ArticlePDF Available

Athena Unbound: The Advancement of Women


Abstract and Figures

Why are there still so few women scientists, especially at the upper levels of the scientific professions? Persisting differences between women's and men's experience in science make this question as relevant today as when sociologist Alice Rossi posed it more than three decades ago at a conference on women in science at the Massachusetts Institute of Technology (Rossi, 1965).
Content may be subject to copyright.
Henry Etzkowitz
Carol Kemelgor
Brian Uzzi
Athena Unbound
The Advancement of Women in Science and Technology
Why are there so few women scientists? Persisting differences between
women’s and men’s experiences in science make this question as relevant
today as it ever was. This book sets out to answer this question, and to
propose solutions for the future.
Based on extensive research, it emphasizes that science is an intensely
social activity. Despite the scientific ethos of universalism and inclusion,
scientists and their institutions are not immune to the prejudices of
society as a whole. By presenting women’s experiences at all key career
stages – from childhood to retirement – the authors reveal the hidden
barriers, subtle exclusions and unwritten rules of the scientific workplace,
and the effects, both professional and personal, that these have on the
female scientist.
This important book should be read by all scientists – both male and
female – and sociologists, as well as women thinking of embarking on a
scientific career.
henry etzkowitz is Director of the Science Policy Institute and
Associate Professor of Sociology at the State University of New York at
carol kemelgor is a psychotherapist and psychoanalyst in private
practice in Westchester County, New York, and Director of the Center for
Women in Science, at the Science Policy Institute, State University of
New York at Purchase.
brian uzzi is Associate Professor of Business and Sociology at the
Kellogg Graduate School of Management at Northwestern University.
To my mother
BA Hunter College 1933
Magna Cum Laude, Geology
the advancement of women
in science and technology
Science Policy Institute,
State University of New York at Purchase
Science Policy Institute,
State University of New York at Purchase
Kellogg School, Northwestern University
michael neuschatz, American Institute of Physics
elaine seymour, University of Colorado
lynn mulkey,University of South Carolina
joseph alonzo, Rockefeller University
The Pitt Building, Trumpington Street, Cambridge CB2 IRP
40 West 20th Street, New York, NY 10011-4211, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
© Henry Etzkowitz 2000
This edition © Henry Etzkowitz 2003
First published in printed format 2000
A catalogue record for the original printed book is available
from the British Library and from the Library of Congress
Original ISBN 0 521 56380 1 hardback
Original ISBN 0 521 78738 6 paperback
ISBN 0 511 00944 5 virtual (netLibrary Edition)
Introduction: Women in science: Why so few? 1
1The science career pipeline 5
2Women and science: Athena Bound 15
3Gender, sex and science 31
4Selective access 49
5Critical transitions in the graduate and post-graduate
career path 69
6Women’s (and men’s) graduate experience in science 83
7The paradox of critical mass for women in science 105
8The ‘kula ring’ of scientific success 115
9Women’s faculty experience 131
10 Dual male and female worlds of science137
11 Differences between women in science 147
12 Social capital and faculty network relationships 157
13 Negative and positive departmental cultures 179
14 Initiatives for departmental change 187
15 International comparisons 203
16 Athena Unbound: Policy for women in science 225
Appendix 251
Bibliography 257
Index 269
We express our appreciation to the National Science Foundation and
the Sloan Foundation for research support.
Women in science: Why so few?
Why are there still so few women scientists, especially at the upper
levels of the scientific professions? Persisting differences between
women’s and men’s experience in science make this question as
relevant today as when sociologist Alice Rossi posed it more than three
decades ago at a conference on women in science at the Massachusetts
Institute of Technology (Rossi, 1965).
The years since Rossi’s groundbreaking analysis have witnessed the
revival of the feminist movement and the increased entry of women
into many professions. Women have become lawyers and doctors
in significant numbers, albeit unevenly distributed into high and low
status subfields of these professions. Despite significant advances,
there is a continuing disproportionate lack of women in most scientific
and engineering disciplines, especially at the upper reaches of the
One such scientist, Leslie Barber, a female Ph.D. in molecular
biology, decided to end her career as a research scientist shortly after
being awarded the doctorate. She reflected upon the mixed experience
of her male and female peers in a recent article (Barber, 1995). On the
positive side, she found widespread evidence of encouragement for
girls and women to pursue scientific professions from the media and
from parents and teachers.
On the negative side, in comparing the career trajectories of the ten
members of her graduate research group, equally divided into five men
and five women, Barber noted significant differences. Whether or not
the men had done well in their graduate careers, they had forged ahead
in their professional lives. Among the women, three ‘have left research
altogether, while the other two languish in post-doctoral positions,
apparently unable to settle on a next step.’ Barber was initially
surprised that, despite the unique story that each woman offered to
explain her situation, the traditional pattern of relative exclusion of
females from the scientific professions had been reproduced in her
graduate cohort.
