The Gender Gap Is Taking Us to Unexpected Places

In one of the most revealing studies in recent years, a 2016 survey of 137,456 full-time, first-year students at 184 colleges and universities in the United States, the U.C.L.A. Higher Education Research Institute found “the largest-ever gender gap in terms of political leanings: 41.1 percent of women, an all-time high, identified themselves as liberal or far left, compared to 28.9 percent of men.”

The institute has conducted freshmen surveys every year since 1966. In the early days, until 1980, men were consistently more liberal than women. In the early and mid-1980s, the share of liberals among male and female students was roughly equal, but since 1987, women have been more liberal than men in the first year of college.

While liberal and left identification among female students reached a high in 2016, male students remained far below their 1971 high, which was 44 percent.

Along parallel lines, a Knight Foundation survey in 2017 of 3,014 college students asked: “If you had to choose, which do you think is more important, a diverse and inclusive society or protecting free speech rights.”

Male students preferred protecting free speech over an inclusive and diverse society by a decisive 61 to 39. Female students took the opposite position, favoring an inclusive, diverse society over free speech by 64 to 35.

Majorities of both male and female college students in the Knight survey support the view that the First Amendment should not be used to protect hate speech, but the men were more equivocal, at 56 to 43, than women, at 71 to 29.

The data on college students reflects trends in the electorate at large. The Pew Research Center provided The Times with survey data showing that among all voters, Democrats are 56 percent female and 42 percent male, while Republicans are 52 percent male and 48 percent female, for a combined gender gap of 18 points. Pew found identical gender splits among voters who identify as liberal and those who identify as conservative.

“Significant gender differences in party identification have been evident since the early 1980s,” according to the Rutgers Center for American Women and Politics, which provides data on the partisanship of men and women from 1952 to the present day.

It’s clear from all this that the political engagement of women is having a major impact on the social order, often in ways that are not fully understood.

Take the argument made in the 2018 paper “The Suffragist Peace” by Joslyn N. Barnhart of the University of California-Santa Barbara, Allan Dafoe at the Center for the Governance of AI, Elizabeth N. Saunders of Georgetown and Robert F. Trager of U.C.L.A.:

Preferences for conflict and cooperation are systematically different for men and women. At each stage of the escalatory ladder, women prefer more peaceful options. They are less apt to approve of the use of force and the striking of hard bargains internationally, and more apt to approve of substantial concessions to preserve peace. They impose higher audience costs because they are more approving of leaders who simply remain out of conflicts, but they are also more willing to see their leaders back down than engage in wars.

The increasing incorporation of women into “political decision-making over the last century,” Barnhart and her co-authors write, raises “the question of whether these changes have had effects on the conflict behavior of nations.”

Their answer: “We find that the evidence is consistent with the view that the increasing enfranchisement of women, not merely the rise of democracy itself, is the cause of the democratic peace.”

Put another way, “the divergent preferences of the sexes translate into a pacifying effect when women’s influence on national politics grows” and “suffrage plays a direct and important role in generating more peaceful interstate relations by altering the political calculus of democratic leaders.”

Barnhart added by email:

The important thing to remember here is that with any trait, we are talking about averages and distributions and not categorical distinctions. Some men will have lesser preference for the use of force than some women and vice versa. The distribution of traits among the two genders overlaps. So we shouldn’t expect perfect partisan distinction.

Other consequential shifts emerged as women’s views began to change and they became more involved in politics.

Dennis Chong, a political scientist at the University of Southern California, wrote by email that “a gender gap in political tolerance, with women being somewhat more willing to censor controversial and potentially harmful ideas, goes back to the earliest survey research on the subject in the 1950s.”

There are a number of possible explanations, Chong said, including “stronger religious and moral attitudes among women; lesser political involvement resulting in weaker support for democratic norms; social psychological factors such as intolerance of ambiguity and uncertainty which translate to intolerance for political and social nonconformity; and greater susceptibility to feelings of threats posed by unconventional ideas and groups.”

