single index of feminism (eigenvalue = 1.83, accounting for 61% of the variance, α = .71). A similar analysis indicated the comparable partner variables should be averaged to form a single index of partner’s feminism (eigenvalue = 1.52, accounting for 52% of the variance, α = .72).

Relationship Health

Participants responded to 12 items pertaining to their relationships on scales ranging from 1 (never) to 6 (very often). Results of a principal components factor analysis (using varimax rotation) revealed three factors. Six items formed the first factor (eigenvalue = 4.97, variance accounted for = 41%, all factor loadings >.60). The items were, “How often do you feel relaxed with your partner?,” “How often do you and your partner laugh together?,” Do you confide your deepest feelings to your mate?,” “Do you and your mate share similar interests?,” “How often do you and your partner quarrel?,” and “How often do you and your partner get on each other’s nerves?” After recoding, items were averaged to form the relationship quality index, on which high scores reflected greater quality, α = .73.

Four items formed the second factor (eigenvalue = 1.67, variance accounted for = 14%, all factor loadings >.78). The items were, “How often have you considered terminating your relationship?,” “How often do you think your romantic relationship has a good future?,” “How often do you think about finding another partner?,” and “How often do you think that things between you and your partner are going well?” After recoding, items were averaged to form the relationship stability index, on which high scores reflected greater stability, α = .89.

Two items formed the third factor (eigenvalue = 1.21, variance accounted for = 10%, all factor loadings >.76). The items assessed conflict based on disagreement about gender equality (“How often do you and your partner disagree about gender equality?” and “How often do you and your partner disagree about your role in the relationship?”) These items were recoded and averaged to form the gender equality index, on which high scores reflect agreement about gender equality, r(240) = .42, p < .001.

Demographics

In addition to gender and race, participants reported their sexual orientation and relationship status (used for screening). They also indicated the length of their relationship on a scale ranging from 1 (less than a year) to 6 (over 20 years). On average, participants reported being in their current relationship for 21 months (range = 6 months to 4 years).

Procedure

Participants were escorted to individual booths equipped with a desktop PC by an experimenter, who then administered instructions and started Inquisit, a psychological software program. Measures were administered in the following order: relationship status, relationship length, followed by (in random order): relationship quality, relationship equality, and relationship stability. Feminist identity and attitudes toward feminists and career women were then assessed, followed by the remaining demographic variables (gender, race, and sexual orientation). Within each measure, items were randomly presented. Participants who were not in a current relationship were instructed to think of their most recent past relationship during measures that asked about their relationship and partner’s feminist identity and attitudes. As noted, these participants were excluded from the analyses to allow us to focus on current relationship health. Following completion of the survey, participants were thanked and debriefed.

Results and Discussion

Gender Differences

Analyses were conducted only on heterosexual women and men who were currently in a relationship. Table 1 shows descriptive statistics for Study 1’s variables, separately by gender. Not surprisingly, women scored higher on the feminism index than did men, resulting in a large effect size for this gender difference. In addition, women reported more agreement regarding relationship equality than men, yielding a small effect size. This suggests that men experience more conflict regarding their role in the relationship and gender equity issues, compared with women. There were no reliable gender differences for the remaining variables, including perceived partner’s feminism and relationship length.
Table 1

Gender differences (Study 1).

Measure

Women

Men

Group Difference

 

M

SD

M

SD

t

d

Feminism

6.20

1.37

4.90

1.13

7.62**

.92

Partner’s Feminism

5.43

1.20

5.37

1.39

.38

.04

Relationship Quality

4.77

.71

4.62

.71

1.57

.21

Relationship Equality

4.89

1.04

4.60

1.12

2.03*

.28

Relationship Stability

4.85

1.06

4.69

1.07

1.06

.15

Relationship Length

1.80

2.29

1.40

1.24

1.55

.18

Ns = 156 women and 86 men. The range for the feminism and partner’s feminism indexes was 1–10. Relationship quality, equality, and stability were assessed on a 1 (never) to 6 (very often) scale. Relationship length was assessed on a 1 (less than one year) to 6 (more than 20 years) scale. Effect sizes (Cohen’s d) were computed using the pooled standard deviation term for each measure. By convention, small, medium, and large effect sizes correspond to .20, .50, and .80, respectively (Cohen 1988).