A guarded professional prognosis for both men and women could
well be advised for a field such as physics, where the potential numbers
of qualified applicants, vastly overwhelm traditional occupational
demand (Linowitz, 1996). Certainly there has been a shift away from
nuclear weapons and power plants, as well as from ‘big science’
projects such as the cancelled Superconducting Super Collider, which
once gave virtually automatic multiple choices of employment to
Ph.D. physicists. Although not unemployed, young physicists can
often be found utilizing their quantitative and analytical skills in the
back rooms of Wall Street or even in their own financial firms.
But how can the male–female divide in following scientific research
careers, as identified by Barber, be explained for molecular biology,
given the proliferation of biotechnology firms with research positions
in recent years? Why has the increase in women entering graduate
school not been fully translated into female scientists occupying
higher positions in the field? Why has science lagged other professions
in its inclusion of women? The answers to these questions, and the
responsibility for repairing a less than optimal outcome, can be found
primarily within science and secondarily in the larger society
(National Research Council, 1940; Fox, 1994).
The thesis of this book is that women face a special series of
gender related barriers to entry and success in scientific careers
that persist, despite recent advances. Indeed, while some of their
male contemporaries view female scientists as ‘honorary men’, others
see them as ‘flawed women’ for attempting to participate in a
traditional male realm (Longino, 1987; Stolte-Heiskanen, 1987;
Barinaga, 1993).
Female scientists have been at odds over how to respond to these
invidious distinctions. Should they insist that as scientists they are
not different from men? On the other hand, given that science has
historically been a male-dominated profession, should not women
claim that they must have their needs taken into account in how the
field is organized?
We focus the greater part of this book on the quality of women’s
experience in academic science, on the grounds that the university
serves as a gateway into the larger scientific community. Our analysis
is based on extensive systematic fieldwork that focuses both on the
personal accounts of female and male graduate students and faculty
members, and on the statistical analysis of aggregate demographic data
and survey data on person-to-person ties in departments. In interviews
with us, they discussed their experience in research groups and
departments as well as their interaction with male and female peers
and mentors.
Athena Unbound provides a life-course analysis of women in
science from early childhood interest, through university, graduate
school and the academic workplace. The book is based on several
studies: (1) fifty in-depth interviews with female graduate students and
faculty members in five science and engineering disciplines at two
universities; (2) four hundred in-depth interviews and focus groups
with female and male graduate students and faculty members in five
science and engineering disciplines at eleven universities; (3) follow-
up interviews with a sub-sample of graduate students and post-
doctoral fellows interviewed in the previous study; (4) a quantitative
survey of female graduate students and faculty members in five science
and engineering disciplines at one university, focusing on publication
experiences; and (5) interviews with very young children on their
image of the scientist as a gender-related role.
In the following chapter we will begin to address the question raised
in this introduction: why so few women in science? We will present
quantitative evidence documenting how women’s entry into and
leakage from the ranks of graduate school education and university
departments differ from men’s. As society becomes more knowledge-
intensive, ending any exclusion of women from science and tech-
nology becomes more pressing.
1The science career pipeline
In this chapter we discuss the ‘pipeline’ thesis for improving women’s
participation in science. This ‘supply side’ approach assumes that if
sufficient women are encouraged to enter the scientific and
engineering professions, the gender gap in science and technology will
The scientific career track, from elementary school to initial
employment, has been depicted as a ‘pipeline’ like those for the
transport of fluids and gases such as water, oil or natural gas. The rate of
flow into scientific careers is measured by passage through transition
points in the pipeline such as graduation and continuation to the next
educational level.
Nevertheless, the flow of women into science is through, ‘a pipe
with leaks at every joint along its span, a pipe that begins with a high-
pressure surge of young women at the source – a roiling Amazon of
smart graduate students – and ends at the spigot with a trickle of
women prominent enough to be deans or department heads at major
universities or to win such honours as membership in the National
Academy of Science or even, heaven forfend, the Nobel Prize’ (Angier,
1995). Even this negative depiction of the pipeline as a leaky vessel is
too optimistic. As we shall see, many women are discouraged from
pursuing their scientific interests far earlier in their educational career
than graduate training.
Although the rate of women entering scientific professions has
improved significantly, especially in the biological sciences, the
numbers reaching high-level positions are much smaller than
expected. In the United States, for example, decades after the science-
based profession of medicine experienced a significant increase in
female medical students (currently about 40% are women), only 3% of
medical school deans and 5% of department heads are women. Dr
Eleanor Shore, dean for faculty affairs at Harvard Medical School,
recalled, ‘Originally we thought if we got enough women in, the
problem would take care of itself’ (Angier, 1995). But it obviously has
Significant numbers of women enter the ‘pipeline’ and then leave at
disproportionate rates, or function less effectively, as covert resistance
to their participation creates difficulties. At best, the picture of
women’s participation in science in recent decades is mixed. Indeed,
the pipeline analogy is unintentionally appropriate as an implicit
criticism of the way that the recruitment to science takes place.