Studies using moral foundations theory, Chong continued, have

found broad value differences between men and women. Women score higher on values defined by care, fairness, benevolence, and protecting the welfare of others, reflecting greater empathy and preference for cooperative social relations. In today’s debates over free speech and cancel culture, these social psychological and value differences between men and women are in line with surveys showing that women are more likely than men to regard hate speech as a form of violence rather than expression, to support laws against divisive hate speech, and to be skeptical that the right to free speech protects the disadvantaged more than the majority.

In addition, Chong said, “Women are also more likely than men to believe that colleges ought to protect students from exposure to controversial speakers whose ideas may create an inhospitable learning environment.”

Steven Pinker, a professor of psychology at Harvard, writes in his book “The Better Angels of Our Nature,” that “the most fundamental empirical generalization about violence” is that

it is mainly committed by men. From the time they are boys, males play more violently than females, fantasize more about violence, consume more violent entertainment, commit the lion’s share of violent crimes, take more delight in punishment and revenge, take more foolish risks in aggressive attacks, vote for more warlike policies and leaders, and plan and carry out almost all the wars and genocides.

Pinker continues:

Feminization need not consist of women literally wielding more power in decisions on whether to go to war. It can also consist in a society moving away from a culture of manly honor, with its approval of violent retaliation for insults, toughening of boys through physical punishment, and veneration of martial glory.

In an email, Pinker wrote:

We’re seeing two sets of forces that can pull in opposite directions. One set comprises the common interests of men on the one hand and women on the other. Men tend to be more obsessed with status and dominance and are more willing to take risks to compete for them; women are more likely to prize health and safety and to reduce conflict. The ultimate (evolutionary) explanation is that for much of human prehistory and history successful men and coalitions of men potentially could multiply their mates and offspring, who had some chance of surviving even if they were killed, whereas women’s lifetime reproduction was always capped by the required investment in pregnancy and nursing, and motherless children did not survive.

“Mapping the Moral Domain,” a 2011 paper by Jesse Graham, a professor of management at the University of Utah, and five colleagues, found key differences between the values of men and women, especially in the case of the emphasis women place on preventing harm, especially harm to the marginalized and those least equipped to protect themselves.

I asked Jonathan Haidt, a social psychologist at N.Y.U.’s Stern School of Business, about the changing political role of women. He emailed back:

In general, when looking at sex differences in outcomes, it is helpful to remember that differences between men and women on values and cognitive abilities are generally small, while differences between men and women in the activities that interest them, and in their relational styles (especially involving conflict) are often large.

When the academic world opened up to women in the 1970s and 1980s, Haidt continued, “women flooded into some areas but showed less interest in others. In my experience, having entered in the 1990s, the academic culture of predominantly female fields is very different from those that are predominantly male.”

Haidt noted:

Boys and men enjoy direct status competition and confrontation, so the central drama of male-culture disciplines is ‘“Hey, Jones says his theory is better than Smith’s; let’s all gather around and watch them fight it out, in a colloquium or in dueling journal articles.” In fact, I’d say that many of the norms and institutions of the Anglo-American university were originally designed to harness male status-seeking and turn it into scholarly progress.

Women are just as competitive as men, Haidt wrote, “but they do it differently.”

He cited a 2013 paper, “The development of human female competition: allies and adversaries,” by Joyce Benenson, of Harvard’s department of human evolutionary biology. In it, Benenson writes:

From early childhood onwards, girls compete using strategies that minimize the risk of retaliation and reduce the strength of other girls. Girls’ competitive strategies include avoiding direct interference with another girl’s goals, disguising competition, competing overtly only from a position of high status in the community, enforcing equality within the female community and socially excluding other girls.

In summary, Benenson wrote:

From early childhood through old age, human females’ reproductive success depends on provisioning, protecting and nurturing first younger siblings, then their own children and grandchildren. To safeguard their health over a lifetime, girls use competitive strategies that reduce the probability of physical retaliation, including avoiding direct interference with another girl’s goals and disguising their striving for physical resources, alliances and status.

In a November 2021 paper, “Self-Protection as an Adaptive Female Strategy,” Benenson, Christine E. Webb and Richard W. Wrangham, all of the department of human evolutionary biology, report that they

found consistent support for females’ responding with greater self-protectiveness than males. Females mount stronger immune responses to many pathogens; experience a lower threshold to detect, and lesser tolerance of, pain; awaken more frequently at night; express greater concern about physically dangerous stimuli; exert more effort to avoid social conflicts; exhibit a personality style more focused on life’s dangers; react to threats with greater fear, disgust and sadness; and develop more threat-based clinical conditions than males.

