The Science of Human Sex Differences: Implications for Policy on Gender Equity
Gender equity is one of the most widely discussed and hotly contested political issues of our time. Of the many topics which fall under the gender equity banner, three have attracted the lion’s share of attention. First, men are overrepresented in a range of prestigious professions, including most STEM fields. Second, even in professions in which the sex ratio is relatively balanced, men tend to be overrepresented at the top; in business, for instance, male CEOs outnumber female CEOs. Third, averaging across professions and age groups, men earn more than women (although this is sometimes now reversed among young adults, in part because young women tend to be better educated). The most common explanations for these disparities include discrimination, gender socialization, and cultural norms, and the most common policy responses aim to remedy these supposed causes. However, a large and growing body of scientific research points to a very different explanation, namely that men and women naturally differ in a number of ways that affect their occupational choices and outcomes (Geary, 2010; Halpern, 2012; Levy & Kimura, 2009). In the following, I outline the evidence for this view, and consider some of the possible policy prescriptions that follow from it.
Sex Differences in Preferences and Talents
Psychologists have documented a range of psychological sex differences that could influence men and women’s career trajectories. First, the sexes differ somewhat in their interests and life priorities. On average, for instance, men are more interested than women in working with things and abstract rules, whereas women are more interested than men in working with people (Su, Rounds, & Armstrong, 2009). Similarly, men place more importance than women on a high income and career success, whereas women place more importance on work-life balance – again, not always but on average (Lubinski, Benbow, & Kell, 2014). In addition to sex differences in interests and priorities, men and women differ somewhat in their cognitive talents (Halpern, 2012). Women, for example, do better than men on tests of verbal reasoning and writing ability, whereas men do better on tests of visuospatial ability and mathematical reasoning (Wai, Cacchio, Putallaz, & Makel, 2010). To be clear, both sexes can be found at all levels of all of these traits, and the distributions for men and women overlap a great deal. However, for each of the traits, the average score – the centre of the distribution – is somewhat higher for one sex than the other.
Although these sex differences are relatively modest, they may nonetheless have a significant impact on the occupational landscape. This is because, for normally distributed data, even small differences at the centre of the distribution are associated with much larger differences at the extremes (Halpern, 2012; Pinker, 2002). Thus, although the gender gap in average language ability is small, women greatly outnumber men at the extremes of linguistic giftedness. Similarly, although the gender gap in average visuospatial ability is modest, men greatly outnumber women among the minority who are especially talented in this arena. Because employers seek the highest scorers on traits required for their professions, this pattern may result in sizeable gender gaps in the professions in question, even though the differences in the general population are not especially large. Thus, the fact that the average man scores somewhat higher than the average woman on tests of visuospatial ability may help to explain why many more men than women end up in physics and engineering. Likewise, the fact that the average woman places somewhat more importance on work-life balance may help to explain why fewer women than men are willing to work sixty or eighty hour weeks in order to rise to the top of their fields (although, of course, some women are willing and most men are not).
Where do sex differences like these come from? A common suggestion is that they’re products of socialization. But although socialization presumably has some impact, several lines of evidence suggest that there’s a substantial innate component as well. Consider, for instance, the maths and language gender gaps. Jonathan Wai and colleagues (2010) found that, since the early 1990s, boys have outnumbered girls four-to-one among the top .01% in mathematical reasoning, despite intensified efforts to eliminate the maths gap. Meanwhile, girls have consistently outnumbered boys at the highest levels of verbal reasoning and writing ability. Purely environmental explanations for these differences, most notably stereotype threat, have recently been cast into doubt (e.g., Flore & Wicherts, 2015).
Other evidence suggests an innate contribution to the sex differences in preferences and priorities. First, sex differences in the preference for working with things vs. people have remained fairly stable for as long as psychologists have measured them (Su et al., 2009), and are found in every nation where psychologist have looked (e.g., 53 out of 53 nations in one large-scale study; Lippa, 2010). Moreover, the differences appear early in life, perhaps even in newborns (Connellan, Baron-Cohen, Wheelwright, Batki, & Ahluwalia, 2000), and have been linked to prenatal hormonal exposure (Levy & Kimura, 2009). In light of this evidence, it is difficult to maintain that the differences could be wholly a product of cultural forces.