*p < .05.

**p < .01.

Correlations Among Variables

Table 2 shows the bivariate relationships among Study 1’s variables, for women and men. For both genders, feminism and partner’s feminism were highly related, reflecting a similarity effect (Byrne 1971). There were also substantial correlations among the relationship health variables, in support of their construct validity. The top half shows that for women, feminism was negligibly related to relationship health variables, yielding no support for Hypothesis 1. By contrast, partner’s feminism was positively related in each case. Relationship length was negatively related to relationship equality, suggesting that the longer women were in relationships, the more disagreement they experienced regarding fairness and gender roles. The bottom half shows that for men, partner’s feminism was negligibly related to their relationship health, yielding no support for Hypothesis 3. However, men’s feminism was positively related to their relationship equality scores. No other relationships not already noted reached significance.
Table 2

Correlations among variables (Study 1).

 

1

2

3

4

5

1. Feminism

 

.57**

.11

−.04

.04

2. Partner’s feminism

.68**

 

.26**

.19

.20*

3. Relationship quality

.10

−.07

 

.47**

.78**

4. Relationship equality

.25*

−.04

.50**

 

.41**

5. Relationship stability

.12

.02

.73**

.40**

 

6. Relationship length

.02

.06

−.10

−.16

−.01

Ns = 156 women and 86 men. Correlations for women are located above the diagonal; correlations for men are located below the diagonal.

*p < .05.

**p < .01.

Predicting Women’s Relationship Health

Hypothesis 1 states that if feminism were incompatible with romance, feminist women should experience poor relationships, compared with non-feminists. As noted, Table 2 does not support this hypothesis. Nonetheless, as suggested in Hypothesis 2, women who are mismatched on feminism with their partners may have more difficulty in their relationships, compared with women who are matched with similar men. If so, we would expect Feminism × Partner’s Feminism interactions. Due to sample variability in relationship length, and its covariation with relationship equality for women, we included it as an additional moderator to examine whether feminists in longer, as opposed to shorter, relationships might be more inclined to suffer poor relationship health.

To investigate these issues, we standardized all variables and submitted the relationship health variables to hierarchical regression analyses that used relationship length, feminism, and partner’s feminism as main effects, followed by the Feminism × Partner’s Feminism interaction term, followed by the interactions involving relationship length. Because results for each step were similar, we report only the final step in Table 3. (To improve the table’s readability, we excluded the interaction terms with relationship length, which were nonsignificant; see the table note).
Table 3

Regression analyses for women and men (Study 1).

Women

β

t

Men

β

t

Relationship quality (.31)

  

Relationship quality (.34)

  

Relationship length

−.01

.06

Relationship length

−.01

.02

Feminism

−.11

1.14

Feminism

.24

1.61

Partner’s feminism

.33

3.02**

Partner’s feminism

−.20

1.37

Self × partner’s feminism

−.03

.33

Self × partner’s feminism

.16

1.17

Relationship equality (.33)

  

Relationship equality (.45)

  

Relationship length

−.16

1.79

Relationship length

−.18

1.32

Feminism

−.20

2.03*

Feminism

.44

3.15**

Partner’s feminism

.32

3.27**

Partner’s feminism

−.29

2.00*

Self × partner’s feminism

.08

.77

Self × partner’s feminism

.19

1.52

Relationship Stability (.26)

  

Relationship stability (.20)

  

Relationship length

.11

1.21

Relationship length

−.08

.52

Feminism

−.09

.91

Feminism

.14

.94

Partner’s feminism

.29

2.88**

Partner’s feminism

−.03

.20

Self × partner’s feminism

.02

.14

Self × partner’s feminism

.17

1.24

Ns = 156 women and 86 men. Standardized regression coefficients are reported in the table. Overall Rs for each equation are shown in parentheses. Interactions not shown with relationship length were ns for women (all ps > .08) and for men (all ps > .41).

*p < .05.

**p < .01.

The left side of Table 3 reports the results for women. As can be seen, feminism was not a predictor of relationship health, with the exception of a negative link to agreement about relationship equality, β = −.20, p < .05. In contrast, relationship quality, equality, and stability were each reliably and positively linked to partner’s feminism, all βs > .29, ps < .01. The Feminism × Partner’s Feminism interaction terms were uniformly weak. That is, women paired with feminist men reported better relationship health, irrespective of their own feminism. Thus, there was no support for Hypothesis 2.

