Why do women still make up such a small percentage of scientists?

At the Solvay Conference on Physics in 1927, the only woman in attendance was Marie Curie (bottom row, third from left).

A study published this summer by Yale researchers found that physicists, chemists, and biologists are more inclined to perceive a young male scientist with the same qualifications positively than a woman with the same qualifications. Professors at six major research universities were substantially more ready to grant the man a job after being presented with similar accounts of the accomplishments of two fictional applicants. If they did hire the woman, they paid her roughly $4,000 less on average than the man. Female scientists were just as biassed as their male counterparts, which was surprising.

The new research goes a long way toward proving that there is still a bias against women in science. Only around half of all physics Ph.D.s in the United States are awarded to women, and only about one-fifth of those women are American; only 14% of all physics professors in the United States are women. The number of black and Hispanic scientists is much smaller; on average, 13 African-Americans and 20 Latinos of either gender earn Ph.D.s in physics each year. The causes of these shortages are not obscure: many minority students attend secondary schools where they are too far behind in science to catch up, and the impacts of bias at every step of their education are widely established. But what is it that continues to keep women out of STEM professions (shorthand for "science, technology, engineering, and mathematics"), which offer so much in terms of career opportunities, status, intellectual stimulation, and income?

This subject strongly concerns me as one of the first two women to get a bachelor of science degree in physics from Yale (I graduated in 1978). I went to a rural public school where I wasn't allowed to take the few advanced physics and calculus courses because, as my principal phrased it, "girls seldom move on in science and math." I started reading about space and time and teaching myself calculus from a book because I was angry and bored. I was dreadfully unprepared when I arrived at Yale. As our professor rushed through the material, the lads in my basic physics class, who had taken significantly more challenging math and science classes in high school, yawned, while I trembled at how little I knew. As the only woman in the room, I wondered whether or not to raise my hand and risk being ridiculed, causing me to lose sight of the lecture and fall further behind.

After excelling in the department's three-term quantum mechanics sequence and a graduate course in gravitational physics, I graduated summa cum laude, Phi Beta Kappa, with honours in the major, all while teaching myself to programme Yale's mainframe computer. However, I did not pursue physics as a career. I was fatigued at the end of four years from all the lonely hours I spent catching up with my classmates, masking my fears, and battling to finish my problem sets while the lads worked in teams. I'd had it with wearing one way to be considered seriously as a scientist and another way to feel feminine. While some of the men I wanted to date were not turned off by my degree, the majority were.

But, for the most part, I didn't continue in physics because no one, not even the supervisor who oversaw my senior thesis, urged me to apply to graduate school. I slunk away in humiliation, leaving the rough draught of my senior thesis outside my adviser's door, certain that I wasn't talented enough to succeed in physics. I locked my textbooks, lab reports, and problem sets in my father's army footlocker, pained by the dream I had failed to realise, and turned my back on physics and math for good.

I didn't feel forced to reopen that footlocker until 2005, when Lawrence Summers, then-president of Harvard, questioned aloud during a luncheon address why more women don't end up having tenured posts in the hard sciences. I've known Summers since I was in high school, when he judged my debate team, and he's always impressed me as a supporter of intelligent women. I got the impression that he had asked the question because he actually cared about the answer when he indicated, among other things, that congenital differences in scientific and mathematical aptitude at the very top end of the spectrum would account for the paucity of tenured female teachers.I was taken aback by his notion that the problem could be due to biological inequities between men and women, but as I read the angry responses to his remarks, I realised that even I didn't understand why so many women were still abandoning physics and math before earning advanced degrees. I decided to look up old classmates and instructors, read up on research on women's performance in STEM professions, and go back to Yale to see what had changed since I was there. I wanted to know why I had abandoned my dream, and why so many other women continue to abandon theirs.

Of course, the climate has changed in many ways to make science and math more appealing to young women. Female students at the high school that I attended in upstate New York no longer have to learn calculus from a book, and physics is taught by a charming young lady. When I first arrived at Yale in the fall of 2010, everyone boasted that women made up 30 to 40% of undergraduates majoring in physics and related subjects. Those young women studied in a department chaired by Meg Urry, a formidable astrophysicist who earned her Ph.D. from Johns Hopkins, completed a postdoctoral fellowship at M.I.T.'s centre for space research, and served on the Hubble Space Telescope faculty before Yale hired her as a full professor in 2001. (There was no other female faculty member in the department at the time.)

Urry has dedicated herself in recent years to changing her colleagues' ideas of why there are so few women in science by utilising real evidence and anecdotes from her personal experience. In response to the Summers controversy, she wrote an essay for The Washington Post in which she described her gradual realisation that women were leaving the profession because of the "slow drumbeat of being underappreciated, feeling uncomfortable, and encountering roadblocks along the path to success," rather than because they were unqualified.

