VENUS AND MARS OR DOWN TO EARTH: STEREOTYPES AND REALITIES OF GENDER DIFFERENCES

Abstract

Psychological scientists, like lay people, often think in categorical dichotomies that contrast men and women and exaggerate the differences between groups. These value-laden divides tend to privilege one side over the other, often to the advantage of the scientists’ own identity group. Besides balancing perspectives in the academic marketplace of ideas, scientists can recognize the complexity of stigma. Gender, like many categories, entails two fundamental dimensions that characterize intergroup stigma (and all interpersonal perception): perceived warmth and competence. These dimensions identify groups viewed with ambivalence (e.g., traditional women are stereotypically warm but incompetent, whereas professional women are allegedly competent but cold). In gender and in other areas, psychological scientists can go beyond value-laden dichotomies and consider the fundamental, continuous dimensions along which we think about stigma.

I want to begin with a gender-related story about how I got here, and by “here” I mean contributing to this special issue and, more profoundly, entering the field of psychology. My father was a psychological scientist and that certainly had a huge impact on my entering the field. However, I also come from a long line of women, starting with my great grandmother, who have worked on the topic of gender. My great grandmother was a suffragist, and then my grandmother was also a suffragist, and her husband (my grandfather) was a women’s suffrage advocate, and my mother worked in the League of Women Voters for most of her adult life. So, I come from a long line of people who have thought a bit about gender.

In college, when I took psychology courses, I kept noticing that, in the Personality courses in particular, all the individual difference scales had a good end and a bad end. And anyone who read the scale could almost always tell which end belonged to the person who had made up the scale. At that time, most of the people making up the scales were men. I kept reading these scales and thinking, but wait, what about the other side? Is it not okay to be that, too? Is it always a deficit?

Therefore, I realized that someone had to enter the field and point out the other perspective, and that was one of my major motivations for going into psychological science. Although I did not end up focusing that much on gender, I do have some relevant thoughts to offer in this regard. First, I want to encourage us to think about how we frame gender differences; next, I consider some pros and cons of studying gender differences the way we currently do; and finally, I address how we can use two fundamental, continuous dimensions to think about gender.

How We Frame Gender Differences

Psychologists love dichotomies. They love to slice and dice a broader population into two categories. For example, a very popular dichotomy applied in the 1970s was termed “field dependence” (Witkins, Moore, Goodenough, & Cox, 1977). Field independent people were believed to be more autonomous and more able to cognitively restructure the tasks before them. Field dependent people were unduly influenced by the immediate situation. My eureka moment occurred when I learned about field dependence, which is the deficit women were believed to have. However, men were believed to be field independent, which was assumed to be clearly a good thing.

It struck me that women might be viewed as field sensitive and men as field insensitive. Suddenly, the gender differences had new meaning.

Consider some other gender dichotomies: Men are perceptually thorough, mathematically self-confident, linguistically specialized, tough-minded, physically and directly assertive, and justly moral. Perhaps these are recognizable as gender differences that are often framed using different words. Seen from the other side, how about women being perceptually fast? (Does that mean men are perceptually slow?) What about women being cautious about math instead of lacking self-confidence? What about women being generally linguistically skilled (and men being linguistically overspecialized)? Are women often more agreeable and tender minded whereas men are more tough-minded? What about women being subtly, socially assertive, if they are not directly assertive? If the word aggression is substituted for assertive—men are physically aggressive and women are socially aggressive—then neither gender comes out looking so great.

All of these distinctions are actual gender differences that fall out of a collection of meta-analyses assembled by Janet Hyde (2005). It is important to put these significant distinctions in context because they form a distinct minority of all the differences examined. Of the many gender difference studies Hyde and her colleagues could find (and these are just the published ones), 78% of the studies showed very small to at the most medium effect sizes (i.e., in the .00–.35 range for Cohen’s d). Thus, almost 80% of published gender differences are not big; the median is around .21. Moreover, in virtually all of these studies, the distribution of males and the distribution of females mostly overlap. Given this reality, why do we make decisions about people based on distributions that we assume are nonoverlapping and separate? We do so because we love dichotomies.

Some Pros and Cons of How We Study Gender Differences

If people like to talk in these kinds of dichotomies, where do the problems arise? Quite clearly, people’s values and identities matter when they do this kind of science. As scientific labels illustrate, is it field dependence or field sensitivity? Researchers’ own interests and proclivities determine the issues that researchers consider interesting, their methods, their analyses, and even the publications that researchers pursue. Reported findings follow from researchers’ own perspectives. This is not to say that researchers are politically biased and, therefore, that their science is suspect. This is just to say that researchers, like everyone else, naturally pursue what they find interesting, and what they find interesting is informed by their values and their identities.

