Trait by trait, sexes don’t differ much

U. ROCHESTER (US) — Forget Mars and Venus. Both men and women are from Earth, according to a new study that shows the sexes aren’t so dissimilar psychologically. 

From empathy and sexuality to science inclination and extroversion, statistical analysis of 122 different characteristics involving 13,301 individuals shows that men and women, by and large, do not fall into different groups.


On physical characteristics, like strength (top graph), men and women fall into distinct groups with very little overlap. But for most psychological attributes, including masculine attitudes (lower graph), variability within each sex and overlap between the sexes is extensive. The physical strength graph shows statistical analysis of the scores for the National Collegiate Athletic Association’s long jump, high jump, and javelin throw competitions. The masculinity-assertiveness graph is based on self-reported measures of competitiveness, decisiveness, sense of superiority, persistence, confidence, and the ability to stand up under pressure. (Credit: U. Rochester)

In other words, no matter how strange and inscrutable your partner may seem, their gender is probably only a small part of the problem.

“People think about the sexes as distinct categories,” says Harry Reis, professor of psychology at the University of Rochester and a co-author on the study to be published in the February issue of the Journal of Personality and Social Psychology.

“‘Boy or girl?’ is the first question parents are asked about their newborn, and sex persists through life as the most pervasive characteristic used to distinguish categories among humans.”

A closer look

But the handy dichotomy often falls apart under statistical scrutiny, says lead author Bobbi Carothers, who completed the study as part of her doctoral dissertation at Rochester and is now a senior data analyst for the Center for Public Health System Science at Washington University in St. Louis.

For example, it is not at all unusual for men to be empathic and women to be good at math—characteristics that some research has associated with the other sex, says Carothers. “Sex is not nearly as confining a category as stereotypes and even some academic studies would have us believe,” she adds.

The authors reached that conclusion by reanalyzing data from 13 studies that had shown significant, and often large, sex differences. Reis and Carothers also collected their own data on a range of psychological indicators. They revisited surveys on relationship interdependence, intimacy, and sexuality.

They reopened studies of the “big five” personality traits: extroversion, openness, agreeableness, emotional stability, and conscientiousness. They even crunched the numbers on such highly charged and seemingly defining gender characteristics as femininity and masculinity. Using three separate statistical procedures, the authors searched for evidence of attributes that could reliably categorize a person as male or female.

The pickings, it turned out, were slim. Statistically, men and women definitely fall into distinct groups, or taxons, based on anthropometric measurements such as height, shoulder breadth, arm circumference, and waist-to-hip ratio. And gender can be a reliable predictor for interest in very stereotypic activities, such as scrapbooking and cosmetics (women) and boxing and watching pornography (men).

But for the vast majority of psychological traits, including the fear of success, mate selection criteria, and empathy, men and women are definitely from the same planet. Instead of scores clustering at either end of the spectrum—the way they do with, say, height or physical strength—psychological indicators fall along a linear gradation for both genders.

No more ‘this or that’

With very few exceptions, variability within each sex and overlap between the sexes is so extensive that the authors conclude it would be inaccurate to use personality types, attitudes, and psychological indicators as a vehicle for sorting men and women.

“Thus, contrary to the assertions of pop psychology titles like Men Are From Mars, Women Are From Venus, it is untrue that men and women think about their relationships in qualitatively different ways,” the authors write. “Even leading researchers in gender and stereotyping can fall into the same trap.”

That men and women approach their social world similarly does not imply that there are no differences in average scores between the sexes. Average differences do exist, write the authors. “The traditional and easiest way to think of gender differences is in terms of a mean difference,” Carothers and Reis write. But such differences “are not consistent or big enough to accurately diagnose group membership” and should not be misconstrued as evidence for consistent and inflexible gender categories, they conclude.

“Those who score in a stereotypic way on one measure do not necessarily do so on another,” the authors note. A man who ranks high on aggression, may also rank low on math, for example. Caution the authors: “the possession of traits associated with gender is not as simple as ‘this or that’.”

Harmful for couples?

Although emphasizing inherent differences between the sexes certainly strikes a chord with many couples, such simplistic frameworks can be harmful in the context of relationships, says Reis.

“When something goes wrong between partners, people often blame the other partner’s gender immediately. Having gender stereotypes hinders people from looking at their partner as an individual. They may also discourage people from pursuing certain kinds of goals. When psychological and intellectual tendencies are seen as defining characteristics, they are more likely to be assumed to be innate and immutable. Why bother to try to change?”

The best evidence we have that the so-called Mars/Venus gender division is not the true source of friction within relationships, says Reis, is that “gay and lesbian couples have much the same problems relating to each other that heterosexual couples do. Clearly, it’s not so much sex, but human character that causes difficulties.”