Despite the fact that Urry admitted in her op-ed column that she mistook her repeated failures to get recruited or promoted as proof that she wasn't good enough as a young scientist, anyone who meets her now would have a hard time recognising her as lacking in confidence. She has a perplexing smile, bright eyes, and an irreverent sense of humour; five people told me she was the busiest lady on campus.

Urry expected that the female students in her department would understand her and my challenges, but that their support structure would shield them from the same kind of self-doubt. For example, the students support a semi-annual Conference for Undergraduate Women in Physics at Yale under the guidance of Bonnie Fleming, Yale's second female tenured professor of physics. Urry further stated that because there were so many women studying physics at Yale, and so many of them were at the top of their class, the faculty couldn't help but notice that their abilities were comparable to men's. Urry said she would attempt to attend a tea that afternoon where I would be able to interview female students interested in science and gender.

Judith Krauss, the professor hosting the tea (she is the former dean of nursing and currently master of Silliman College, where I was an undergraduate), told me that just a few students would turn up. Krauss and I were taken aback when 80 young women (plus three interested males) poured into the room. Urry was lucky to find a seat by the time he rushed in.

The students were eager to tell their stories. One young woman was concerned when she discovered she was one of just three girls in her AP physics class in high school, and she was even more concerned when the other two dropped out. From the beginning, another student was the lone girl in her AP physics class. "You're a female," her peers mocked her relentlessly. "Physics is not for girls." She had hoped that the teacher would stop the tormenting, but he didn't.

Other women weighed in to express that it was their instructors that teased them the most. When asked why, the teacher said that he couldn't reasonably expect a girl to compete in physics on an equal footing with boys; when asked why, the teacher answered that he couldn't properly expect a girl to compete in physics on an equal footing with boys.

Women who had attended all-girls secondary schools or had grown up in foreign countries were the only ones in the audience who didn't understand what the rest of the crowd was talking about. (At the tea and elsewhere, the lesbian scientists I met with expressed varying reactions to the gender dynamic in the classroom and lab, but shared many of the same worries as the straight women.) One girl, who I assumed was Indian or Pakistani, claimed she came to school having taken a number of advanced classes and didn't hesitate to enrol in the most difficult math course. She addressed the professor, shaken to find herself the lone girl in the class and unable to follow the first lecture: Should I be here? "You shouldn't take the class if you're not confident that you should be here," she said, imitating his derision.

A dozen girls lingered after the tea to discuss. One astrophysics major grumbled, "The boys in my group don't take anything I say seriously." "I despise being confrontational. Is that all there is to it? That was not my upbringing. Is it true that I'll have to be as aggressive in graduate school? "Will I be able to do this for the rest of my life?" Another stated she didn't like going out to a bar with her sister and her sister introducing her as an astrophysics major. "I stomp on her beneath the table." I despise it when people find out I'm a physics major in a bar or at a party. I can imagine the males turning away the moment they find out." Another went on to say that even at Yale, males wouldn't date a physics major, and that she was scared she'd go four years without dating.

I asked Urry if she was as surprised as I was after the students had left. She answered, "More," because she was the chairwoman of the department where the majority of these females were enrolled.

Over the next two years, I heard similar stories from young women in Michigan, upstate New York, and Connecticut. I was shocked to see that the cultural and psychological variables that I encountered in the 1970s not only survive, but appear to be even more damaging in a culture where women are persuaded that nothing stands in their way of achieving success in any area. The expectations to be traditionally feminine appear to be more stronger now than when I was younger.

Watch an episode of the popular television show "The Big Bang Theory," about a group of awkward but endearing male Caltech physicists and their neighbour, Penny, an attractive blonde who has moved to L.A. to make it as an actress, to see how stereotypes continue to shape American attitudes about science, and especially about women in science. Despite the fact that two of the scientists on the show are female, one of them, Bernadette, speaks in a strident voice that could shatter a test tube. Rather than working in a lab, as any true doctoral student would, she waitressed with Penny while pursuing her Ph.D. in microbiology. Mayim Bialik, who plays Amy, a neurobiologist who develops a semiromantic relationship with the immature but brilliant physicist Sheldon, has a Ph.D. in neuroscience and is far from the horrifically dumpy lady depicted on the programme. Of course, "The Big Bang Theory" is a sitcom, so every character is a caricature, but what slightly normal young person would want to work in an environment with misfits like Sheldon, Howard, and Raj? And what slightly sane young lady would prefer to envision herself as dowdy, socially inept Amy instead of trendy, bouncing, math-and-science-illiterate Penny?

