In Science Author: Robert Finn
A recently released study from Harvard University examining the careers of scientists who showed high promise as postdocs has found persistent gender differences in career outcomes. The study, called Project Access, reveals clear evidence of a glass ceiling for women in certain fields, notes differences in publication patterns, and elucidates the way that family-related issues-such as raising children and living in a two-scientist household-disproportionately affect women.
INVESTIGATORS: Gerhard Sonnert, left and Gerald Holton examined 508 men and 191 women scientists.
Among the study's findings is that women in the physical sciences, math, and engineering who received their Ph.D.'s before 1978 are only about half as likely as men to have achieved the rank of full professor. And female scientists with children are at least six times more likely than male scientists with children to be working part-time.
"It is important to remember," the study's authors write, "that the gender gap in science has diminished during recent decades. . . . Nonetheless, gender disparities do remain; they are just more subtle."
Project Access is the first of three major studies of gender disparities in science expected to be released over the next few months. A longitudinal study of a matched sample of 92,904 scientists and engineers who received Ph.D.'s between 1973 and 1989 is under review at the National Research Council (NRC), and is expected to be issued by the end of the year. And Mary Frank Fox, a professor of sociology at the Georgia Institute of Technology in Atlanta, will present the results of her survey of 5,400 doctoral candidates and faculty members at the annual meeting of the American Association for the Advancement of Science in February.
The results of Project Access-conducted by sociologist Gerhard Sonnert, an associate in the Harvard physics department, with the assistance of Gerald Holton, Harvard's Mallinckrodt Professor of Physics and Professor of History of Science-are detailed in two books recently published by Rutgers University Press. Who Succeeds in Science? The Gender Dimension was released last month. And Gender Differences in Science Careers: The Project Access Study was put out in July.
Notes Holton of the two books, "The first one is kind of a monograph of the sociological sort, and the second one is more of a how-to book: how to maximize one's chances and how to minimize ill luck and ill fortune."
Both books make use of questionnaires completed by 699 scientists (508 men, 191 women), all of whom had received or were finalists for prestigious postdoctoral fellowships from the National Science Foundation, NRC, or the Bunting Institute of Radcliffe College (the United States' oldest and largest center of advanced studies for women artists and scholars, including scientists). Two hundred of the individuals surveyed (92 men, 108 women) also participated in two- to three-hour open-ended interviews about their careers.
In addition to discussing statistical information on the entire sample, Who Succeeds in Science? profiles 20 of the interview subjects, 10 of whom have had successful academic research careers and 10 others who have left academic research. In each case, five of those profiled are men and five are women.
"I, for one, think that this will certainly show . . . the traps that even beckon to people of high promise," comments Holton. "If the high-promising ones make those mistakes, the ones of low promise are more likely to make those mistakes."
Sonnert and Holton examined their data in the context of two opposing models that have been used to explain gender disparities in science. The "deficit" model says that women are treated differently, and the "difference" model says that women act differently. Evidence of formal or informal institutional barriers to women would support the deficit model, and evidence that, for example, men are more aggressive-either innately or through gender-role socialization-would support the difference model.
PROGRESS NOTED: Mary Good says women can now get a foot in the door. Since blatant gender discrimination was made illegal in the U.S. early in the 1970s, the overwhelming influence of the deficit model has waned, maintain Sonnert and Holton. Chemist Mary L. Good, now undersecretary for technology administration at the United States Department of Commerce, believes that "there has been a lot of progress, and the progress has particularly been rather good in the whole issue of access, of getting your foot in the door."
"I don't think it's either-or," explains Sonnert. "Both models can apply to the same scientist and to different extents in one scientist or another. The key point we're making is these two models would actually compound in bringing about career outcomes. One of the more successful women scientists that we feature fought her main battles against herself. She was plagued by the idea that she did not belong there and a lack of self-confidence. She battled this and she overcame it eventually. This would be an example where the difference model applied."
But things are not always even that clear. Referring to the Project Access findings, Catherine Didion, executive director of the Washington D.C.-based Association for Women in Science (AWIS), notes, "Many more women than men thought that they should have had more confidence or been more assertive in previous career obstacles. And you could go back and forth as to whether that is a difference or if that's the result of a deficit in not having the opportunity to learn the skills." One area in which there is a clear difference between men and women scientists is in patterns of publication. "We found what I think almost every other study found, also," says Sonnert. "Men published more articles than women did." The study's data show that male scientists produced an average of 2.8 publications per year, while women produced an average of only 2.3.