Sex Differences in Variability
Men and women differ in another, very different way as well – one that may also be relevant to their occupational outcomes. Specifically, in a wide range of traits, men are more variable than women (Lehre, Lehre, Laake, & Danbolt, 2009). To illustrate, consider height. We all know that men, on average, are taller than women. But men are also more variable in height – that is, the gap between the tallest and shortest man is greater than that between the tallest and shortest woman.
Importantly, men also appear to be more variable in a number of psychological traits relevant to people’s occupational choices and outcomes. First, men are more variable on tests of scientific achievement, mathematics, spatial ability, and mechanical reasoning (Hedges & Nowell, 1995). This means that, although most men and women fall in the middle of the distribution for these skills, there are somewhat more men than women at the very highest levels of ability and somewhat more men than women at the very bottom. Men also appear to be more variable in general cognitive ability or IQ, with more men at the top and more men at the bottom, once again (Johnson, Carothers, & Deary, 2008; although see Iliescu, Ilie, Ispas, Dobrean, and Clinciu, 2016, for a recent failure to replicate this common finding).
In principle, greater male variability could help to explain the preponderance of males at the top of the occupational ladder (Levy & Kimura, 2009; Pinker, 2002). Most people don’t have the cognitive talents required to be CEOs or top STEM researchers, and among the few who do, some are men and some are women. It is possible, however, that somewhat more are men than women. At the same time, the fact that there are more men than women at the bottom of the ability distribution may help to explain the fact that there are also more men than women at the bottom of the social heap: more homeless men, more men in prison, and so on. This side of men’s social situation is often overlooked in discussions of gender equity.
There are several reasons to think that sex differences in variability have an innate basis. First, it’s unclear what social or cultural influences could simultaneously shunt more men to the top and more men to the bottom of the distribution for traits such as mathematical ability or IQ. Second, sex differences in variability emerge early in childhood, before children even start preschool (Arden & Plomin, 2006). Third, the differences appear not only in psychological traits but in traits which are impervious to stereotypes and social pressure, such as height (Lehre et al., 2009). And fourth, sex differences in variability are found not only in humans but in a wide range of species. Male red deer, for example, are more variable than females in overall size, and male barn swallows are more variable in tail length. The most plausible and parsimonious explanation for these findings is that greater male variability is an offshoot of the male-typical developmental program, rather than an effect of culture that coincidentally mimics what we see in the rest of the animal kingdom.
Differences Don’t Necessarily Imply Discrimination
We’ve seen, then, that men and women differ in interests, talents, and levels of variability. These differences have a number of potential implications for policy makers. One concerns efforts to deal with the differential representation of the sexes across occupations. A common assumption is that when women are underrepresented in any desirable profession, or are underrepresented in a profession’s upper echelons, this must be due to discrimination of some kind. But the science of sex differences challenges the automatic inferential leap from differences to discrimination. Certainly, if men and women were identical in interests, talents, and variability, such an inference might be warranted. Given that they’re not identical, however, there’s no reason to expect that equality of opportunity will automatically translate into equality of outcome. On the contrary, the natural expectation would be that men and women’s outcomes will differ, as a result of freely made decisions about which paths through life would best suit their interests and talents. This isn’t to deny that discrimination might also contribute to any lingering disparities. But this can no longer be taken for granted.
To what extent, then, is discrimination implicated? Certainly, some studies suggest that it’s still an issue in traditionally male-dominated areas, such as STEM fields (e.g., Moss-Racusin, Dovidio, Brescoll, Graham, & Handelsman, 2012). It’s important to note, though, that not all the evidence points in this direction. Many women report experiencing no discrimination, but instead being strongly encouraged to rise through the ranks (Pinker, 2008). Furthermore, various studies find evidence of bias in favour of women in STEM, not against – perhaps unsurprising given the strong push for equal representation (Ceci & Williams, 2011; Williams & Ceci, 2015). Finally, the popular concept of implicit bias or unconscious discrimination has increasingly come under fire in recent years, with many psychologists raising questions about its validity and significance (summarized in Singal, 2017). The fact that the evidence is so mixed tells us, if nothing else, that discrimination couldn’t be a major cause of the disparities. If it were, it would presumably be easier to detect and thus the research would paint a much more consistent picture.