Although scientists are stereotyped as geeks in the United States, a talent for arithmetic is generally considered as a sign of intuition and creativity in other cultures. In an effort to identify top performers, the American Mathematical Society provided data from a number of prominent international competitions in 2008. Almost all of the American contestants were immigrants' children, and only a few were female. Bulgaria, for example, sent 21 girls to the International Mathematical Olympiad between 1959 and 2008, whereas the United States sent only three from 1974 to 2008, when it first entered the tournament; no women ever made the American squad until 1998. The study's authors claim that native-born American students of both sexes avoid math groups and contests because "only Asians and nerds" would willingly undertake math. "In other words, pursuing mathematics for fun is considered uncool in the social environment of American middle and high schools, and doing so can lead to social ostracism." As a result, gifted girls, even more so than boys, conceal their mathematical abilities in order to fit in with their peers."

The conclusions of the study are applicable to science as well. Urry told me that the women from Italy and France "dress really nicely, what Americans would call exposing," at the space telescope institute where she used to work. A Frenchwoman will be wearing a short skirt and fishnets, as is customary for them. Men in those countries appear to be able to distinguish a woman's sexual identity from her scientific identity. "It seems that American men can't accept a woman as both a lady and a scientist; it's either one or the other."

It appears beyond debate that the discrepancy in men and women's representation in science and math is due to culture rather than heredity. In the early 1980s, a large cohort of American middle-schoolers took the SAT math exam; guys outpaced girls by 13 to 1 among those who scored higher than 700. However, even in middle school, a SAT score of 700 or more does not always indicate actual mathematical inventiveness or mastery of higher-level subjects. These were all students from the United States. The examination of the top performers in international contests by the mathematical society went much further in studying genius by examining the performance of young women from diverse cultures. What did the study's conclusion say? "The scarcity of women at the very top echelons is due, in large part, to variable factors that vary with time, country, and ethnic group," says the report. To begin with, some countries find and nurture females with exceptional mathematical talent at a considerably higher rate than others." Furthermore, the ratio of boys to girls scoring 700 or above on the math SAT in middle school is now barely three to one. How could their scores have increased so steadily in such a short time if girls were so bound by their biology?

In elementary school, girls and boys perform equally well in math and science. But by the time they reach high school, when those subjects begin to seem more difficult to students of both sexes, the numbers diverge. Although the percentage of girls among all students taking high-school physics rose to 47 percent in 1997 from about 39 percent in 1987, that figure has remained constant into the new millennium. And the numbers become more alarming when you look at AP classes rather than general physics, and at the scores on AP exams rather than mere attendance in AP classes. The statistics tend to be a bit more encouraging in AP calculus, but they are far worse in computer science. Maybe boys care more about physics and computer science than girls do. But an equally plausible explanation is that boys are encouraged to tough out difficult courses in unpopular subjects, while girls, no matter how smart, receive fewer arguments from their parents, teachers or guidance counsellors if they drop a physics class or shrug off an AP exam.

It has long been recognised that cultural cues can influence a student's capacity to perform well on an exam. A group of University of Michigan students with similar backgrounds and math abilities were placed into two groups in a widely referenced 1999 research. In the first, students were told that males outperform women on arithmetic examinations; in the second, they were guaranteed that, contrary to popular belief, there was no difference in male and female performance. A arithmetic test was administered to both groups. The men outscored the women by 20 points in the first round, but only by 2 points in the second.

It's also feasible that gifts in science and math aren't measurable by standardised exam scores. On the math SATs, less than a third of white American males in engineering, computer science, math, and the physical sciences scored higher than 650, while more than a third scored below 550. Hard work, persistence, and encouragement appear to be just as vital as raw skill in the middle ranks. Even at the greatest levels, test scores may be useless; according to reports, Richard Feynman's I.Q. was a mediocre 125.

It's possible that the most effective indicator of whether a woman continues in science is whether or not anyone encourages her to do so. On my first physics midterm during my freshman year at Yale, I received a 32. My parents pushed me to change my major. They only cared that I could make a living until I married a guy who could support me, and physics seemed unlikely to help me achieve either goal.

I trudged up Science Hill, asking Michael Zeller, my professor, to sign my withdrawal form. I took the elevator to Professor Zeller's floor, then wound my way through corridors decorated with portraits of the all-male faculty and notices for lectures with titles that I couldn't understand. I knocked on my professor's door and stammered that I had received a 32 on my midterm and that I needed him to sign my drop slip.

"Why?" he inquired. In two of his physics classes, he scored Ds. Not on midterms, but in classes. The storey seemed like something a kind professor would make up to help his least gifted pupil feel less stupid. The D's in his case were clearly anomalies. The 32 in my instance meant that I wasn't very strong at physics.