In sum, Study 1’s findings do not support perceptions that feminist women suffer from poor relationships, although they tended to have less agreement about equality issues than non-feminists. By contrast, results showed that male partners’ perceived feminism was a positive predictor of relationship quality, equality, and stability. Because self and partner’s feminism were strongly related, feminism may indirectly promote relationship health, through the selection of like-minded partners.

Predicting Men’s Relationship Health

The right side of Table 3 reports the results for men. If feminism is incompatible with romance, men with feminist partners should be more likely to experience poor relationships, compared with men who have non-feminist partners (Hypothesis 3). In addition, men who are mismatched on feminism with their partners might show decreased relationship health (Hypothesis 4). However, Table 3 does not support these hypotheses, with one exception. Men paired with feminist partners reported less agreement about equality issues, β = -.29, p < .05. This result substantiates women’s reports that if they were a feminist, they disagreed about equality in the relationship. However, feminist men reported more agreement about equality issues, β = .44, p < .01. This result substantiates women’s reports that having a feminist partner decreased conflict regarding equality. No other results reached significance, including the Feminism × Partner’s Feminism interactions. Thus, Hypothesis 4 was not supported.

In sum, Study 1’s findings do not support perceptions that men paired with feminist women suffer from poor relationships, although they tended to have less agreement about equality issues than men paired with non-feminists. In addition, feminist men were less likely to experience conflict regarding relationship equality. Because self and partner’s feminism did not interact, it appears that being a male feminist and having a feminist partner may have opposing effects on agreement about relationship equality. These results are the mirror image of women’s reports, and because they are the only indication that feminism troubles relationships, they will be important to replicate in Study 2.

Study 2

Although Study 1’s results overwhelmingly suggest that feminism does not predict poor romantic relationships, our findings are limited to college-aged adults. Because young adults tend to be relatively sexually inexperienced, we also did not examine the accuracy of beliefs that a female partner’s feminism can decrease sexual satisfaction for men. Specifically, Rudman and Fairchild (2007, Study 3) found that women and men who endorsed beliefs such as “men perform better sexually when they are in charge” and “romance depends, in part, on men being in charge,” showed low enthusiasm for feminism. This suggests that female assertiveness and autonomy, attributes that are instrumental for gender equality, are perceived as promoting sexual conflict. Study 2 afforded a check on the accuracy of this perception.

More generally, we employed an online study to examine our hypotheses using adults who have more life experience, and thus may be more likely to show an incompatibility between feminism and romantic relationships. One reason to expect this pattern is the fact that older women may have come of age during the second wave of feminism (the first wave was suffrage), which strongly promoted female independence from men (e.g., through the popular slogan, “A woman without a man is like a fish without a bicycle”). By contrast, young women grew up believing that “women can have it all” (i.e., a successful marriage, family, and a high-powered career). Persons coming of age in a particular historical period share a similar consciousness, and can be thought of as political generations (Schneider 1988). Finally, young women’s experience is inarguably limited, compared with older women, vis-à-vis intimate relationships, balancing them with careers, and with sexual discrimination (Frieze et al. 1991). It was possible that older feminist women would have expectations for gender fairness that, if not met, could breed dissatisfaction. They may also have higher expectations of their partners (e.g., for emotional support and intimacy), compared with non-feminists. Over time, if these expectations are not fulfilled, disappointment may result (McNulty and Karney 2004; Sedikides et al. 1994; Vangelisti and Daly 1997). In sum, in Study 2 we sought a more diverse sample to further examine the influence of feminism on romantic relationships.

Method

Participants

Four hundred and seventy-one volunteers (327 women, 144 men) were recruited from various web sites (age range = 18 to 65 years). People who were not in a current relationship (91 women, 55 men) or who reported not being exclusively heterosexual (42 women, 15 men) were excluded for the focal analyses, leaving a sample of 289 volunteers (208 women, 81 men) Of these, 209 (72%) were White, 28 (10%) were Asian, 16 (6%) were Hispanic, and the remainder reported another ethnic identity. In addition, 264 (90%) were US residents. Their mean level of education was 14 years (range = 11 to 18 years). Their mean age was 26; SD = 9.00 (women’s M = 26; range = 18–65; men’s M = 27; range = 18–55). Because not all respondents completed every measure, degrees of freedom vary in the reported analyses.

