by Kenneth W. Krause.
Kenneth W. Krause is a contributing editor and “Science Watch” columnist for the Skeptical Inquirer. Formerly a contributing editor and books columnist for the Humanist, Kenneth contributes regularly to Skeptic as well. He may be contacted at firstname.lastname@example.org.
Why are people so afraid of the idea that the minds of men and women are not identical in every way?—Steven Pinker, 2002.
The mere suggestion that one group of people is cognitively or emotionally distinct from another can leave many of us speechless and squirming in our seats. The effect is intensified, of course, in the regrettable event of historical discrimination, and especially so when the differences are alleged to be innate.
Scientists of many stripes have bravely confronted, struggled with, and evidently resolved the issue as it pertains to “race.” Such classifications lack sturdy biological bases, the current consensus holds, and their very existence relies on nothing more concrete or dependable than cultural convention and political expediency (1).
Gender or sex (I use the terms interchangeably here) is similar in some respects, but clearly distinct in others. Some biological differences between men and women are both unmistakable and abundantly appreciated. Combat can erupt, however—perhaps most furiously in intellectual circles—over questions involving mental differences and, assuming their existence, over their proposed causes.
In a recent column, I investigated the origins of female “underrepresentation” in high-end STEM fields. The latest analyses had suggested that, rather than being discriminated against, qualified women tended to choose people-centered over thing-centered professions. That is, the somewhat narrow mental trait examined was interest.
Other studies have explored broader gender differences in personality—a related and at least equally sensitive domain. In a highly influential 2005 paper, for example, University of Wisconsin-Madison professor of psychology and women’s studies, Janet Hyde, rebuked the popular media and general public for their apparent fascination with an assumed profusion of deep psychological variances between genders (Hyde 2005).
After reviewing 46 meta-analyses on the subject, Hyde proposed a new model. The gender similarities hypothesis (GSH) holds that “males and females are similar on most, but not all, psychological variables.” Because most differences are negligible or small, and because very few are large, Hyde contended, “men and women as well as boys and girls are more alike than they are different.” Physical aggression and sexuality were offered as exceptions.
But in a new study, “The Distance Between Mars and Venus,” Hyde’s renowned hypothesis was directly and expressly challenged by a trio of Europeans led by Marco Del Giudici, evolutionary psychologist at the University of Turin, Italy (Del Giudici 2012). Having subjected a sample of 10,261 American men and women between ages 15 and 92 to an assessment of multiple personality variables, Del Giudici obtained results he and his team described as “striking.”
The “true extent of sex differences in human personality,” he argued, “has been consistently underestimated.” Del Giudici now compares personality disparities to those of other psychological constructs like vocational interests and aggression. When properly measured, he reports, gender personality differences are “large” and “robust.” Indeed, roughly 82 percent of his cohort delivered personality profiles that could not be matched with any member of the opposite sex.
So by what method should researchers measure these distinctions? The Europeans broke new ground by combining three techniques. First, to enhance reliability and repeatability, they estimated differences based on latent factors rather than observed scores. Second, instead of employing the so-called “Big Five” variables (extraversion, agreeableness, conscientiousness, neuroticism, and openness to experience), Del Giudici and company applied 15 narrower traits in order to assess personality with “higher resolution.” Finally, they chose multivariate over univariate effect sizes—thus aggregating rather than averaging variances—to more accurately reveal “global” sex differences.
Hyde swiftly posted her response. Roundly disparaging Del Giudici’s statistical method, she charged that it “introduces bias by maximizing differences.” In the end, she continued, the Europeans’ “global” result is merely a single and “uninterpretible” dimension that only “blur[s] the question rather than offering higher resolution.” The GSH stands intact, she insists. The true expanse between genders, Hyde argued, is anything but astronomical: Instead, it more resembles “the distance between North Dakota and South Dakota.”
Either way, a third researcher teased, “you’ll still have a mighty long way to walk.” Richard Lippa, professor of psychology at California State University, Fullerton, proposed an attractive analogy in the Italian’s defense. Consider sex differences in body shape, he suggested. The approach underlying Hyde’s GSH would average certain trait ratios—shoulder-to-waist, waist-to-hip, torso-to-leg length, for example—and likely declare that men and women have similar bodies. By contrast, Del Giudici’s multivariate method would probably generate the much more intuitive conclusion that “sex differences in human body shape are quite large, with men and women having distinct multivariate distributions that overlap very little.” The Italian offered a similar and equally effective analogy comparing male and female faces.
Del Giudici finds Hyde’s “single dimension” criticism ironic indeed because his method’s essential point, he says, was to integrate multiple personality factors rather than isolate them. Most dramatically in univariate terms, those traits included sensitivity, warmth, and apprehension (higher in women), and emotional stability, dominance, vigilance, and rule-consciousness (higher in men).
Nor does he see an interpretability problem. The Italian’s “weighted blend” of 15 personality traits, he argues, provides a concrete and meaningful description of global differences informing us of a 10 to 20 percent overlap between male and female distributions. He denies as well Hyde’s claim that his techniques were either controversial or prone to maximizing bias. To the contrary, he told me, the Europeans simply “thought hard about the various artifacts that can deflate sex differences in personality, and took steps to correct them.”
