|
|
|
Women in Science
|
In 2005, I was asked to give a lecture on the subject “Women
in Science” as part of the “Postgraduate Lecture Series” at the Centenary
Institute in Sydney. This series used to be strictly scientific, but has more
recently been organised by an enthusiastic but somewhat mischievous medical
graduate now doing a PhD. He decided to make the series more interesting by
making the topics more controversial. The Institute Director would be asked to
pontificate on “Great Scientific Discoveries of the Past 500 Years” or “Medical
Progress in the 20th Century”, and colleagues would attempt to explain topics
such as “How to Make Money in Science”. In 2004, I was asked to talk about
“Immunological Controversies”, which proceeded without too much of the same.
However in 2005, I could see “Women in Science” degenerating into an unpleasant
shouting match, and it was with considerable trepidation that I agreed to take
the subject on.
Upon Googling the topic (how else to start preparing a lecture on a general
subject these days?) I discovered a number of sites that dealt with famous women
in science — usually starting with Marie Curie. These were generally lovely
homilies about how brilliant women had made it through despite prejudice (which,
the articles implied, would not be operating in our more enlightened times). But
I felt that this approach would not be of much help to the women graduate
students at the Centenary Institute. After all, who regards themselves as being
in the same league as Marie Curie? If we have problems in our careers, they
could as easily be attributed to a lack of brilliance as to prejudice.
I also rather suspected that I was invited to give this lecture so that I could
complain, as I have learned through bitter experience never to do in mixed
company, of instances when my gender was the target of overt, or far more often
covert, discrimination. I therefore dodged the issue by treating the lecture as
a purely factual exposition of the position of women in science, with a
diversion into some possible contributing factors, garnered from published
research in social psychology. So what follows is my attempt to describe the
position of women in science today, summarised from this one-hour lecture. I
have concentrated on biology and more particularly on immunology, my field of
scientific study, and I have tried to find Australian data to compare to US and
European data, although I would be the first to admit that this is by no means a
statistically rigorous presentation. Most of the data were taken from web sites,
which I have tried to acknowledge conscientiously, and I checked some but by no
means all the primary papers to which these sites refer.
|
|
|
A/Prof. Barbara Fazekas de St Groth, from the Centenary Institute of Cancer Medicine and Cell Biology. |
The National Science Foundation of the USA publishes figures showing the number
of bachelor’s degrees awarded to women and men in science and engineering each
year since 1966. The number of men awarded degrees has remained fairly constant
since the mid ‘70s, whereas the number of women awarded degrees has doubled, and
equalled that of men for the first time in 2001. Twice as many doctoral degrees
in science and engineering were awarded to men as women in 2001, but once again,
the curve for males had been relatively fl at for many years, whereas the curve
for women has moved steadily upwards. If no other factors supervened, one would
expect that the number of women moving into more senior positions in science and
engineering would also be steadily increasing, and that is true for 5 or so
years after the award of a doctorate. After that, the women start to disappear.
Those who stay take an average of 3 extra years to reach full professorship, and
face an increasingly large gap in salary — up to 25% less than that of male
colleagues of equivalent rank by age 60. On every measure — academic rank, years
taken to achieve that rank, salary— women consistently lag behind men.
Nonetheless, the US does relatively well in employing women in top-ranking
university positions, with 17% of these positions being occupied by women. In
the UK and Germany, these figures are 7% and 4%, respectively, although France
and Italy do better, with 16% and 15% of top university positions being occupied
by
women. Even in Sweden, a supposed paradise in terms of social equality, only 10%
of top university positions are held by women. In EU countries, approximately
50% of university undergraduates are female, but their representation drops
progressively to 40% of PhD students, 30% of assistant professors, 20% of
associate professors and about 10% of full professors.
The American Association of Immunologists (AAI) has run a Women’s Committee for many years. They maintain a database of women in immunology, so that
|
meeting
organisers can be reminded about candidates for invitation, and journal editors
can look for associate editors. (Unfortunately, they recently decided to
restrict this service to members). The AAI has collected figures on the
representation of women in US Immunology Departments. The proportions vary from
30% at Stanford and Berkeley, 17% at Harvard, to 5% at the University of
Minnesota. The proportion of female graduate students is between 45% and 60% at
all institutions for which figures are available. The Australasian Society for
Immunology (ASI) does not keep similar figures, nor is there a Women’s
Committee. However by looking at membership lists, I was able to estimate that
PhD students and research assistants are 70% female, post-docs are about 45%
female, and full professors are 14% female. At the Centenary Institute, 69% of
our students, 36% of postdocs and 11% of senior scientists are female (that’s me
in the latter category, the one female laboratory head for the past 12 years).