Materials

Self and Partner’s Feminism

We used Study 1’s measures to assess feminism and partner’s feminism. As in Study 1, principal components factor analyses indicated one factor for self-reported feminism, (eigenvalue = 1.95, accounting for 65% of the variance, α = .69), and one component for perceived partner’s feminism (eigenvalue = 1.83, accounting for 63% of the variance, α = .71).

Relationship Health

We adopted Study 1’s relationship quality and stability measures, αs = .81 and .90, respectively. We added an additional item to increase the reliability of the gender equality index (“Do you and your partner agree about women’s rights?”). This item was averaged with “Do you and your partner agree about gender equality?” and “Do you and your partner agree about your role in the relationship?” to form the gender equality index, α = .78, on which high scores reflect greater agreement about gender equality.

Finally, we added three items to measure sexual satisfaction, “My relationship is sexually satisfying,” “The sexual side of my relationship could use improvement,” and “How often have you considered having a sexual relationship with someone other than your partner?” After recoding, reliability analyses suggested averaging these items to form the sexual satisfaction index, α = .70, on which high scores reflect greater sexual satisfaction.

Demographics

In addition to age, gender, race, and education, participants reported their sexual orientation and relationship status (used for screening). They also indicated the length of their relationship. On average, participants reported being in their current relationship for approximately 4 years (range = 6 months to 20 years).

Procedure

The study was administered via the Rutgers University Implicit Social Cognition Lab website between March 16 and December 27, 2006. Participation was open to the public and was completely voluntary. Participants were recruited from various websites, including Craig’s List, Social Psychology Network, and Psychology Research on the Net. In addition, requests for participants were posted using Google AdWords and on forums for various Yahoo! and Google Groups.

After reading the consent form, participants were directed to an interactive Flash movie that administered the survey. Demographic information was assessed first (gender, race, sexual orientation, age, education level, and US residency), followed by relationship status and relationship length. Participants then completed the following indexes: relationship quality, relationship equality, relationship health, and sexual satisfaction. Feminist identity and attitudes toward feminists and career women were then assessed.

Results and Discussion

Gender Differences

As in Study 1, our focal analyses were conducted only on heterosexuals who were currently in a relationship. Table 4 shows descriptive statistics for Study 2’s variables, separately by gender. As in Study 1, women scored higher than men on the feminism index. Unlike Study 1, women perceived less feminism in their partners than did men. Further, women did not report more agreement regarding relationship equality. Thus, Study 1’s suggestion that men experience more conflict in the relationship regarding gender roles and equality was not replicated. New to Study 2, women reported greater sexual satisfaction than men, resulting in a moderate effect size. There were no reliable gender differences for the remaining variables.
Table 4

Gender differences (Study 2).

Measure

Women

Men

Group Difference

 

M

SD

M

SD

t

d

Feminism

6.18

1.48

5.42

1.52

3.82**

.50

Partner’s feminism

5.28

1.42

5.73

1.58

2.28*

−.31

Relationship quality

4.01

1.11

4.12

.96

.73

−.10

Relationship equality

5.06

1.02

4.96

.95

.74

.10

Relationship stability

4.48

1.31

4.17

1.33

1.78

.23

Sexual satisfaction

4.55

1.17

4.00

1.37

3.36**

.44

Relationship length

4.01

4.79

4.20

5.44

.29

−.03

Age

26.25

8.80

27.41

10.39

.95

−.12

Education

3.30

1.01

3.26

.89

.33

.04

Ns range from 204–208 women and from 79–81 men. The range for the feminism and partner’s feminism indexes was 1–10. Relationship quality, equality, and stability were assessed on a 1 (never) to 6 (very often) scale. Relationship length was assessed on a 1 (less than 1 year) to 6 (more than 20 years) scale. Effect sizes (Cohen’s d) were computed using the pooled standard deviation term for each measure. By convention, small, medium, and large effect sizes correspond to .20, .50, and .80, respectively (Cohen 1988). Education was reported categorically. Means for women and men translate to approximately 14 years each.