I trudged up Science Hill, asking Michael Zeller, my professor, to sign my withdrawal form. I took the elevator to Professor Zeller's floor, then wound my way through corridors decorated with portraits of the all-male faculty and notices for lectures with titles that I couldn't understand. I knocked on my professor's door and stammered that I had received a 32 on my midterm and that I needed him to sign my drop slip.

"Why?" he inquired. In two of his physics classes, he scored Ds. Not on midterms, but in classes. The storey seemed like something a kind professor would make up to help his least gifted pupil feel less stupid. The D's in his case were clearly anomalies. The 32 in my instance meant that I wasn't very strong at physics.

He advised, "Just swim in your own lane." When he saw I was perplexed, he revealed that he had been a member of the Stanford swim team. His stroke was on par with anyone else's. But he always came in second. "Your problem, Zeller," the coach explained, "is that you constantly glancing around to see how the other guys are doing." You'll win if you keep your attention on your own lane and swim your quickest."

I deduced that he wasn't going to sign my drop slip.

He said, "You can do it." "Stick it out," says the narrator.

I completed the course. I fought to complete my problem sets week after week until they no longer felt insurmountable. More equations decorated with comet-like exclamation marks and theorems whose brilliance I noted with exploding novas of hot-pink asterisks appear as I go deeper into my four-inch-thick freshman physics textbook. The book's markings transport me back to a time when, sitting in my cramped dorm room, I suddenly realised a principle that governs how objects interact, whether here on Earth or thousands of light years away, and I marvelled that such vastness and complexity could be reduced to the equation I had highlighted in my book. Is there anything more thrilling than grasping a completely new way of seeing, a reality that is more real than the actual itself?

I received a B in the course the following semester and an A the following semester. By the beginning of my senior year, I was at the top of my class and had the most research experience. However, not a single professor inquired about my plans for graduate school. When I mentioned shyly to Professor Zeller that my dream was to apply to Princeton and become a theoretician, he shook his head and said that if you went to Princeton, you had better put your ego in your back pocket because those guys were so brilliant and competitive that your ego would be crushed, which made me feel as if I wasn't brilliant or competitive enough to apply.

Even the math professor who oversaw my senior thesis didn't encourage me to pursue a Ph.D. I'd been missing parties, skipping dinners, and losing sleep for nine months trying to figure out why waves of sound, light, or anything else travel in a spherical shell, like the skin of a balloon, in any odd-dimensional area, but like a solid bowling ball in any even-dimensional environment. When I finally had the solution, I triumphantly knocked on my adviser's door. Yet I don’t remember his thanking me in any manner. I couldn't wait to ask if my ability to solve the problem meant I had a chance as a theoretical physicist. But I realised I wasn't in the position to ask.

When I called the same professor, mathematician Roger Howe, years later, he happily agreed to meet with me to talk about women in science and math. We met at his office, which is still home to a giant banner of great mathematicians (all male) in the foyer, albeit someone has tacked a tiny poster of "famous women in math" next to the women's lavatory on the top floor. Even though he was the youngest full professor at Yale at the time I studied with him, Howe appeared incredibly young. He recommended we get a sandwich, and while we sat waiting for our panini, I told him that one of the reasons I didn't go to graduate school was because I rated my abilities in comparison to his. After all, I'd had such a tough time addressing the task he'd set before me.

He appeared perplexed. "However, you figured it out."

"Yes," I said. "At the end of the day, I knew exactly what I was doing." But it took such a long time for me to figure it out."

"But that's the way it is," he explained. "You don't notice it until it's too late, and then you wonder why you didn't notice it sooner."

However, I had to drop my real-world analysis class.

Howe shrugged his shoulders. There are many different types of math personalities. Different mathematicians excel in various disciplines.

I inquired as to whether he had noticed any differences in how male and female students approach math issues, or if they had distinct "math personalities." He said no. On the other hand, he couldn't get inside the heads of his students. In math, he had two female pupils continue, and both had done well.

I inquired as to why there were no female math professors on Yale's faculty even now. Howe corrected me when I said there were no tenured women. The department voted in 2010 to employ a woman for a tenure-track position. (While that lady has yet to be considered for tenure, the faculty did appoint a senior female professor this year.) As I already stated, that is still a small number. He fixed his gaze on the horizon. "I guess I just haven't come across that many ladies whose work I admire." I used to watch him picture n-dimensional toruses cradled in his hands while he considered his answer. "Perhaps women are victims of misconception," he finally said. One of his colleagues at another institution recently confided to him that when they were first starting out, there were two people in his field, a woman and a guy, and this colleague felt the man must be the better mathematician, but the woman has gone on to do better work.