These differences manifest in a number of behaviors and characteristics, Benenson, Webb and Wrangham argue:

We found that females exhibited stronger self-protective reactions than males to important biological and social threats; a personality style more geared to threats; stronger emotional responses to threat; and more threat-related clinical conditions suggestive of heightened self-protectiveness. That females expressed more effective mechanisms for self-protection is consistent with females’ lower mortality and greater investment in child care compared with males.” In addition, “females more than males exhibit a lower threshold for detecting many sensory stimuli; remain closer to home; overestimate the speed of incoming stimuli; discuss threats and vulnerabilities more frequently; find punishment more aversive; demonstrate higher effortful control and experience deeper empathy; express greater concern over friends’ and romantic partners’ loyalty; and seek more frequent help.

In an email, Benenson added another dimension to the discussion of sex roles in organizational politics:

From an early age, women clearly dislike group hierarchies of same-sex individuals more than men do. Thus, while boys and men are more willing to compete directly with both higher and lower status individuals, girls and women prefer to interact with same-sex individuals of similar status. This does not mean however that girls and women don’t care about status as much as boys and men do. For both sexes, high status increases the probability that one lives longer and so do one’s children. The result of these two somewhat conflicting motives is that girls and women seek high status but disguise this quest by avoiding direct contests. This gender difference likely impacts how women seek to shape organizational culture.

The strategies Benenson and her colleagues describe, Haidt pointed out,

lead to a different kind of conflict. There is a greater emphasis on what someone said which hurt someone else, even if unintentionally. There is a greater tendency to respond to an offense by mobilizing social resources to ostracize the alleged offender.

In “Feminist and Anti-Feminist Identification in the 21st Century United States,” Laurel Elder, Steven Greene and Mary-Kate Lizotte, political scientists at Hartwick College, North Carolina State University and Augusta University, analyzed the responses of those who identified themselves as feminists or anti-feminists in 1992 and 2016.

Based on surveys conducted by American National Election Studies, Elder, Greene and Lizotte found that the total number of voters saying that they were feminists grew from 28 percent to 34 percent over that period. The growth was larger among women, 29 percent to 50 percent, than among men, 18 percent to 25 percent.

Some of the biggest gains were among the young, 18-to-24-year-olds, doubling from 21 percent to 42 percent. Most striking is the data revealing the antithetical trends between women with college degrees, whose self-identification as feminist rose from 34 percent to 61 percent, in contrast to men with college degrees, whose self-identification as feminist fell from 37 percent to 35 percent.

Anti-feminist identity, the authors found,

is not just a mirror image of feminist identity but its own distinctive social identity. A striking difference between feminist and anti-feminist identification is that while gender is a huge driver in feminist identification in 2016, there is essentially no gender gap among anti-feminists. Indeed, bivariate analysis shows that 16 percent of women and 17 percent of men identify as anti-feminists.

In addition, Elder, Greene and Lizotte wrote, “while young people were more likely to identify as feminists than older generations in 2016, young people, particularly young women, also have a higher level of anti-feminist identification compared to older groups.”

The other patterns of anti-feminist identification, according to the authors, are “more the mirror image of feminist identification” with “Republicans being more likely to identify as anti-feminists compared to Democrats, and stay-at-home parents/homemakers, those who identify as born again, and those who attend church frequently being more anti-feminist.”

To provoke further discussion, I will end with the argumentative economist Tyler Cowen, of George Mason University and “Marginal Revolution.” In December 2019, Cowen wrote a column for Bloomberg, “Women Dominated the Decade,” subtitled “The 2010s were pretty thrilling if you liked music, books, TV or movies by or about women.”

Cowen, who acknowledges describing “feminization in not entirely glowing terms” — indeed one would have to say hostile terms — is also, in other contexts, unequivocally enthusiastic about “what I see as the No. 1. trend of the decade: the increasing influence of women.”

“I had the best of both worlds,” Cowen writes, “namely to grow up in the ‘tougher’ society, but live most of my life in the more feminized society.”