*p < .05.

**p < .01.

Correlations Among Variables

Table 5 shows the bivariate relationships among Study 2’s variables, for women and men. These relationships are remarkably similar to those shown in Table 2, for Study 1. For women, feminism was again negligibly related to relationship health, whereas partner’s feminism was positively related to these variables (including sexual satisfaction). As in Study 1, feminism and partner’s feminism were strongly related. New to Study 2, sexual satisfaction positively covaried with relationship health for both genders, with the exception that men showed a weak link between sexual satisfaction and gender equality. Also new to Study 2, relationship length was negatively related to relationship quality and sexual satisfaction for both genders. For women, it was also negatively related to relationship stability. Unlike Study 1, women did not show a negative link between relationship length and agreement about gender equality. The overall pattern suggests that relationship length should be included as a moderator in the remaining analyses. Because it was strongly related to participants’ age (rs = .75 and.60, for men and women, respectively, ps < .001), and to level of education for both genders (both rs > .50, ps < .001), we did not include age or education to avoid multicollinearity and to preserve statistical power.
Table 5

Correlations among variables (Study 2).

 

1

2

3

4

5

6

7

1. Feminism

 

.69**

−.08

−.04

−.03

−.01

−.05

2. Partner’s feminism

.74**

 

.16*

.28**

.17*

.21**

−.07

3. Relationship quality

.01

.05

 

.60**

.80**

.68**

−.30**

4. Relationship equality

.12

.01

.48**

 

.58**

.41**

−.08

5. Relationship stability

.03

.15

.69**

.37**

 

.65**

−.17*

6. Sexual satisfaction

.05

.21

.58**

.12

.43**

 

−.40**

7. Relationship length

−.02

−.14

−.23*

−.19

.04

−.41**

 

Correlations for women are located above the diagonal; correlations for men are located below the diagonal.

*p < .05.

**p < .01.

Predicting Women’s Relationship Health

We followed Study 1’s strategy for predicting relationship health variables. Table 6 shows that, echoing Study 1, women reported greater relationship quality, equality, and stability to the extent that their partner was a feminist; new to Study 2, they also reported greater sexual satisfaction, all βs > .29, ps < .01. Thus, the importance of having a feminist partner for women’s relationship health was replicated in Study 2. However, and in stark contrast to Study 1, we found support for Hypothesis 1 in Study 2, with feminism a negative predictor of women’s relationship quality, equality, stability, and sexual satisfaction, all βs < −.23, ps < .01. Relationship length did not qualify these findings, although it was a negative predictor of quality, stability, and sexual satisfaction (echoing Table 5). As in Study 1, the Feminism × Partner’s Feminism interaction terms were negligible, suggesting no support for Hypothesis 2. Thus, it appears that feminism and partner’s feminism may have opposing effects on women’s relationship health.
Table 6

Regression analyses (Study 2).

Women

β

t

Men

β

t

Relationship quality (.42)

  

Relationship quality (.32)

  

Relationship length

−.30

3.81**

Relationship length

−.16

1.00

Feminism

−.35

3.78**

Feminism

−.12

.68

Partner’s feminism

.39

4.11**

Partner’s feminism

.13

.72

Self × partner’s feminism

.01

.19

Self × partner’s feminism

.22

1.83

Relationship equality (.43)

  

Relationship equality (.33)

  

Relationship length

−.09

1.21

Relationship length

−.04

.24

Feminism

−.43

4.67**

Feminism

.27

1.45

Partner’s feminism

.57

6.03**

Partner’s feminism

−.20

1.18

Self × partner’s feminism

.04

.52

Self × partner’s feminism

.07

.43

Relationship stability (.33)

  

Relationship stability (.33)

  

Relationship length

−.24

2.90**

Relationship length

.11

.68

Feminism

−.24

2.46*

Feminism

−.27

1.54

Partner’s feminism

.30

3.02**

Partner’s feminism

.40

2.26*

Self × partner’s feminism

.03

.35

Self × partner’s feminism

.18

1.50

Sexual satisfaction (.49)

  

Sexual satisfaction (.48)

  

Relationship length

−.40

5.38**

Relationship length

−.41

2.74**

Feminism

−.24

2.72**

Feminism

−.20

1.26

Partner’s feminism

.32

3.51**

Partner’s feminism

.33

1.99*

Self × partner’s feminism

.02

.31

Self × partner’s feminism

.11

.32

Standardized regression coefficients are shown. Overall Rs for each equation are shown in parentheses. Interactions not shown with relationship length were ns for women (all ps > .12) and for men (all ps > .14).