I finally came out and asked how he felt about my project. How did it stack up against all of the other student research projects he had to oversee?

His brows were raised, as if he were expressing the mathematical symbol for perplexity. In fact, in his entire career, he had only supervised two or three students. "It's quite rare for an undergraduate to pursue an individual project in mathematics," he explained. "By that standard, I'd think what you accomplished was excellent."

"Exceptional?" I said again. So why hadn't he told me before?

He was taken aback by the question. After all, he was now the director of undergraduate studies, so I inquired if he had ever particularly urged any undergraduates to pursue Ph.D.s. He said, however, that he never pushed anyone to pursue a career in mathematics. "It's a very difficult life," he admitted. "You must have pleasure in it." Being a mathematician comes with a lot of stress. It's a difficult existence, a difficult culture."

Meg Urry blew raspberries when I told her that Howe and several other of my instructors indicated they don't encourage anyone to continue in physics or math because it's such a difficult field. "Oh, please," she exclaimed. "They're in charge of themselves." They're nicely compensated. They are passionate about what they do. Why not inspire others to pursue their passions?" "There's usually a woman who comes up to me and says the same thing you said, that I wanted to be a physicist but no one pushed me," she says. If even one person had told me, 'You can accomplish this,' I would have believed it. She burst out laughing. "More positive reinforcement is needed for women, while more negative reinforcement is needed for men. Men vastly overestimate their ability to study as well as their earning potential. "Oh, I'm not good, I won't earn much, anything you want to give me is OK," women say.

One student informed Urry that she didn't think she was good enough for graduate school, and Urry wondered why, given that she had nearly all A's at Yale, which has one of the most difficult physics programmes in the country. "She didn't believe she was qualified, whereas I've written a number of letters for men with B averages." She will not claim that earning a Ph.D. is simple. "It's a slog." 'How is this going to be for me?' said a young woman. Yes, there are easier things to do, I must admit. That isn't to say I shouldn't encourage her to try. To do what I do, you don't have to be a genius. 'Oh, Meg, you have to be a genius to be an astronomer,' my adviser remarked when I told him what I wanted to do. I was their finest physics student. He was implying that I wasn't a genius, that I wasn't good enough. So, all those theorists out there are Feynman or Einstein? "I don't believe so."

I met five young Yale alumnae at a Vietnamese restaurant in Cambridge not long ago. Three of the women were studying physics and astronomy in graduate school at Harvard, and two were studying oceanography at M.I.T. None of the five stated they were anxious about surviving graduate school, but they were all concerned about how they would teach and perform research once they had children.

One woman commented, "That's where you lose all the female physicists."

"Yeah, getting your kid into child care at M.I.T. is considerably harder," added another.

Most young women are unaware, according to Urry, that being a female scientist allows her greater flexibility than most other occupations. On her first day at the Goddard Space Flight Center, she met her husband. She added, "And we have a perfectly equal relationship." "He doesn't say he's assisting me when he looks after the kids." Nobody is pretending that balancing a scientific profession with raising children is simple. Having a family while establishing a profession as a doctor or a lawyer isn't simple, but that doesn't stop women from pursuing those careers. Urry believes that having a family is often used as an explanation for women leaving science when, in reality, they have been discouraged to the point of quitting.

All Ph.D. candidates must compete for a junior job, write grants, and undertake sufficient research to gain tenure. Yet, in the tenure race, women must overcome obstacles that are higher than those faced by their male counterparts, frequently without recognising it.

In the mid-1990s, three senior female professors at M.I.T. began to believe that similar patterns of marginalisation had harmed their careers. They complained to the dean, who formed a committee made up of six senior women and three senior males to look into their concerns. The committee concluded that the marginalisation of female scientists at M.I.T. "was often accompanied by differences in salary, space, awards, resources, and response to outside offers between men and women faculty, with women receiving less despite professional accomplishments equal to those of their colleagues" after conducting the investigation and studying the data. The dean agreed with the findings of the committee. Nonetheless, his colleagues administrators "resisted the premise that there was any problem that originated from gender prejudice in the treatment of the women professors," according to the committee's report. Some claimed that M.I.T.'s male culture was to fault, and that little could be done to modify it." In other words, women did not pursue careers in research because science — and scientists — were dominated by men.

The committee's most striking conclusion was that female scientists faced discrimination in the final quarter of the twentieth century that was qualitatively distinct from the more evident forms of sexism addressed by civil rights laws and affirmative action, but no less genuine. "I have found that even when women win the Nobel Prize, someone is bound to tell me they did not deserve it, or the discovery was really made by a man, or the important result was really made by a man, or the woman really isn't that smart," said Nancy Hopkins, one of the professors who initiated the study, in an online forum. In 2011, this is what discrimination looks like."