At recent US Keystone meetings in the field of immunology, 27% of plenary
speakers were female. At our own ASI meetings, which rotate around the capital
cities in a 7 year cycle, we have averaged 19% female plenary speakers over the
past 10 years, but some cities stand out from the mean. Melbourne, which has the
largest number of ASI members of any capital city in Australia, has managed just
one female plenary speaker in the last 3 ASI meetings, of a total of 39 plenary
speakers. Adelaide, on the other hand, managed 7 female plenary speakers out of
30. There has been a concerted effort to increase the number of women on the
editorial boards of international immunology journals, partly in response to
lobbying from the AAI Women’s Committee. The average is now about 20%, up from
10% in 1994. (In that time, the number on the editorial board of Immunology and
Cell Biology, the immunology journal run by ASI, has decreased.)
One factor that is often quoted as explaining women’s under-representation in
the higher echelons of science is that there is a pipeline from graduation to
retirement, and it takes many years for increased representation in the junior
ranks to flow through to more senior levels. However, the representation of
women in senior positions has been quite stable for the past 10 years, despite
the increased availability of junior women to be promoted into those positions,
so the pipeline is clearly very leaky. There is also a perception, often voiced
when women’s representation in science is discussed, that women “choose not to
compete”. However, this could just as easily become an excuse for institutions
to choose not to promote them. What are the factors that prevent women from
reaching senior positions at the same rate as men? Clearly the responsibilities
of childbearing and childcare are a major factor, as they are in the rest of the
workforce. In science, women with children advance more slowly than those
without, whereas the opposite is true for men. But even women without children
advance more slowly than men. I don’t propose to go into the obvious need for
more social support for working women with children in this context. This is
part of a wider social problem in the distribution of work, which seems to be
getting less rather than more equitable as the professions work harder.
Overriding this whole picture is the question how society treats men and women
differently, and how this plays out in the world of science. In the field of
social psychology, these patterns of human behaviour are referred to as “Gender
Schemas”, and I am indebted to the website of Virginia Valian at Hunter College
NY (1) for much of the discussion that follows. Gender schemas (also referred to
as gender stereotypes) describe how the differences in behaviour between men and
women are self-perpetuating and self-reinforcing. These schemas are unconscious,
and operate independently of good intentions, so female colleagues will be no
more conscious of this form of gender-based discrimination than men.
|
Typical experiments supporting the existence of gender schemas have made use of
self-assessment of the monetary value of work performed by college students, in
the absence of any outside input. Male self-assessments can be more than 50%
higher than the self-assessments of females, and when asked to perform what they
judge to be a reasonable amount of work for a given amount of money, males
typically perform less work, and perform this less carefully. Males and females
also prefer to see men in leadership roles. In one particular study (2) college
students were shown slides displaying 5 people seated around a table and asked
to identify the leader. In same gender groups, the person at the table head was
always identified. In mixed gender groups, the man at the head was always
identified as the leader, but if a woman was at the head, a man seated elsewhere
was labelled the leader about 50% of the time. The gender of the observer made
no difference to the judgements made. In another study (3), two trained actors
followed a script in which one adopted a “friendly but assertive” leadership
role. The leader could be either male or female. Naive observers, both male and
female, gave more positive than negative facial expressions to men playing the
leadership role, and the reverse was true for women playing the same role.
Virginia Valian gives another example on her website (1) to which many women can
probably relate. This is where a woman makes a suggestion, which is ignored, as
if she hadn’t spoken. Five minutes later, a man makes the same suggestion, to
universal applause. If the woman mentions what has happened, her colleague, with
the best intentions in the world, says ‘Don’t make a mountain out of a
molehill’. But as Virginia Valian goes on to say, “…many molehills turn into
mountains, and small imbalances add up to disadvantage women. Success is largely
the accumulation of advantage, exploiting small gains to get bigger ones.”
Gender schemas start to operate early. In studies of perception of maths ability
in American junior high school students, taken at an age when there is no
substantial gender difference in high school maths performance tests, both boys
and girls agree that boys’ skills are higher. Self-evaluation is lower, relative
to performance, for girls, whereas the reverse is true for boys. For selecting a
career path, high school girls are more influenced by marks and self-evaluated
ability, the advice of parents and teachers, job opportunity, desire for career
flexibility and the desire to make the world a better place. In contrast, boys
are more influenced by the perceived status and remuneration of the chosen job.