*p < .05.

**p < .01.

Predicting Men’s Relationship Health

In Study 1, feminist men showed more agreement about gender equality in the relationship, whereas, in partial support of Hypothesis 1, men paired with feminist partners showed less agreement about gender equality. Table 6 shows we did not replicate this pattern in Study 2, using a more diverse sample. Instead, having a feminist partner predicted men’s relationship stability, β = .40, p < .05, and sexual satisfaction, β = .33, p < .05. Thus, in contrast to Hypothesis 3 (and to prevailing beliefs; Rudman and Fairchild 2007), men may benefit from having a feminist partner–just as women do. As in Study 1, the Feminism × Partner’s Feminism interaction terms were weak. Thus, there was no support for Hypothesis 4. Finally, Table 6 echoes Table 5 in that relationship length negatively predicted men’s sexual satisfaction. There were no other reliable main effects, and the interactions involving relationship length were nonsignificant.

In sum, Study 2 provided evidence that feminist partners may benefit men’s relationship stability and their sexual satisfaction, contradicting negative beliefs about female assertiveness and autonomy creating problems in the bedroom (Rudman and Fairchild 2007). Study 2 also replicated and extended Study 1’s results by showing that male partners’ feminism was a positive predictor of women’s relationship quality, equality, stability, and sexual satisfaction. By contrast, the finding that women’s own feminism negatively predicted these variables was new to Study 2. Although it is in line with our suggestion that feminists with more life experience may be dissatisfied in their relationships, relationship length (a proxy for age) did not modify this pattern. Analyses using age as a moderator instead of relationship length also showed no reliable interaction effects. Puzzled by these null findings, we turned to the possibility that the negative relationships between women’s feminism and relationship health variables stemmed from a statistical artifact.

Adjusting for Suppressor Variable Effects

Because Table 5 shows that feminism and partner’s feminism were highly related for women (r = .69), whereas feminism was not bivariately related to relationship health variables, we suspected a suppressor variable effect. A suppressor variable is indicated when zero-order correlations between it and the criterion variable are weak (in this case, feminism and relationship health variables), but strengthened after accounting for a predictor variable that is related to both it and the criterion variable—in this case, partner’s feminism (Darlington 1968). Adding the suppressor variable to the equation increases the relationship between the criterion variable and the first predictor (partner’s feminism) because it “suppresses, or controls for, irrelevant variance, that is, variance that is shared with the predictor and not related to the criterion” (Pedhazur 1982, p. 104).

Typically, the direction of the suppressor variable’s effect is negative; as a result, people high on this variable are discounted, whereas people who score low are weighed more heavily. Specifically, when using standardized variables, women who score above the feminism mean receive positive z scores, whereas those who score below receive negative z scores. When submitting scores to a regression equation, multiplying positive z scores by a negative beta effectively lowers women who scored high on feminism, whereas multiplying negative z scores by a negative beta raises women who scored low on feminism. Thus, “people who are high on the suppressor variable are penalized, so to speak, for being high, whereas those who are low on the suppressor variable are compensated for being low” (Pedhazur 1982, p. 105). In Study 2, women’s feminism was likely a suppressor variable for the relationship between partner’s feminism and the relationship health variables.

Table 7 presents evidence that supports this hypothesis. Columns 1–2 show that the zero-order correlations are negligible between feminism and the relationship health variables, whereas they are positive between partner’s feminism and the relationship health measures (see also Table 5, repeated here for illustration purposes). Columns 3–4 show the results of bivariate regressions using these two original variables. As can be seen, the positive zero-order relationship for partner’s feminism translates to a much larger beta after accounting for own feminism, which becomes a reliably negative predictor. These effects are classic hallmarks of suppressor variables.
Table 7

Regression analyses for women adjusting for suppressor variable effects (Study 2).