Not everyone agrees that the discrimination discovered at M.I.T. counts as discrimination. The M.I.T. study, according to Judith Kleinfeld, an emeritus professor in the psychology department at the University of Alaska, isn't convincing since the number of faculty members involved is too small, and university administrators refuse to provide the data. Even if female professors have been shortchanged or overlooked, their marginalisation could be due to the same types of departmental squabbles, personality clashes, and "mistaken perceptions" that make male faculty members feel slighted. "Discrimination perceptions are nothing more than subjective sensations," Kleinfeld scoffs.

However, larger studies demonstrate that the sense of discrimination is frequently accompanied with a very genuine disparity in resource allocation. The American Institute of Physics published a survey of 15,000 male and female physicists from 130 countries in February 2012. Even after the researchers adjusted for characteristics other than sex, female scientists received less funding, lab space, office support, and grants for equipment and travel in almost all societies. "Women physicists could be the majority in some possible future," the researchers concluded, "yet still face obstacles in their careers due to sometimes unconscious bias."

Jo Handelsman's research focuses on soil microorganisms and insect gut microbes, but she has also worked to increase the participation of women and minorities in science since the early 1990s. She had long thought that the same subtle prejudices that had been shown in the general population were at work among scientists, but she had no evidence to back up her claims. "People replied, 'Oh, that might happen in the Midwest or the South, but not in New England, and certainly not in my department, because we just graduated a woman.' "That only happens in economics," they'd say. I have women in my lab,' said a male scientist to Handelsman. "The women in my class are smarter than the males!" "They go to their experience with a sample size of one," she explained. She burst out laughing. "Scientists can be so irrational."

In 2010, Handelsman teamed up with Corinne Moss-Racusin, then a postdoctoral associate at Yale, to start work on the study, which was published last year and documented gender bias in American faculty members in three scientific fields — physics, chemistry, and biology — at six major research institutions across the country.

Moss-Racusin devised a study with collaborators from the psychology, psychiatry, and School of Management that involved sending out identical résumés to professors of both genders, with a cover page stating that the young applicant had recently completed a bachelor's degree and was now looking for a position as a lab manager. Half of the 127 participants were given a résumé for a student called John, while the other half were given Jennifer's résumé. The applicant's qualifications were enough for the job in both cases (with favourable letters of recommendation and coauthorship of a journal paper), but not enormously persuasive — the applicant's G.P.A. was only 3.2, and he or she had dropped one scientific class. On a scale of one to seven, each faculty member was asked to rank John or Jennifer in terms of competency, hireability, likability, and the professor's willingness to mentor the student. Professors were then asked to select a wage range within which they would be willing to hire the candidate.

The outcomes were astonishing. In every category except likability, where Jennifer scored nearly half a point higher, John was rated half a point higher than Jennifer, regardless of the respondent's age, sex, area of speciality, or degree of seniority. In addition, John was offered a beginning salary of $30,238, compared to $26,508 for Jennifer. When Handelsman and Moss-Racusin present the graph to a group of psychologists, "we hear a collective gasp, the importance is truly that big," she said.

I asked Handelsman if she was shocked that senior female faculty members, regardless of age, showed the same level of bias as male professors, and she responded no; she had seen too many research with comparable results. She was also unsurprised that, despite the presence of more female biologists in most departments, the bias against women in biology was as prevalent as in physics or chemistry. She claims that while biologists may see women in their labs, their preconceptions are established by images and attitudes they've been exposed to since birth. In some ways, Handelsman is relieved that the women she investigated were just as biassed as the men. "You can watch the tension in the room diminish" when she gives a talk and exposes the results, she said. 'We all do it,' I can say. It isn't just you. This isn't just something the evil guys do.'

I asked Handelsman about the typical complaint that John is a more powerful name than Jennifer. With a shake of her head, she expressed her dissatisfaction with the situation. "Believe me, it's not just a matter of syllables," she explained. "Studies have been conducted to determine whether names express the same traits to survey respondents, and John and Jennifer are usually regarded as conveying the same amount of respectability and competence." The fact that faculty members prefer Jennifer to John adds to the insidiousness of the concealed bias. As the authors point out, their findings are consistent with those of other studies that show that people's biases are the result of "repeated exposure to pervasive cultural stereotypes that portray women as less competent while simultaneously emphasising their warmth and likability in comparison to men."

