“Depressing Study Finds Gender Stereotypes Haven’t Changed Since the 1980s,” proclaimed the New York magazine website the other day. The women’s site Bustle echoed the gloomy view: “Gender Stereotypes Just As Prevalent in 2016 As In The 1980s, New Study Finds, So Maybe Things Aren’t As Great As We’d Like To Believe.”

Yet a closer look at the study in question shows a far more complicated picture. While some beliefs about male and female traits and roles have indeed changed little since a similar survey in 1983, there has been a marked shift toward egalitarian attitudes on some important issues. There also seems to have been a marked shift toward more negative perceptions of men — which is arguably depressing, but probably not in the way the study’s authors and most of the commentators would like you to think.

The study, published in the Psychology of Women Quarterly, was carried out by psychologist Elizabeth J. Haines of New Jersey’s William Paterson University and her colleagues. Subjects, recruited online, were asked to rate on a 0-to-100 scale the likelihood of various traits, behaviors, and occupations for a hypothetical individual described in different versions of the survey as a typical man, a typical woman, or a typical person. The design closely matched a 1983 study involving college students; while the current sample had a much broader age range, the authors say the responses did not vary significantly by age.

What has changed least, as evident from the full dataset, is expectations about jobs and domestic tasks. The likelihood of being an elementary school teacher was rated 64 for a woman and 36 for a man, nearly the same as in 1983; the likelihood of being a firefighter, 61 for a man and 18 for a woman. In both years, women were estimated to be about twice as likely as men to cook, shop for groceries and do the laundry, and about 70 percent more likely to care for children. (Interestingly, all domestic activities were rated as slightly less likely for both men and women in 2014 than in 1983.)

But is that “stereotyping” or observation of reality? While the Women’s Health website caustically summed up the study as reflecting the belief that men should be responsible for “bringing home the bacon and fixing the car” and women for “raising the kids, cooking, and cleaning,” the respondents were not expressing prescriptive opinions but estimating actual probabilities. If anything, their ratings underestimated gender disparities in many jobs: in reality, a woman is four times more likely than a man to be an elementary school teacher, while the ratio of male to female firefighters is a staggering 24:1.

Where things get interesting is in ratings on perceived personality traits and related roles. People still see men as somewhat more independent, but the gap between the scores for men and women on that trait has shrunk from nearly 20 points in 1983 to 10 points today. This is due to a small rise in median female scores (from 58 to 60, which may well be a blip) and a seven-point drop in male ones. Men are still given a slight edge on “makes decisions easily” (61 points versus 57 for women), but that’s down from a 14-point gap in 1983; again, the change is due to a sharp drop in male scores and a small increase in female ones. And on “stands up under pressure,” men and women in the new survey got virtually equal ratings—whereas in 1983, men had a 16-point advantage.

On the likelihood of being “a leader,” men are still ahead by eight points, but that’s down from 17 points in 1983. And for “makes major decisions,” the gap has shrunk from 17 points to five, nearly obliterated by individual variation.

Contrary to the “nothing has changed” tenor of many reports, the new survey shows a dramatic decline in the male-provider norm. In the 1983 survey, the likelihood of being a “financial provider” was scored at 83 for a man, 46 for a woman; the 2014 numbers were 72 and 58. What’s more, men and women in 2014 were seen as equally likely to “assume financial obligations,” while in 1983 the score for men on this item was 22 points higher.

How a Rebellious Scientist Uncovered the Surprising Truth About Stereotypes by Claire Lehmann

Sydney. London. Toronto.

QuilletteQuillette

And here’s another fascinating nugget. In 1983, men and women were rated equally likely to “plan for the future,” with median scores around 75 for both. In 2014, the score for women remained unchanged — but the score for men dropped below 62. Bizarrely, the New York report specifically cited this item as one of the categories in which “gender stereotypes stayed consistent when compared to the 1983 study.”

Meanwhile, Bustle noted with dismay that “there seemed to be a small increase in stereotyping women’s gender roles.” This refers to the fact that the gaps between men’s and women’s ratings on some traits coded as “feminine” was somewhat larger in the 2014 study than in the earlier one. But is this really an increase in the stereotyping of women, or is there something else going on?

It is noteworthy that all the “feminine” traits in question —“warm,” “understanding,” “aware of the feelings of others,” “a source of emotional support”— are ones generally considered desirable in both sexes. In fact, in the 1983 survey, they were rated as highly likely in men, though more so in women: Thus, the likelihood of a man being “a source of emotional support” was rated 66, compared to 70 for a woman. In 2014, the rating for women was up to 77, while the rating for men had dropped to just over 50. On “aware of the feelings of others,” the median score for women barely changed while the score for men plummeted from 66 to 46.

Interestingly, both men and women were seen as less likely to be kind, warm, understanding, and helpful in 2014 than in 1983; but on all these items except “helpful,” the drop was far greater for men. The one “feminine” item on which ratings have risen for both sexes is “able to devote itself to others”; but it was a 5-point increase for men and nearly a 10-point boost for women.

While these findings should be treated with caution, given that each survey had a sample of fewer than 200, the trend toward a more jaundiced view of males seems unmistakable —particularly among women. (According to the authors, the higher attribution of “feminine” traits to women was particularly marked among female respondents.) For the most part, compared to 1983, men in 2014 were scored lower on both “feminine” positive traits and on “masculine” ones such as leadership, independence, self-confidence, decision-making, and ability to handle pressure.

There are, of course, several ways to look at this trend. Many feminists would probably argue that women are rightly holding men to higher standards. Some conservatives would probably argue that as a result of feminism and other societal changes, men have become less masculine and more self-centered. But let’s be honest: if we were talking about a shift toward less positive perceptions of any other group, the overwhelming response would be concern about growing social bias.

The authors of the study are concerned that lingering or even increasing stereotypes may undercut powerful women in fields such as politics. Citing U.S. Democratic presidential candidate Hillary Clinton as an example, Haines told Reuters, “When women act with power, we have a strong backlash reaction to them because they violate our expectation of what a woman should be like—nice, kind and accommodating.” (Oddly, as proof, Reuters cited a poll of likely Democratic voters in Florida showing that women are much more likely than men to prefer Clinton.)

But is there evidence that this is how stereotypes always work — especially when, as in the new survey, the perception of women as more likely to have positive “feminine” traits does not seem to preclude the attribution to them of positive “masculine” qualities such as leadership? Could stereotyping sometimes cause powerful women to be seen as kinder and more altruistic than powerful men? Recent research, such as the work of political scientists Deborah Jordan Brooks, Jennifer Lawless and Danny Hayes, suggests that today gender is more an asset than an obstacle for female politicians.

Yes, it’s likely that women who are perceived as too hard and cold are sometimes penalized because of societal expectations of female “niceness.” But surely, there are also times when the tendency to stereotype men as less understanding, warm, and capable of providing emotional support can result in unfairness to men. And some of that stereotyping is likely due not to patriarchy or lack of feminist progress, but do the direction feminism has taken in the last thirty years.