 

1

2

3

4

5

6

Feminism

Partner’s Feminism

Feminism

Partner’s Feminism

Feminism

Residualized Partner’s Feminism

r

r

β

β

β

β

Relationship quality (.31)

−.08

.16*

−.36**

.41**

−.08

.30**

Relationship equality (.43)

−.04

.28**

−.43**

.58**

−.04

.43**

Relationship stability (.25)

−.03

.17*

−.25**

.33**

−.03

.25**

Sexual satisfaction (.29)

−.01

.21**

−.25**

.37**

−.01

.29**

Standardized regression coefficients are shown. Columns 1–2 show the bivariate correlations. Columns 3–4 show suppressor variable effects when feminism and partner’s feminism are used as predictors. Columns 5–6 show results when feminism and residualized partner’s feminism (adjusting for feminism, the suppressor variable) are used as predictors. Overall Rs for the analyses in Columns 5–6 are shown in parentheses.

*p < .05.

**p < .01.

We then residualized the partner’s feminism rating, which controls for own feminism rating. In essence, this procedure “purifies” the partner’s feminism rating of irrelevant variance that is not related to criterion variables (because own feminism is not related to them). In our case, the result is a relationship between partner’s feminism and relationship health that is cleansed of a “halo” or similarity effect (i.e., rating the partner as similar to one’s self; Henik and Tzelgov 1985). Results of a bivariate regression using this residualized measure are shown in the last two columns of Table 7. As can be seen, partner’s feminism remained a predictor of relationship health measures, whereas own feminism was not (indeed, the betas are identical to the zero-order correlations shown in Column 1). These analyses suggest that women’s own feminism does not, in fact, have a negative influence on relationship health. Instead, it was acting as a suppressor variable-leading to a pattern of results that reflect a statistical artifact rather than an actual relationship between feminism and relationship health. The fact that suppressor variables emerge in relatively large samples (Pedhazur 1982) provides a reason why this effect would appear for Study 2’s women (N = 208), but is less likely to emerge for Study 1’s women (N = 156) or the men in either study (both Ns < 87).

Are Feminist Stereotypes Accurate?

Because we wished to avoid memory problems and possible group differences based on sexual orientation, we selected only heterosexual participants who reported being in a current relationship for our focal analyses. Nonetheless, we had sufficient data from both samples regarding relationship status and sexual orientation to check on the accuracy of beliefs about feminists. Negative stereotypes suggest they are likely to be single, lesbian, and unattractive, but are they? As a means of assessing attractiveness, participants in both samples also responded to three items, “I consider myself to be attractive,” “People often tell me I am attractive,” and “People compliment me on my looks” on scales ranging from 1 (strongly disagree) to 6 (strongly agree). They also rated their popularity with the other gender. The items were, “It is not difficult for me to get a date,” “I am frequently hit on for sex,” “I seem to be very popular with the opposite sex,” and “I was popular (dating-wise) in high school.” A principal components analysis suggested these seven items formed a single factor in both the lab and online studies (eigenvalues = 3.84 and 3.08, accounting for 68% and 61% of the variance, respectively). Therefore, they were averaged to form a single index of sex appeal, α = .84 (laboratory study) and α = .83 (online study).

After establishing comparable results for both samples, we combined them (N = 1259, 784 women) to present the findings. Among women, there were virtually no relationships between feminism and heterosexuality, r = -.07, ns, or sex appeal r = -.03, ns. There was a small but reliable tendency for female feminists to be more (rather than less) likely to be in a relationship, r = .10, p < .01. Thus, we found no support for Hypothesis 5. On the contrary, negative feminist stereotypes appear to be inaccurate, and thus, their unfavorable implications for relationships unfounded. Comparable analyses for men revealed negligible relationships between feminism and relationship status, r = .03, ns, and feminism and sex appeal, r = −.03, ns. There was a small but reliable tendency for male feminists to report being homosexual or bisexual, as opposed to heterosexual, r = -.13, p < .01.

General Discussion

In two studies, we examined the credibility of negative beliefs about feminists vis-à-vis their relationships with men (Goldberg et al. 1975; Rudman and Fairchild 2007; Swim et al. 1999; Unger et al. 1982). Combining the lab and online samples, we found no support for the accuracy of stereotypes suggesting that feminist women are likely to be lesbian, single, or sexually unattractive–in fact, they were more likely to be in a romantic relationship than non-feminist women. Thus, we found no evidence for the accuracy of negative stereotypes that, if true, would likely impinge on women’s relationships with men.