Perceptions can also influence performance. When females taking an undergraduate
maths test are told that females typically score lower than males, they score
lower. When they are told that there are no gender differences, they score the
same as males (4).
Do these perceptions carry over into science, where ranking of scientific
articles and CV’s is performed “without bias”? Unfortunately, they do. In one
study, first done in 1968 and then replicated in 1983 (5), college students were
asked to rate identical scientific articles according to specific criteria. The
authors’ names were clearly male or female, but were reversed for each group of
critics. Articles supposedly written by women were consistently ranked lower
than when these same articles were thought to have been written by a male. In a
similar study, department chairs were asked to make hypothetical hiring
decisions and to assign faculty rank on the basis of CV’s. For CV’s with male
names, chairs recommended the rank of associate professor. However, the
identical CV with a female name merited only the rank of assistant professor.
|
In 1997, these gender schemas came to the attention of readers of the journal
Nature when it published a study by Christine Wennerås and Agnes Wold from
Goteborg University in Sweden (6). The authors of the study found clear gender
bias in the way in which postdoctoral research awards were made by the Swedish
Medical Research Council. Women needed to be about 2.2 times more productive
than male counterparts to be as successful in securing financial support. How
could a selection committee that prided itself on fairness make such an obvious
mistake? The committee decisions were based not only on published work, but by
reference to factors that already discriminated against women, namely the
recommendations of heads of department, and institutions. A subsequent UK study
showed no statistically significant bias in awards, although they noted that
women published slightly fewer papers, but had a higher impact per paper and
overall. Nonetheless, there was a very significant difference in the percentages
of men and women who applied for awards compared to what would be predicted on
the basis of academic position (7).
In the US in 1994, 15 women with tenure in the School of Science at MIT decided
to obtain quantitative data to either support or
refute anecdotal stories of gender bias. Those 15 women represented 8% of
tenured faculty. In response to their request, the Dean of Science appointed a
committee made up of one tenured woman from each department, plus three tenured
men who were current or former department chairs. In their landmark report,
released in 1999 and endorsed by the President of MIT, inequalities were found
in salaries, pensions and allocation of space and resources (8). When
interviewed, junior women said they felt supported by their departments. Their
most common concern was the extraordinary difficulty of combining family and
work. However as women progressed through their careers, they became
increasingly marginalised and excluded from positions of real power in their
department. Each generation of young women, including those who were currently
senior faculty, began by believing that gender discrimination was ‘solved’.
However, gradually their eyes were opened to the realization that the playing
field was not level at all. The study concluded that bias is subtle and largely
unconscious and that the most lasting correction will come only when the numbers
of women have grown significantly (8).
Sometimes bias is not so subtle. At Harvard recently, the President of the
University, Lawrence Summers, suggested that women fail to advance in science
because they are innately less able than men. Nancy Hopkins, an MIT professor,
brought Summers’ comments to public attention, and the discussion once again
spilled over into the pages of Nature, as it had done after the Swedish Medical
Council Study more than 8 years earlier. Interestingly, two male professors of
physics from Harvard have been in the forefront of studies highlighting gender
differences in the way in which scientists perform. Gerald Holton’s “Project
Access” (9) studied a group of 700 male and female scientists, who had all shown
great promise early in their careers. By selecting an elite group, the
investigators hoped to highlight gender differences that would be even greater
in more “average” scientists. Over the course of the study, 88% of men attained
tenure, compared with 40% of women. However, the women produced an average of
2.3 publications per year, compared with 2.8 publications per year for men.
Senior biologists from two prestigious research universities were then selected
to review the dossiers of biologists selected from the Project Access pool. The
reviewers were told to assign quality ratings on the basis of CV, the
bibliographies and reprints of six articles or chapters that each individual
thought represented his or her best work. They used a scale of from 1 to a top
grade of 5, similar to that of many granting agencies. The average quality
rating given by the evaluators was 3.67 for women versus 3.27 for men. Moreover,
in the annual Institute for Scientific Information (ISI) Science Journal
Citation Reports, the women’s articles had received substantially more citations
— 24.4 citations per article, compared with 14.4 for men. “With all due caution
about the use and abuse of citations in the literature, at the very least one
can say that overall the women’s papers had greater visibility, and, one may
infer, were found to be of value to the field” said Holton.
