16 March 2014, 14:42

Doing math doesn't help girls to get better jobs - study

Doing math doesn't help girls to get better jobs - study

Women make up around a quarter of all so-called STEM workers in science, technology, engineering, and math. Even though more girls are getting college degrees in these fields, male science and engineering graduates are employed in STEM jobs at twice the rate of women. The hurdles women face in entering these fields crop up at a young age, NYT says.

They are socialized to think women don't go into science and math. Just 20 percent of the high school students taking the Advanced Placement exam in computer science last year were girls, and in two states, not a single female student took it. Women make up 53 percent of college graduates but just 41 percent of those from science and engineering programs. Part of the problem is clearly that they face deep-seated discrimination when trying to get hired in these fields. The economist Larry Summers famously suggested once that so few women become scientists and engineers because of discrimination, preference and even differences in innate ability. A new study confirms this theory.

Researchers from Columbia Business School, the Booth School of Business at the University of Chicago, and the Kellogg School of Management at Northwestern University set up a lab experiment in which "managers" hired people to complete mathematical tasks that, on average, men and women performed equally well.Then test subjects had to decide who to hire. "Our results reveal a strong bias among subjects to hire male candidates," the researchers note, which was true of both men and women. With no information about the job "applicants" other than their appearance, the managers (of both sexes) were twice as likely to hire a man over a woman. This was because women were expected to perform worse on the math problems, even though it was a task they were equally like to do well.

Things improved when those doing the hiring were given full information about how the candidates did on the task, but even then discrimination wasn't totally eliminated. Perhaps worse, the employers' perceptions were changed when the information about performance was given to them by an "objective" source, i.e. by the people running the experiment. But when the candidates provided it themselves, there was still a bias toward choosing men.

So much for women's lack of ability getting in the way — even when they perform equally well, they're less likely to be hired into a math position. The researchers also note that a third explanation for the dearth of women, that they opt out of these jobs and prefer other work, doesn't apply to their findings either. "[W]e designed an experiment in which supply-side considerations did not apply," they write, "and those possible differences in preference could not lead to differences in performance quality." All signs point to heavy bias against women getting in the way.

But another factor is that even when women do get hired, they are getting pushed out much faster than men. Women leave these jobs 45 percent more often in the early years than men. Some are leaving because the conflict between these high-demand jobs conflicts with the fact that women are still usually the primary caregivers in a family. Sixty-two percent of female STEM workers — those who stick it out — don't have kids, compared to 57 percent of men.

To top it all off, even women who study in the field, break into these jobs, and stay with it don't get the same rewards. Women who work full-time in a STEM job make $15,900 less each year than their male coworkers.

The professors, Ernesto Reuben of Columbia Business School, Paola Sapienza of the Kellogg School of Management at Northwestern and Luigi Zingales of the Booth School of Business at the University of Chicago, tried another version of the experiment, which they labeled "Cheap Talk." In this version, the job candidates were allowed to predict their own performance. Men tended to exaggerate their acumen, while women downplayed theirs. But the managers failed to compensate for that difference, and were again twice as likely to choose a man, NYT writes.

The bias persisted even when managers were given hard data about the applicants' ability to perform the tasks in question. Managers were still one-and-a-half times more likely to hire a man. When they knowingly chose the lower-performing candidate, two-thirds of the time they were choosing the male applicant.

The managers were also given an "implicit association test," or I.A.T., to measure their gender bias when it comes to math and science. "The very people who are biased against women about math, they're also less likely to believe that men boast," Mr. Zingales said. "So they're picking up a negative stereotype of women, but not a negative stereotype of men."

The study showed that hard evidence could reduce prejudice, Zingales said, but that it was even more important for managers to understand their own preexisting beliefs. "Anyone can do an IAT, and if they know that they are biased they should correct for that," he said.

Zingales said the findings could apply to graduate school admissions as much as to the workplace. And, he said, they shed light on why many women opt out of science and technology majors before they even reach graduation — they may assume that the negative response they are getting is based on their actual performance.

"People don't even learn," he said, "that they are equally capable."

    and share via