Feminism and Relationship Health

The primary goal was to investigate whether feminism predicts conflict in heterosexual relationships, because past research suggested that women and men alike shy away from feminism partly for this reason (Rudman and Fairchild 2007). Restricting our analyses to heterosexuals in a current romantic relationship, we found scant evidence that women’s feminism troubles their intimate relationships. In the laboratory study, feminist women did not report decreased relationship quality and stability, although they tended to report more conflict regarding equality in the relationship, compared with non-feminist women. In the online study, feminist women appeared to report decreased relationship health, including sexual satisfaction, but only after accounting for partner’s feminism. Subsequent analyses showed these findings were due to a suppressor variable effect. Thus, there was virtually no evidence that feminism is incompatible with romance.

In fact, we consistently found that feminist men are important for women’s relationship health. In each study, women reported greater relationship quality, equality, and stability to the extent they perceived their partner to be a feminist; they also reported greater sexual satisfaction in Study 2. Thus, feminist male partners may be important for healthy romantic relationships. Because feminist women tend to select them as partners, feminism for women may have an indirectly positive influence on their relationships.

Feminism may also be healthy for men’s relationships. First, feminist men in Study 1 reported greater agreement about relationship equality. Second, men in Study 2 reported greater relationship stability and sexual satisfaction to the extent their partner was a feminist. Although results for Study 1 suggested that female partners’ feminism negatively predicted agreement about relationship equality, Study 2’s men did not replicate this finding. Thus, the overall pattern suggests that for men, feminism (for self and partners) may be beneficial for their relationships, rather than problematic.

Limitations and Future Directions

Because having a feminist partner was a consistently positive predictor of women’s relationship health, and a positive predictor of men’s relationship stability and sexual satisfaction in Study 2, it is important to investigate the dynamics of this benefit. For example, future research might investigate whether feminists tend to select like-minded, feminist partners or whether they shape their partner’s beliefs. A related question concerns predictors of male feminism—to date, an under-investigated topic. In Study 2, education was positively related to men’s feminism, r(136) = .26, p < .01, as it was for women’s feminism, r(313) = .17, p < .01. Thus, higher education may promote feminism for both genders. But there are no doubt other important developmental factors that should be examined, including parental and peer attitudes toward feminism.

Moreover, the characteristics of relationships involving at least one feminist partner should be investigated to determine how feminism bestows relationship health. For example, it might be the case that feminist men are more supportive of their female partner’s ambitions than traditionalists are. Consistent with this view, male partner’s feminism was related to women’s education level in Study 2, r(204) = .21, p < .05. The comparable relationship for men was weak, r(77) = .07, ns, suggesting that other factors underlie men’s satisfaction with feminist partners. For example, men with feminist partners may enjoy having a partner with which to share the economic burden of maintaining a household.

In the present research, we relied on people’s assessment of their partner’s feminism. Future research should collect data from both partners to test the accuracy of these assessments. Although researchers have found that predicting relationship stability is not significantly improved by having both partners’ assessments (Attridge et al. 1995), whether this would hold true for perceptions of feminism is unclear. Nonetheless, our objective of testing whether perceived partner feminism disrupts men’s relationships was not undermined by our strategy.

The fact that feminists are unfairly stereotyped suggests a political motive underlying negative beliefs. Whenever women challenge male dominance, they are likely to be targeted for abuse, and particularly along sexual dimensions, perhaps to discourage other women from embracing feminism and collective power (Faludi 1991). Because this strategy appears to be effective (Rudman and Fairchild 2007), it will be important for future research to examine whether educating people might alleviate their concerns that the Women’s Movement has disrupted heterosexual relations. Far from supporting beliefs that feminism and romance are “oil and water,” we found that having a feminist partner was healthy for both women’s and men’s intimate relationships. Contrary to popular beliefs, feminism may improve the quality of relationships, as opposed to undermining them.

Acknowledgment

This research was partially supported by Grants BCS-0109997 and BCS-0417335 from the National Science Foundation.

Copyright information

© Springer Science+Business Media, LLC 2007