However, the women themselves saw things differently. They reported that they
considered themselves as being average almost twice as often (35% of women
versus 18% of men), and men reported themselves more often as being above
average (70% of men versus 32% of women). There were also differences in
perceptions of gender differences in terms of how women work as scientists.
Whether or not these differences are real, twice as many women believed that
they exist (51% of women versus 26% of men). The same proportions held true when
women were asked whether gender affected their choice of a research subject or
their way of thinking in science. When asked whether gender differences exist in
terms of the methods adopted to pursue a scientific project, women answered
affirmatively more than three times as often as men. Among the men and women
interviewed, a very frequent observation was that male scientists are more
aggressive, combative, and self-promoting on the road to career success. The
women were more likely to emphasize that one of the best things about research
was the intellectually stimulating process. Holton believes that women
scientists uphold, to a statistically significant degree, what might be called
the traditional standards of good science, namely the pursuit of fundamentals,
with care, objectivity, and reproducibility. When questioned, women reported
themselves as more cautious and careful in their method, and more attentive to
details on the way to a conclusion. Numerous women acknowledged that they were
tending towards perfectionism, not only to avoid failure and criticism, but to
seek a broader, more comprehensive picture, and to produce more complete and
synthetic papers.
“One way to escape some of the rough-and-tumble competition of life at science’s
frontier is to choose …a niche where, with luck, one can work on a problem that
is important and yet not at the center of those volcanic eruptions,” said
Holton. And that is just what women have often done. As Marie Curie said when
she was asked why she decided to work on what was later called radioactivity: “I
chose this field because there was no bibliography.” This approach may also shed
light on the finding that while in Graduate School, women are more
collaborative, but after the postdoc period, they become less collaborative than
men. Women are more vocal about valuing broad or comprehensive projects that are
also characterized by integrity and thoroughness, rather than cutting by corners
to gain career advantage. These self-reports and interviews potentially explain
the productivity gap between women and men, as well as the higher citation rates
of women’s publications.
|
Another champion of women in science is Howard Georgi, also a Professor of
Physics at Harvard. He has written articles about how selection procedures for
academic positions tend to also select for assertiveness and single-mindedness.
Not only are the distributions of assertiveness and single-mindedness skewed
towards men, but there are also very strong cultural biases that make it more
difficult for women to be assertive and single-minded. We are all familiar with
these hurtful stereotypes: Mr X is assertive, Ms Y is pushy, Mr X thinks
creatively, Ms Y is easily distracted. As Prof. Georgi says “Just because we
have a system that produces good scientists does not mean that the system is not
eliminating many others who could be equally good”.
So what can we do about this? So far, the only cases where women’s
representation at management level has changed dramatically come from companies
which can actively manage their workforce. One example is Motorola, who made a
corporate decision to increase the percentage of women in senior positions.
There were complaints from men in the beginning, but after the percentage of
women in management reached 40%, even the men noticed that the workplace was
more pleasant and productive. Unfortunately, it seems highly unlikely that the
representation of women at the more senior levels of science will reach 40%
within our lifetimes, if ever. The issue of women’s career advancement seems to
have slid off the agenda in Australia. Certainly, raising the issue of
advancement immediately brings cries of unfairness and positive discrimination.
In the US, things are somewhat different. In 2004, the NIH instituted a new
award — the NIH Director’s Pioneer Award for novel biomedical research
strategies. When all 9 awards went to men, there was a public outcry (10). The
sole female member of the selection committee protested that they had made their
choice in an unbiased way, but that few applications had been received from
women. In 2005, responsibility for the award was transferred to the National
Institute of General Medical Sciences, which has more experience with grants and
was prepared to redesign the program. Some 40% of referees are now female, up
from 5% last year. Instead of requiring a nomination, which automatically biases
against women, applicants were allowed to nominate themselves. The application
form was also reworded to emphasize that women and minorities were especially
encouraged to apply, and that the award was available to early- and mid-career
scientists. In 2005, 26% of applicants and finalists were female, up from 21% of
applicants and only 10% of finalists in 2004.