But that doesn't mean men are better scientists, Holton points out: "Men as a whole tend to optimize [the postdoctoral period] in one way, namely by writing a large number of articles, therefore putting themselves on the map as producers. Sometimes this is called 'salami slicing' [dividing a single study up into as many 'least publishable units' as possible] in the extreme.
"Women, we discovered-and this was quite unexpected to me-seemed somehow to put the eventual effect of publication rate secondary to their deep interest in doing good science no matter how long it must last. Therefore they turned out far fewer articles. But what we found was that these articles are much more cited than the ones that men do," and thus have a larger impact on the field. Among biologists surveyed in a pilot study by Sonnert, for example, articles by women averaged 22.4 citations, while articles by men were cited an average of 14.4 times.
The gender differences in publication pattern strike a chord with Lydia Villa-Komaroff, a molecular biologist who is currently an associate professor of neurology at Harvard University and Children's Hospital in Boston. Some years ago, she suffered through a bruising tenure battle at the University of Massachusetts Medical Center in Worcester. "One of the reasons that the tenure process was so difficult at [Massachusetts] was that, in fact, my productivity at that time was quite poor," she recalls. One factor, she explains, was that she refused to submit any paper for publication until it was "just right."
Another clear difference between men and women, the study found, is in the area of family issues. Sonnert and Holton talk about the difficulty women scientists have in solving the "three-clock problem": A woman must synchronize her biological clock, her career clock, and her partner's career clock. But women increasingly are finding solutions to these problems. Says Sonnert, "There is the intuition that for a woman scientist to have a child, or even to be married, is detrimental to her career; but we didn't find such a strong link in our study if we looked at global indicators. I don't think you can give any global advice-[such as] if you're a woman in science do not marry or do not have kids-because some of our subjects were able to combine these two and did very well in science. Others didn't do that well, but it's a highly complex matter."
PRODUCTIVITIY: Mary Frank Fox found that women with children publish at least as those without. Mary Frank Fox agrees. "The common wisdom on the subject is this: Good scientists are either married men or they're women who either aren't married or don't have children," she maintains. "But both my data and the data of others fly in the face of that conventional wisdom. Married women in science publish as much or more as unmarried women. And women with children publish as much or more as childless women" (M.F. Fox, "Women and scientific careers" in S. Jasanoff et al., eds., Handbook of Science and Technology Study, Thousand Oaks, Calif., Sage Publications, 1995).
Then there's the "two-body" problem. Women scientists are far more likely to be married to another scientist than men are. In the Project Access sample, 62 percent of the married women, but only 19 percent of the married men, had a spouse with a doctorate.
"In many cases," notes AWIS's Didion, "the men are a few years older than the women," and thus more established and less able to relocate to help advance a wife's career. "So you have these legions of spouses . . . who have put themselves in a different career pattern by not going to the best places to do a postdoc."
Harvard's Villa-Komaroff is married to physician Anthony L. Komaroff, a specialist in internal medicine who is six years older than she is. The two-body and three-clock problems have been a persistent theme during their 25-year marriage.
The problems first became apparent when, as an undergraduate at the University of Washington in Seattle, Villa-Komaroff transferred to Goucher College in Baltimore to be close to her future husband, who had a position at the National Institutes of Health. Later she chose to go to the Massachusetts Institute of Technology for her graduate studies, largely because she wished to follow her husband to Boston. But until recently, the couple have been able to manage their careers satisfactorily, they maintain.
"Our careers are somewhat staggered," explains Villa-Komaroff. "We've been extremely fortunate in that until this point we have been able to find things that we both find extremely satisfying in the same town." They've lived in Boston for 25 years, one of the easiest cities in the U.S. to conduct dual scientific and professional careers because of its high concentration of colleges and universities. "Now I have a new opportunity at Northwestern [University, in Evanston, Ill.], and Tony has a wonderful position here, and we're going to be commuting. We're not sure how long this is going to go on, but we're going to give it a try.
"The biological clock," continues Villa-Komaroff, "I think that's been the most difficult for us. In our case, it's meant that we have opted not to have children of our own."
"To me the most surprising thing was that the obstacles against success in careers-particularly for women, but not only for women-were small in themselves in effect, but large in numbers," observes Holton. "So what really counted was that there wasn't-except in those extreme wings-one great stroke of ill luck or good luck or ill treatment. What happened is that a small set of misfortunes or disadvantages throughout the career accumulated in the same direction, so as to deflect the women in one direction."
A scientist whom Sonnert and Holton refer to as "Irene" in Who Succeeds in Science is a case in point. "While Irene's husband went from good jobs to better jobs and ended up as a full professor," they write, "Irene had a sequence of less-elevated jobs, was denied tenure, and occupied increasingly marginal science positions until she had to drop out completely." Irene now holds an administrative position in a university.