Because of this perfectly natural tendency, I argue that we need a variety of perspectives; we need diversity among scientists. As the people in the disability movement say, “Nothing about us, without us.” Of course, there is a marketplace of ideas, and any researcher has to be up to the scientific standards to get published. As we all know, peer review is hardly pro forma. But within these agreed upon standards, psychological science needs a variety of perspectives, because people so spontaneously look at things from their own point of view.

What is the downside of emphasizing gender or other group differences? People tend to jump from a continuum to categories. They maximize the differences between categories and minimize the differences within categories. Descriptive differences can become self-fulfilling prophecies. Or they can become a case of stereotype threat in which people worry about fulfilling the stereotype with which they have been labeled. Prescriptive differences go farther: People are enjoined to follow these differences whether they fit the distinction or not. And if they do not, there is well-documented backlash and constraints on the behavior of people to fit into these dichotomous categories (for one review, see Fiske, 1998).

What is more, people are just not “either/or.” Much of my own work has focused on the ambivalence of so many of these perceptions. Given this focus on, in the case of gender, male agency and female communion, Peter Glick and I developed a theory about the different kinds of stigma that can result—the most obvious being hostility toward agentic women. But there is also a subjectively benevolent, paternalistic kind of sexism toward women who are more communal (Glick & Fiske, 1996). There is also a comparable phenomenon directed at men: There is some hostility and resentment toward male agentic dominance and some benevolence toward men’s weaknesses (Glick & Fiske, 1999).

Two Fundamental, Continuous Dimensions of Gender and Stigma

The point is that these are fundamentally ambivalent forms of stigma. Often, people do not also consider the fact that even positive attributes can be damaging when they are assigned to a whole group and expected about that group. For example, by saying women are so communal and regarding it as a great virtue, there are expectations that these attributes apply to all women, which is not true. Moreover, this ambivalence turns out to be more general that just gender relations.

The communal-agency dimensions turn out to be fundamental, going across categories well beyond gender (Fiske, Cuddy, & Glick, 2007). When people encounter another person or group, the first questions people ask are “Are they on my side or not? Are they friend or foe?” And if they are a friend, then they are warm, trustworthy, and sincere. But if they are a foe, they are cold, untrustworthy, and callous. The second fundamental dimension is whether the other(s) are able to enact those intentions—if they can, then they are competent, able, skillful, and capable. These two dimensions account for as much as 85% of the variance in interpersonal impressions and intergroup impressions (see Box 1a for examples and Fiske et al. 2007, for a review). This is an incredibly powerful principle. It is not ours alone, but we have been working on it in the inter group relations area.

Box 1.

For ease of presentation, these 2 × 2 tables reduce continuous variables represented by cluster analyses in a two-dimensional space.

a. Fundamental Dimensions With Illustrative Social Groups (e.g., Fiske et al., 2007)
	Low competence	High competence
High warmth	Older people, people with disabilities (mental or physical)	Middle-class people, Americans, heterosexuals
Low warmth	Homeless people, addicts, welfare recipients	Rich people, professionals, entrepreneurs
b. Fundamental Dimensions With Female Subtypes (Eckes, 2002, Study 1)
	Low competence	High competence
High warmth	Typical women, housewives, secretaries, “wallflowers”	Confident types, society ladies
Low warmth	Bourgeois, “chicks”	Feminists, career women, intellectual women
c. Fundamental Dimensions With Male Subtypes (Eckes, 2002, Study 2)
	Low competence	High competence
High warmth	Senior citizens, radicals, hippies, softies	Professors, intellectuals, confident types
Low warmth	Cads, punks, bums, rockers	Career men, managers, yuppies, typical men
d. Fundamental Dimensions With Gay Subtypes (Clausell & Fiske, 2005)
	Low competence	High competence
High warmth	Flamboyant, feminine	Artistic, in-the-closet
Low warmth	Leather/biker, cross-dressers	Hypermasculine, activists
e. Fundamental Dimensions With Mental Illness Subtypes (Moore & Fiske, 2005)
	Low competence	High competence
High warmth	Mental retardation, Down’s syndrome	Attention-deficit disorder, eating disorder
Low warmth	Schizophrenia, psychopathology	Depression, bipolar disorder, obsessive–compulsive disorder
f. Fundamental Dimensions With Black Subtypes (Fiske, Bergsieker, Russell, & Williams, 2009)
	Low competence	High competence
High warmth	Disabled Blacks, Poor Blacks	Christian Blacks, Elderly Blacks, Black mothers, Middle-class Blacks
Low warmth	Uneducated Blacks, “Niggas”	Educated Blacks, Rich Blacks
g. Fundamental Dimensions With Immigrant Subtypes (Lee & Fiske, 2006)
	Low competence	High competence
High warmth	Italian immigrants, Irish immigrants	Canadian immigrants, European immigrants, documented immigrants, third-generation immigrants
Low warmth	Undocumented immigrants, Mexican immigrants, African immigrants	Japanese immigrants, Chinese immigrants, Korean immigrants