When you combine that unconscious institutional bias with the psychological bias many young female scientists report suffering against their own skills, the results are very alarming. Handelsman found the mentorship results to be the most devastating of all the data she uncovered in her research. "If you add up all the little interactions a student has with a professor — asking questions after class, an adviser recommending which courses to take or suggesting what a student might do for the coming summer, whether he or she should apply for a research programme, whether to go on to graduate school, all those mini-interactions that students use to gauge what we think of them so they'll know whether to go on or not, all those mini-interactions that students use to gauge what we think of them so You'd think they'd figure it out on their own, but they don't." With a shake of her head, Handelsman expressed her dissatisfaction with the situation. "All the small kicks that women get, as opposed to all the replies that males get that make them feel more a part of the party," says the author.

Some detractors contend that if women do not pursue careers in science, there is no real harm done. Vanderbilt University psychologists David Lubinski and Camilla Persson Benbow spent decades studying thousands of mathematically gifted 12-year-olds. What was their conclusion? The girls tended to be "more rounded" from the outset, keen to work with people, plants, and animals rather than objects. Although the boys were more likely to pursue jobs in math or science, the women were equally likely to pursue postgraduate degrees and high-level positions in subjects such as law, medicine, and the social sciences. Men and women in their mid-30s looked to be equally satisfied with their life choices and saw themselves as equally successful.

Nonetheless, I believe the premise that women are underrepresented in science because they believe they will be happier in "people" jobs is misguided.

The issue is that most girls — and boys — decide they don't enjoy math and science before they discover its inherent beauty, which is exacerbated by the unimaginative manner in which science and math are taught. Last year, the President's Council of Advisers on Science and Technology issued an urgent need for significant reform if the United States is to fulfil the demand for one million more STEM workers in the next decade than it is now on course to create.

But, beyond upgrading our curriculum, we must ensure that we do not lose girls at every turn as a result of their low self-esteem, misunderstandings about who goes on in science, and wrong judgments of their abilities.

As difficult as such transformation may seem, it is not insurmountable. Danica McKellar's book "Math Doesn't Suck," as well as her follow-up publications "Kiss My Math," "Hot X: Algebra Exposed," and "Girls Get Curves: Geometry Takes Shape," may have done more to inspire girls to stay with math than any government task group. McKellar's math books may go a little too far in pandering to adolescent girls' stereotypical obsessions (the problems involve best friends, beads, and Barbies rather than baseballs and speeding cars), but the overwhelmingly positive response they've gotten demonstrates the impact that reworking the contents of standard math and science problems can have in dispelling the myth that boys won't like smart girls.

Persuading educators, researchers, and administrators that expanding the pool of female scientists and making the culture more livable for them does not degrade standards is the key to improvement. If society need a specific number of scientists, Urry explained, and you can just look for them among the males in the community, you will have to search far further down the barrel than if you can also look among the ladies in the population, particularly those at the top of their barrel.

Her peers, she added, must also appreciate the potential of women who discover a passion for science later in life. According to studies, a child's early interest in science has little to do with their abilities. You can grow up to be a science freak and not be excellent at research, or you can come to science late and turn out to be a genius, according to Urry.

Girls can overcome an initial disadvantage working with machines, tools, and electrical equipment with a little effort and confidence. While studies suggest that spatial ability is a consequence of experience, boys regularly outperform girls in exams that measure the spatial skills required for lab work and engineering. The administration at Massachusetts' Olin College of Engineering is committed to ensuring that 50% of each entering class is made up of women. The first semester at Olin, all entering students must take a machining course. "Everyone is faced straight on with gender inequalities in the lab," says Yevgeniya Zastavker, a faculty member who performs biophysics research and investigates the role of gender in science. We put them in mixed-gender groups and ask them to create a tool or a product. We intervene if the gender relations get strange, and that one intervention early on has a huge impact."

Urry scoffed at my lab anecdotes about dripping acid on my stockings, which dissolved and blew up in smoke, being thrown across the room by a jolt from an ungrounded oscilloscope, and not being able to recreate the Millikan oil-drop experiment. Even in college, she had been a mess in the lab. She didn't realise she knew as much as her classmates until she took a more difficult lab and spent hours looking over a circuit diagram, figuring out that they had set up an experiment incorrectly.

"I'm soldering stuff and thinking to myself, 'Hey, I'm really good at this.' I understand the fundamentals. It's like a work of art. It took me a long time to understand that I'm actually rather skilled with my hands. All of these small-motor abilities are the result of all of the years I spent sewing, knitting, and creating. 'That material truly prepares you for working in a lab,' we should inform young women.

Scientists of all genders must recognise that they can't always perceive how their bias affects their day-to-day lives, as the Yale study revealed. Handelsman's study, according to Abigail Stewart, director of the University of Michigan's Advance programme, which aims to improve the lives of female and minority faculty members, "shakes the passionately held belief of most scientists that they are dedicated to accurately identifying and nurturing merit in their students." "Evidence suggesting we are less likely to detect and support talent (even small skill, as in this study) rattles our confidence and (I hope) makes us more aware of our limitations in recognising talent in unexpected places."