Linda S. Wilson, president of Radcliffe College, and a former research chemist-who wrote the preface to Who Succeeds in Science-agrees that a series of small events can have a large effect: "A lot of the discrimination women experience is in the form of micro-inequities as opposed to blatant obstacles or bad behavior. An awful lot is an accumulation of little things, no one of which is by itself worth going to the mat about, but the sum total becomes an obstacle you can't get over."
According to Holton, the influence of such "luck," as chronicled in the career histories of his subjects, indicates that scientists should search for serendipity as much in their careers as in their science. "Very few people, men or women, know how to recognize a lucky break. The idea that luck plays a role rather than your virtue and your long hours and your long training, that one should be on the lookout for lucky breaks and take advantage of them, it sounds almost against the Puritan ethic."
Sonnert and Holton examine the effects of luck in scientific career paths with reference to a "kick-reaction model" (J.R. Cole and B. Singer, "A theory of limited differences: Explaining the productivity puzzle in science" in H. Zuckerman et al., eds., The Outer Circle: Women in the Scientific Community, New York, W.W. Norton and Co. Inc., 1991). "A kick is any event in the environment that has a potential effect on the individual's career, be it positive or negative," write Sonnert and Holton. "Likewise, the individual's reaction to a kick can be positive or negative. Over the course of a career, the pattern of kicks and reactions changes." Notes Sonnert, "Negative or positive kicks can be subtle. Several women told stories about how some important decisions are made at a very informal level, maybe not even in the office, but after hours in a bar. And these were things they might not get invited to or might not feel comfortable with. So they would miss out on potential good kicks-that is, being involved in the decision-making."
"Clearly, there's no question that the informal channels in science are important," says Mary Good. But, she says, women have it within themselves to remedy this disparity. "The men will invite themselves or figure out ways to be included. For the most part, women have a difficult problem doing that. My view on that is that women simply have to take the initiative to be sure they are included."
Project Access also documented clear indications of a glass ceiling for women in science, at least in some fields. In the physical sciences, mathematics, and engineering (PSME), women hold lower academic rank than men who received their doctorates at the same time. And, women who received their Ph.D.'s before 1978 are significantly less likely than men to have achieved the rank of full professor both in PSME and in the social sciences (see accompanying table).
But women in the biological sciences seem to be faring somewhat better, which Sonnert and Holton attribute to a "critical-mass" effect. According to this idea, once the percentage of women in a field reaches a certain level, men's and women's career outcomes start to equalize. However, says Sonnert, "I would expect that there is a gender differential in biology. I don't think our data can be used to say, well, the problem has gone away in biology because our biologists did equally well whether they were men or women."
MOVING ON UP: NRC's Charlotte Kuh studies career development. But NRC's soon-to-be-released longitudinal study may not support a critical-mass effect. Charlotte V. Kuh, executive director of NRC's Office of Scientific and Engineering Personnel, observes that although there are very few women in engineering, they advance in their careers as fast as men. "If you were looking for a critical-mass effect, what you would see is that the numbers of women increased and then outcomes became more equal," she states.
Although veteran observers of women in science queried by The Scientist generally react favorably to Sonnert and Holton's data, interpretations, and list of recommendations for policymakers and aspiring scientists, there has been criticism. Mary Good, for example, remarks that the study's definition of success-a research career in academia-seems unduly restrictive.
As far as her field goes, says Good, "In chemistry, 80 percent of Ph.D.'s go to industry and always have. So to define success as only a research position in a prestigious university doesn't meet my definition at all. It would be interesting to see how many of them became V.P.'s for research."
"We don't presume that people who go into industry do less important work or less successful work," replies Sonnert. "What we faced was a practical problem more than anything, because we were not able to measure success for industrial scientists. The field is so diverse that there is no yardstick, really."
He points out that scientists in industry tend to publish less than academic scientists, for example, and their titles are more diverse-and much harder to quantify-than the academic scale ranging from postdoc to full professor.
Despite persistent gender disparities, Holton believes that things are improving for more recent women Ph.D.'s. Formal barriers to women in science have fallen, and women now are reaching critical mass in more and more fields. "Compared to 50 years ago,and even compared to the women who got their degrees and their fellowships before 1970, it is as night and day. Neither of us wants to say these are the good times for women in science, but they certainly are better than even a generation or two back."
Robert Finn, a freelance science writer based in Long Beach, Calif., is online email@example.com.
- (The Scientist, Vol:9, #22, pg.3, 9 , November 13, 1995)
- (Copyright © The Scientist, Inc.)