This gives us at least some reason to think that today’s occupational gaps are primarily a consequence of differences in men and women’s interests, talents and levels of variability. And if that’s the case, then efforts to close the gaps by combatting discrimination are presumably doomed to failure (as well as being rather unfair on those tacitly accused of discrimination). As psychologists Stephen Ceci and Wendy Williams (2011) concluded, on the basis of their research on this topic, “the ongoing focus on sex discrimination in reviewing, interviewing and hiring represents costly, misplaced effort. Society is engaged in the present in solving problems of the past” (p. 3157). Policies tackling the current sources of women’s underrepresentation may be more productive. This includes, for instance, efforts to make career trajectories in STEM fields more compatible with the demands of motherhood for women with children.
Differences Aren’t Necessarily a Problem
The science of sex differences raises another, more fundamental question as well, namely whether eliminating occupational disparities is a worthwhile goal in the first place. If the disparities were due to discrimination or disinformation, it clearly would be. To the extent, though, that they’re a product of people’s free and informed choices, the issue becomes a lot trickier. Gender parity is presumably not a good in itself, but a good only in as much as it increases aggregate happiness. But if occupational disparities are a result of men and women pursuing their own best interests, it’s unclear whether gender parity would contribute to the goal of raising net happiness.
To begin with, men and women could have different outcomes yet still be happy with their lives. Consistent with this suggestion, one landmark study showed that, among the academically gifted, men and women had somewhat different goals and took somewhat different paths, but ended up equally happy (Lubinski et al., 2014). Apparently, then, even among those best positioned to achieve their career goals, gender parity isn’t necessary for happiness.
Not only may it be unnecessary, but policies that artificially induce gender parity –incentives, gender quotas, and the like – may in fact lower aggregate happiness. To the extent that these policies work, they necessarily mean that some people will end up in work that doesn’t best suit their tastes or talents. As the educational psychologist Linda Gottfredson observed, “if you insist on using gender parity as your measure of social justice, it means you will have to keep many men and women out of the work they like best and push them into work they don’t like” (cited in Holden, 2000, p. 380). This is less likely to nurture aggregate wellbeing than it is to harm it.
Many see sex differences as inherently problematic and as evidence of rampant sexism. There’s reason to believe, however, that in some cases, gender disparities are a symptom not of a sexist society but of a healthy one. A large body of research indicates that, in nations with greater gender equity and wealth, sex differences are often larger, rather than smaller (Schmitt, 2015). The reason for this isn’t entirely clear, but one possibility is that when people are relatively free to follow their preferences and nurture their individuality, nascent differences between the sexes are magnified (Schmitt, Realo, Voracek, & Allik, 2008; Sommers, 2013). If this is right, then efforts to eradicate these differences, rather than furthering gender equity, may instead involve attacking a positive symptom of gender equity.
Which Way Forward?
In summary, various lines of evidence suggest that men and women differ naturally in preferences, talents, and levels of variability, and that this may be the primary cause of gender disparities in occupational choices and outcomes in modern Western nations. This has important implications regarding the appropriate political response to these disparities. The approach that would be most compatible with maximizing happiness may be to strive for equality of opportunity for the sexes in every profession, but then to respect men and women’s decisions regarding their own lives and careers, even if this doesn’t result in perfect gender parity. Policies aimed at ensuring equality of outcome – quotas, affirmative action, incentives for people to pursue careers they wouldn’t otherwise pursue – may exact a toll in terms of net happiness (if they override people’s preferences) and economic efficiency (if people are hired based on their demographic profile rather than pure merit). In other words, such policies may inadvertently place a statistical outcome (a 50:50 sex ratio) above individual and social wellbeing. This would be an unfortunate state of affairs, to say the least.
Arden, R., & Plomin, R. (2006). Sex differences in variance of intelligence across childhood. Personality and Individual Differences, 41, 39-48.
Ceci, S. J., & Williams, W. M. (2011). Understanding current causes of women’s underrepresentation in science. Proceedings of the National Academy of Sciences, 108, 3157-3162.
Connellan, J., Baron-Cohen, S., Wheelwright, S., Batki, A., & Ahluwalia, J. (2000). Sex differences in human neonatal social perception. Infant Behavior and Development, 23, 113-118.