The findings support the “gender similarities hypothesis” put forth by University of Wisconsin psychologist Janet Hyde. Using different methods, Hyde has challenged “overinflated claims of gender differences” with meta-analyses of psychology studies, demonstrating that males and females are similar on most, though not all, psychological variables.

Biology and behavior

Those results were not a surprise for Carothers. Raised by two physical education teachers, the self-described tomboy grew up with “all kinds of sporting equipment… I did not question stereotypical attitudes, I just knew that they did not necessarily fit me and the folks I hung out with.” That experience, she says, fueled a lifelong interest into the biological basis of behavior. When she discovered in graduate school that she could apply her prowess in statistics to exploring sex differences, the project became “a marriage of two interests.”

The authors acknowledge that the study is based largely on questionnaires and may not fully capture real life actions. “Methods that more pointedly measure interpersonal behaviors (how many birthday cards have they sent this year, how many times a month do they call a friend just to see how he or she is, etc.) may more readily reveal a gender taxon,” they write.

By the same token, however, as gender roles are liberalized, the authors speculate that new studies may show even less divergence between men and women in the United States. The opposite may be the case in cultures that are far more prescriptive of male and female roles, such as Saudi Arabia, Reis and Carothers predict.

Source: University of Rochester

chat8 Comments


  1. Joe

    I am not sure I’d agree with his interpretations. Also, there are differences which I don’t think his study looked at. It’s good to know there are many similarities between men and women, but I think there are many differences, too, which this study apparently didn’t look at.

  2. required

    Design your own study already. Don’t forget to show variation in your results, along with those that perhaps prove your theory, substantiated by your hypothesis, which will of course have its own operational definition and accountable statistics. And feel free while doing so.

  3. Anthony

    Good to see clarity applied once again to a place where over generalization is rife. Even when clear differences can be extracted such as in physical capability measurements, I think it is a sobering fact to consider the following;

    Whilst the fastest person (running) on earth may consistently be a man, the fact remains the fastest woman on earth runs faster that more than 99% of all people (including men). In effect this highlights that often individual differences are far greater than differences found in the extremes or within population.

    When ever we come across an individual we must accept that such generalisations have little or no value, after all we never meet a statistically significant cohort in person.

    I ask our last two “comment’ors” how they have attempted to avoid these pitfalls when making there own generalisations ?

  4. sally

    Brilliant. Not surprised at all.
    I have been arguing for some time against the sex-typing of the characteristics themselves.
    They are just characteristics and can all be found in everyone to varying degrees.
    It’s also hard to completely delineate what a person is born with, and what is developed socially.

  5. Simon

    Great to see some an evidence challenging the unhelpful dichotomy between genders. I have long thought it unfortunate that the Mars/Venus idea, which many people embrace, subsumes the much more interesting aspects of human relationships. While no doubt mental categories are useful shortcuts to understanding, they do us a disservice in this context. Many people simply do not understand the concept of a statistical average. For example, if on average, women were slightly more empathetic than men, on average, this is rather useless information when dealing with this individual man or that individual woman. The information we need is about this individual. Reading Mars vs Venus will not help this. Congratulations to the researchers for an interesting study. I hope it is expanded and any methodological limitations addressed, as this is important work.

  6. Jill

    Very refreshing findings. We had the decade of ‘gender behaviour caused by socialisation’, followed by the decade of ‘brain studies show significant gender differences’. Such an interesting topic

  7. Vince

    Did the study look at the variety of traits that each sex considered in it’s mate selection strategy and the position on a spectrum each trait occupied as per sex? If there’s any area in which there is the most divergence between the sexes, I believe, it is specifically here. What about the influence of hormones that vary significantly between the sexes such as testosterone on specific traits or predispositions? By and large, putting these together, it is difficult to conceive of empirical data that supports a minor divergence of the sexes

  8. Vince

    Actually, on looking at it a second time, it appears it did look at mate selection criterion. This indeed I find hard to believe. Moreover, having lived in the United States for about 4 years but having grown up in a more traditional culture, I think my observations largely do validate the claim that “the opposite may be the case in cultures that are far more prescriptive of male and female roles”. The reason for the difference however, I don’t believe to arise primarily from different cultures so “prescribing roles”. My experience in the US confirms what I suspect to be the case; that the general zeitgeist especially amongst women that consider themselves “liberated from traditional roles” mirrors the exact problem such studies would face in traditional cultures; only such problems are now reversed. Couple this with the fact that questionnaires were used largely in the experimentation; in other words, the data was based on what respondents thought to be true about themselves. It’s not a stretch to see why what respondents think to be true about themselves will be a function of what they perceive as what they ought to think about themselves which as I have maintained would be influenced largely by the dominant ideology, even counter-ideology of the times.

We respect your privacy.