Urry, like Stewart, believes Handelsman's findings will help precipitate the changes she has been advocating for for years. "I've long believed that knowing the existence of implicit prejudice is crucial. If you believe that the playing field is level, whatever action you take will favour women. However, if you are aware that women are undervalued, you must act, or you will lose qualified personnel."

Above all, we must ensure that women — and men — do not grow up in a society where scientists are portrayed as geeky male misfits. Gender inequalities in high-school physics enrollment rates appear to be connected with the percentage of women in the greater community who work in STEM disciplines, according to Catherine Riegle-Crumb, an associate professor at the University of Texas at Austin. "Say, a big mural with Rosalind Franklin in the front and Watson and Crick in tiny proportion in the back," Handelsman, who is awaiting Senate confirmation as associate director of science in the White House Office for Science and Technology Policy, told me she would love to see murals of women scientists painted on the walls of Yale's classrooms.

The good news is that real change is slowly but surely taking place as more institutions recognise gender bias and implement initiatives to address it. When I was at Yale, Peter Parker, who was the director of undergraduate studies in physics at the time and for many years afterward, told Urry that he wasn't shocked that the department's students and faculty were all male. Urry noted that in his latter years, he would shout with delight that, for example, 21 out of 49 physics majors in the junior class that year were women. Roger Howe recently emailed me to tell that he had a bright female student and asked if I could contact her and offer some guidance and encouragement. M.I.T. currently has a large number of female administrators, 19 years after those three senior women began comparing their experiences and demanding improvements. There is a greater availability of day care. It is more acceptable for faculty members to have children before achieving tenure. And deans and department chairmen appear to be committed to boosting the number of women in academia.

Urry, who left Yale's physics department as chairwoman this summer but will shortly become president of the American Astronomical Society, is unsure whether her department's commitment to gender equality will continue or stop. She invited me to a picnic hosted by the physics and astronomy departments to welcome back their graduate students and teachers on an autumn Friday. Professors sipped wine from disposable glasses while catching up with colleagues they hadn't seen since the summer. Hungry graduate students scanned tables piled high with salad bowls, barbecue preparations, pies, cakes, and a plate of brownies cooked by Urry's husband when he realised she had overslept that morning. Four young ladies — one black, two white, and one Asian from Australia — told me how they had persevered while so many others had given up.

One of them said, "Oh, that's simple." "We're the women who don't give a damn," says one of them.

Do you give a damn about —?

"What the public expects of us."

"Or don't."

"Or the fact that men don't take you seriously because you dress like a woman." I figure it's your issue if you don't take my science seriously because of how I appear."

"Let's face it, grad school is a hazing for everyone, male or female," one of the ladies stated. However, if your class has enough ladies, you can help each other get through."

The young black woman told me she attended a historically black college for her undergraduate studies before enrolling in a master's programme aimed at helping minority students acquire research skills and one-on-one mentoring relationships that would help them move to a Ph.D. programme. Her first year at Yale was difficult, but she was supported by her mentors. "Success is the best revenge," she added, citing her mother's advice.

As numerous studies have shown, success in math and the hard sciences is almost entirely dependent on culture — a culture that teaches girls that math isn't cool and that no one will date them if they excel in physics; a culture in which professors rarely encourage their female students to pursue advanced degrees; a culture in which graduate school success is a matter of isolation, competition, and absurdly long hours in the library.

And yet, as I listened to these four young ladies laugh at the prejudices and anxieties that had deterred so many others, I felt encouraged that even these few had made it this far, that theirs will be the faces that the future generation envisions when they think of a female scientist.

On Oct. 6, a figure about girls taking high-school physics was misrepresented in an article about the situation of women in science and mathematics. It was the percentage of girls taking high-school physics among all high-school physics students of both sexes that increased to 47% in 1997 from 39% in 1987, not the percentage of females taking high-school physics among all high-school girls. The status of a woman hired for a tenure-track post at Yale University's mathematics department in 2010 was similarly misrepresented in the article. She has yet to be offered tenure. It was not denied to her.

A statistic about girls taking high-school physics was misrepresented in an article on Oct. 6 about the status of women in science and mathematics. The percentage of females taking high-school physics among all high-school students of both sexes increased to 47 percent in 1997 from 39 percent in 1987, not the percentage of girls taking high-school physics among all high-school girls. The status of a woman hired for a tenure-track position at Yale University's mathematics department in 2010 was likewise incorrect in the article. She has not yet been offered a position. It wasn't a problem for her.