Flore, P. C., & Wicherts, J. M. (2015). Does stereotype threat influence performance of girls in stereotyped domains? A meta-analysis. Journal of School Psychology, 53, 25-44.
Geary, D. (2010). Male, female: The evolution of human sex differences (2nd ed.). Washington, DC: American Psychological Association.
Halpern, D. F. (2012). Sex differences in cognitive abilities (4th ed.). New York, NY: Psychology Press.
Hedges, L. V., & Nowell, A. (1995). Sex differences in mental test scores, variability and numbers of high-scoring individuals. Science, 269, 41–45.
Holden, C. (2000). Parity as a goal sparks bitter battle. Science, 289, 380-380.
Iliescu, D., Ilie, A., Ispas, D., Dobrean, A., & Clinciu, A. I. (2016). Sex differences in intelligence: A multi-measure approach using nationally representative samples from Romania. Intelligence, 58, 54-61.
Johnson, W., Carothers, A., & Deary, I. J. (2008). Sex differences in variability in general intelligence: A new look at the old question. Perspectives on Psychological Science, 3, 518-531.
Lehre, A.-C., Lehre, K. P., Laake, P., & Danbolt, N. C. (2009). Greater intrasex phenotype variability in males than in females is a fundamental aspect of the gender differences in humans. Developmental Psychobiology, 51, 198-206.
Levy, J., & Kimura, D. (2009). Women, men, and the sciences. In C. H. Sommers (Ed.), The science on women and science (pp. 202-284). Washington, DC: AEI Press.
Lippa, R. A. (2010). Sex differences in personality traits and gender-related occupational preferences across 53 nations: Testing evolutionary and social-environmental theories. Archives of Sexual Behavior, 39, 619-636.
Lubinski, D., Benbow, C. P., & Kell, H. J. (2014). Life paths and accomplishments of mathematically precocious males and females four decades later. Psychological Science, 25, 2217-2232.
Moss-Racusin, C. A., Dovidio, J. F., Brescoll, V. L., Graham, M. J., & Handelsman, J. (2012). Science faculty’s subtle gender biases favor male students. Proceedings of the National Academy of Sciences, 109, 16474-16479.
Pinker, S. (2002). The blank slate: The modern denial of human nature. New York, NY: Viking.
Pinker, S. (2008). The sexual paradox: Troubled boys, gifted girls, and the real difference between the sexes. New York, NY: Scribner.
Schmitt, D. P. (2015). The evolution of culturally-variable sex differences: Men and women are not always different, but when they are… it appears not to result from patriarchy or sex role socialization. In V. A. Weekes-Shackelford & T. K. Shackelford (Eds.), The evolution of sexuality (pp. 221-256). New York, NY: Springer.
Schmitt, D. P., Realo, A., Voracek, M., & Allik, J. (2008). Why can’t a man be more like a woman? Sex differences in Big Five personality traits across 55 cultures. Journal of Personality and Social Psychology, 94, 168-182.
Singal, J. (2017). Psychology’s favorite tool for measuring racism isn’t up to the job. New York Magazine. Retrieved from http://nymag.com/scienceofus/2017/01/psychologys-racism-measuring-tool-isnt-up-to-the-job.html
Sommers, C. H. (2013). What ‘lean in’ misunderstands about gender differences. The Atlantic. Retrieved from https://www.theatlantic.com/sexes/archive/2013/03/what-lean-in-misunderstands-about-gender-differences/274138/
Su, R., Rounds, J., & Armstrong, P. I. (2009). Men and things, women and people: A meta-analysis of sex differences in interests. Psychological Bulletin, 135, 859-884.
Wai, J., Cacchio, M., Putallaz, M., & Makel, M. C. (2010). Sex differences in the right tail of cognitive abilities: A 30 year examination. Intelligence, 38, 412-423.
Williams, W. M., & Ceci, S. J. (2015). National hiring experiments reveal 2:1 faculty preference for women on STEM tenure track. Proceedings of the National Academy of Sciences, 112, 5360-5365.
 STEM stands for science, technology, engineering, and mathematics.
Monographic Gifted women, fragile men
School of Psychology, University of Nottingham Malaysia Campus