Pinker’s provocative query denouncing our fear of sex differences was largely rhetorical, of course. He answered the question soon after asking it: “The fear,” he acknowledged, “is that different implies unequal.” If we momentarily assume that gender personality differences are substantial, the next issue to confront might be whether those differences are driven more by culture or biology. In either case, certain groups may be forced to rethink some much-cherished ideas and practices.
Lippa recently probed the ultimate “nature vs. nurture” question in a review of two meta-analyses and three cross-cultural studies on gender differences in both personality and interests (Lippa 2010). In the end, he discovered that women tend to score significantly higher over time and across cultures in the Big Five categories of agreeableness and neuroticism, and, as others have found since, that they gravitate more toward people-centered than thing-centered occupations.
The Californian then described two basic sets of non-exclusive theories under which such evidence is typically evaluated. Biological theories, of course, focus on genes, hormones, neural development, and brain structure, for example. These models are inspired by our knowledge of evolution. Social-environmental theories, by contrast, concentrate on stereotypes, self-conceptualization, and social learning. Here, cultural influences are thought to dominate.
Supporters of distinct sub-theories would no doubt evaluate the evidence in varying ways. But significant gender differences that are consistent across cultures and over time, Lippa contends, are more likely to reflect underlying biological rather than social-environmental causes. Similarly suggestive, the author says, is the fact that such divergences tend to be greater in relatively ‘modern,’ individualistic, and gender-egalitarian societies.
In his new paper, Del Giudici chose not to directly engage the difficult question of underlying causes. Nonetheless, he reminds us that evolutionary principles—sexual selection and parental investment theories, in particular—provide us with ample grounds to “expect robust and wide-ranging sex differences in this area.” “Most personality traits” he continues, “have substantial effects on mating- and parenting-related behaviors.”
Even so, Hyde answers, more than one evolutionary force may be at play here. Although sexual selection can produce sex differences, she admits, other forms of natural selection can render sex similarities. “The evolutionary psychologists,” she reckons, “have forgotten about natural selection.”
On these limited questions, truly common ground seems scarce indeed. Why should the authors interpret the evidence so differently? Of course no member of any group or human institution is impervious to personal or philosophical biases. One might reasonably expect academics to be more objective than others, but—for what it’s worth—that has seldom been my experience as a science writer.
In his review, Lippa argued generally that “[c]ontemporary gender researchers, particularly those who adopt social constructionist and feminist ideologies, often reject the notion that biologic factors directly cause gender differences.” And more pertinently here, he claims that Hyde has long “ignored ‘big’ differences in men’s and women’s interests,” and that the GSH “is, in part, motivated by feminist ideologies and ‘political’ attitudes.”
Hyde denies the accusation categorically: “The GSH is not based on ideology,” she told me. “It is a summary of what the data show … data from millions of subjects.” One might note of the Wisconsinite’s pioneering paper, however, that a great deal of concluding space was consumed decrying the perceived social costs of gender difference claims—especially to women, rather than further illuminating or summarizing the data.
Del Giudici appears to find the issue of bias somewhat less motivating. If sex differences are small, he suggests, we have little to explain and more time to discuss incorrect stereotypes—“this is the main appeal of the GSH.” The author agrees that “ideology has played a part in the success of the GSH.” Nonetheless, he maintains that the aforementioned “methodological limitations have played a larger role.”
In his closing comments to me, the Californian echoed much of what Steven Pinker has so courageously recognized in recent years with regard to the broader subject of group divergences. The ongoing examination of sex differences in personality may or may not be tainted by feminism or other ideologies. But given the inquiry’s great sensitivity and profound implications, Lippa’s comments—crafted in the finest tradition of true skepticism—bear repeating here:
“I believe this is not a topic where ‘ignorance is bliss.’ We have to examine the nature of sex differences objectively… We should, as researchers, be open to all possible explanations. And then, as a society, we have to decide whether we want to let the differences be whatever they may be, or work to reduce them.”
Words to inquire by. So let the research into gender differences continue, as the Europeans urge, unfettered by irrelevant politics or pet, self-serving causes. I suspect we have little to fear. But let science characterize our differences objectively, whatever their nature and degree. Then, if necessary, we’ll decide together—as an open and informed community—how best to cope with them.
(1) Two excellent books have recently reviewed the scientific and cultural particulars of “race” for a popular audience: Ian Tattersall and Rob De Salle. 2011. Race? Debunking a Scientific Myth. Texas A&M University Press, and Sheldon Krimsky and Kathleen Sloan, eds. 2011. Race and the Genetic Revolution: Science, Myth, and Culture. Columbia University Press.
Del Giudice, M., Booth, T., and Irwing, P. 2012. The distance between Mars and Venus: Measuring global sex differences in personality. PLoS ONE 7(1): e29265.
Hyde, J.S. 2005. The gender similarities hypothesis. American Psychologist 60: 581-592.
Lippa, R. A. 2010. Gender differences in personality and interests: When, where, and why? Pers. and Soc. Psych. Compass 4/11: 1098-1110.
Pinker, Steven. 2002. The Blank Slate: The Modern Denial of Human Nature. NY: Penguin Viking.