I should end this survey with a message of optimism and hope, but I have to say
that I don’t believe that things are improving in Australia. This may be a
reflection of where I am in my career — having survived the struggle of a young
family, I have come out the other side breathing a sigh of relief, but have now
found further difficulties of which I was previously unaware. Being the only
senior woman in my institution does not help. There seems to be general doubt
that there is a gender equity problem, and it is certainly not on the political
agenda (or not to the extent that it was when I started my PhD 25 years ago). I
think the problem has to be addressed at an institutional level, as at Motorola,
but academia is bound by beliefs and traditions that do not sit easily with the
sort of workforce management possible in a commercial organization. The concepts
of merit and academic freedom are diametrically opposed to measures that take
gender into account, which would not be a problem if these concepts were not
gender-biased. But as mentioned above, the definitions of merit and academic
freedom as usually applied, are unfortunately covertly gender-biased. We need
more understanding and acceptance of gender schemas and their effect on our
behaviour, as well as the means to make the practices of institutions more
accountable in terms of gender equity. Providing positive incentives for
institutions or departments that actively pursue gender equity would be a good
start.
Some of the areas in which the undervaluing of women’s ability negatively
affects their careers are in appointment, promotion, awarding of grants,
invitations to present at meetings and publications. In the first four areas at
least, there is an opportunity to do something at a local level. As mentioned
above, the AAI and many other US-based societies have Women’s Committees that
provide publicly available lists of women scientists and their areas of
expertise. In 1993, I chaired the organizing committee for a joint meeting of
our own society, ASI, with the US-based Society of Leukocyte Biology. The
provisional program came back from my US counterpart with a comment that the
percentage of female plenary speakers was far too low — couldn’t we find more
women? This was a concept that was so far off the radar here that it had not
even been mentioned in our own meetings about the conference program.
One of our difficulties here in Australia is that there are also few senior
women scientists to serve on numerous appointment, promotion, grant and meeting
organising committees. It is hard for senior women to cover all these bases and
still survive in their own careers. In addition, if senior women appointed to
these committees are no more aware of gender schemas than their male colleagues,
they will interpret the concepts of merit in exactly the same way as men. I
don’t think we will make significant progress until both men and women
acknowledge that the system does discriminate against women, and that for
reasons not only of equity but also of efficiency, productivity and support for
innovation, it would be better if the system did not discriminate. I hope that
Wisenet can achieve something in this area, at least by providing the moral
support and encouragement that we all need but often don’t get from our
colleagues. I also hope that this article helps to stimulate debate, and that we
can all come up with useful strategies to improve the situations of women in
science in Australia.
References
(1)
http://www.hunter.cuny.edu/genderequity/;
http://www.hunter.cuny.edu/gendertutorial/.
(2) Porter, N., & F. Geis. 1981. Women and nonverbal leadership cues: When
seeing is not believing. In: Gender and Nonverbal Behavior (C. Mayo & N. Henley,
eds.), Springer-Verlag, New York, NY, pp. 39–61.
(3) Butler, D., & Geis, F. L. (1990). Nonverbal affect responses to male and
female leaders: Implications for leadership evaluations. Journal of Personality
and Social Psychology, 58, 48–59.
(4) Spencer, S. J., & Steele, C. M. (1994). Under suspicion of inability:
Stereotype vulnerability and women’s math performance. Unpublished manuscript.
SUNY Buffalo and Stanford University.
(5) Sandler, Bernice R., with the assistance of Roberta M. Hall. ``The Campus
Climate Revisited: Chilly for Women Faculty, Administrators, and Graduate
Students’’. Copyright 1986 by the Project on the Status and Education of Women,
Association of American Colleges, Washington, DC, 1986.
(6) Wennerås, C., & Wold, A. (1997) Nepotism and sexism in peer-review. Nature,
387, 341–343.
(7) Grant, J., Burden, S. & Breen, G. (1997). No evidence of sexism in peer
review. Nature, 390, 438.
(8) http://web.mit.edu/fnl/women/women.html
(9) Sonnert, G. & G. Holton. (1996). Career patterns of women and men in the
sciences. American Scientist, 84, 63–71.
(10) Mervis, J. (2004). Male sweep of new award raises questions of bias.
Science, 306, 595.
Bio: Associate Professor Barbara Fazekas de St Groth MBBS, BSc(Med), PhD is an NHMRC Principal Research Fellow, and Head of the T Cell Biology Research Program at the Centenary Institute of Cancer Medicine and Cell Biology in Sydney. She also has two sons aged 15 and 12, and could never have survived in science without the constant support of her husband, who works part time.