Our model captures some crucial aspects of stigma, both scientists’ stigma and also lay people’s. Our warmth-by-competence space does represent the most obvious diagonal, the one most often meant by stigma. Stigma usually contrasts pride in the ingroup (“we are the best”) and contempt toward the outgroup, which in this case would be low on both warmth and competence (see Box 1a). What our model adds to this standard dichotomy are the ambivalent clusters. For one, people who are stereotypically high warmth but low competence typically include older people and people with disabilities; they all elicit pity. In contrast, people perceived to be highly competent but not so warm in the United States at this time stereotypically include people who are rich, entrepreneurs, or professionals. The data replicate across student and adult samples, across the country, and in other countries as well (Cuddy, Fiske, & Glick, 2007; Cuddy et al., 2009).

Focusing on the gender aspect of it, Thomas Eckes (2002) took this analysis of stigma one step further by looking at different kinds of female subgroups (see Box 1b). Translated from German, typical women, housewives, secretaries, and “wallflowers” fall into what might be considered “traditional” female roles; they are stereotypically warm but incompetent. In contrast to this are the feminists, lesbians, female athletes, and female professionals; they are stereotypically competent but cold. For subgroups of women, these two dimensions differentiate them in important ways. Eckes also examined different subgroups of men. Notably, the subgroups of men are more often clustered around the kinds of jobs they have (see Box 1c).

We have also found that these two fundamental dimensions appear in subgroups of gay men (Clausell & Fiske, 2005; see Box 1d). Paralleling the male and female gender subgroups, the feminine gay men are allegedly warm, but not so competent and the hypermasculine gay men are allegedly competent but not very warm. And it is apparently not good to be a leather biker—they are stereotyped as neither smart nor nice. Finally, almost no one appears to be highly warm and highly competent part of the space except perhaps artistic gay men or straight-acting ones; so, it suggests that gay men are just not seen as a likely warm and competent ingroup.

Overall, dimensions provide a more nuanced view of stigma. Groups can be stigmatized even if viewed positively on one dimension and by implication negatively on the other dimensions. Consider people with various kinds of mental illness or mental disability (Fiske, Moore, & Russell, 2006; see Box 1e). People who are viewed as mentally disabled are allegedly incompetent but nice. People with schizophrenia are viewed as neither competent nor nice. And people with bipolar disorders, as in the stereotypic mad genius, are seen as highly competent but not nice.

As for Black people, our research indicates that contemporary Americans split African Americans by social class, with poor Blacks (and poor Whites) allegedly being neither competent nor nice, but Black professionals being like other professionals: stereotypically competent but cold (Cuddy et al., 2007). Black Americans subtype themselves as well (Fiske, Bergsieker, Russell, & Williams, 2009; see Box 1f). We have also analyzed distinct subgroups of immigrants by ethnicity; some are seen as competent but not warm (e.g., Asians), some are seen as incompetent but warm (e.g., Irish), and some are seen as neither (e.g., Latinos and Africans; Lee & Fiske, 2006; see Box 1g).

These stigma dimensions matter because emotional prejudices come quickly on their tail—pity, envy, and disgust—and the discrimination directed at these different groups is quite predictable from the emotions. For example, consider the difference between being attacked and being neglected. Both kinds of discrimination occur, but the two kinds of discrimination are often aimed toward different kinds of groups (Cuddy et al., 2007). What is more, the stigma dimensions often will force trade-offs; so an outgroup can be smart but cold or nice but stupid. These ambivalent combinations are not beneficial. And some stigmas obviously are worse than ambivalent: There are groups who are seen as having no redeeming features; some data, including neuroimaging data, indicate that such people are basically dehumanized (Harris & Fiske, 2006, 2009).

Conclusion

Stigmas differ. Scientists must go beyond viewing stigma as simply “I hate them, and I love us.” Nor is it just that “my end of the scale is good, and your end of the scale is bad.” For gender, in particular and for many other groups, the similarities are often greater than the differences between the groups. The differences divide us and oversimplify a complex, textured reality. When there is stigma, it is likely to be on multiple dimensions with ambivalent and insidious results.