In addition to the positive meaning of steady flow and assured
delivery, a pipeline also connotes a narrow, constricted vessel with few
if any alternative ways of passage through the channel. At each age
grade, the entry ways for women become narrower and increasingly
restrictive. As more are excluded, the talent pool for the next level to
draw upon becomes smaller.
Although the genders are almost equally represented in the early
stages of the pipeline they increasingly diverge at the later stages,
resulting in a much smaller proportion of women than men emerging
from the pipeline. At the point of career choice, many women are
diverted from the academic and research tracks, even though some
who are trained as scientists pursue science-related careers such as
scientific writing or administration. The U.S. science pipeline runs
through a distinctly different educational landscape than its
counterparts in many other countries, and it is worth taking a moment
to describe the system here.
In contrast to most European, Latin American and other countries
where a specialized course of study on one or a few related areas makes
up virtually the entire undergraduate curriculum, the U.S. educational
system does not expect students to make an early choice of careers.
Even though an increasing number of secondary and even middle
schools have occupational themes such as healthcare, art and science,
all offer a general education. The flexibility of the U.S. undergraduate
degree allows room for secondary education to remain unspecialized.
Students typically graduate from high school after twelve years of
primary and secondary education at the age of seventeen or eighteen.
Where to go to college or university becomes a serious issue in the third
year of high school, although student and parental anxieties about
getting into a prestigious college or university have pushed these
concerns ever earlier. Again in contrast to countries with national
systems of examinations at secondary school leaving, the U.S. high
school offers an education that can vary widely in quality among
schools and even within the same school. High school is still the
quintessential U.S. social scene depicted in television programs and
motion pictures of a youth culture focused on peer status, looks and
athletic ability. Intellectual merit is not a leading status distinction
except in a very few leading public and private high schools.
Universities also vary widely in quality of education and prestige, in
contrast to Europe where university-level institutions are, more or
less, expected to be on the same level. There is also a tradition in the
U.S. of students going to university away from home, if it can be
afforded. This makes the college decision a major turning point in
life. It also marks the entry of the student into a nationwide
educational and prestige gradation market. To take account of the
wide differences in quality among secondary schools, an external
system of exams offered by a non-profit corporation rather than a
government agency was established in order to help universities sort
potential students from a wide variety of backgrounds. Once
university intake broadened from a select set of students attending
college preparatory public and private high schools, as had been the
case in the 1920s, to a mass education system, uniform measures were
needed and the College Board examinations were established for this
The College Board examinations focus on general abilities in
mathematical and analytical reasoning and are not directly tied to the
high school curriculum. Therefore, a separate educational industry has
grown up offering courses and tutoring to prepare students for these
examinations, whose sponsors persist in insisting that formal
preparation will do no good. Through these exams, high school grades,
recommendations and sometimes an essay to be written on ‘life goals’,
‘the most influential book I have read’ or some such topic, combined
with interviews by an alumnus or a college admissions officer, an
initial selection is made.
High school graduates are sorted into more than 3,000 institutions of
higher education, ranging from four-year baccalaureate colleges to
universities offering Ph.D. degrees. However, this selection is still
malleable since college students increasingly take time off from their
studies to travel or work for a while and then decide to apply for
Almost 70% of U.S. high school graduates now continue on to post-
secondary education. This is still in sharp contrast to the U.K. which
has only in the past decade seen a rise from 10% to 30%, with an
expected rise to 40% of secondary school leavers continuing on to
university during the next decade.
In the U.S., general education continues from high school into the
university. ‘Distribution requirements’ insure that students take one
or more courses in the various spheres of knowledge such as science,
art, history, languages and mathematics. In addition, many colleges
and universities require students to take certain courses, typically in
writing and the history of western civilization, as part of a general
education program. In other countries such broad knowledge and skills
are expected to be acquired in secondary schools, leaving the university
career completely to specialized and professional training.
In the U.S. specialization begins at the baccalaureate level with
declaring a major. ‘A major’ is a group of related courses in a
disciplinary area such as history or biology, although it can also be
an interdisciplinary group of courses in an area such as biology and
society. An individual course typically consists of a sixteen-week
series of class meetings totalling around three hours per week. It
may combine lectures, class discussion and laboratories. Evaluation
is likely to be some combination of laboratory exercises, short
examinations or quizzes, a mid-term examination and/or a final
examination. A research paper may also be required.
The course is the basic building block of undergraduate education
and the credits attached to it, typically three or four, are added up to the
requisite 120 for the degree with the major representing perhaps a third
of that total. The European model would instead be the degree course
with a set of requirements, lectures and examinations geared to
measuring an end result rather than discrete pieces along the way,
through the course.
The science major in the U.S. follows an intermediate format
between the general U.S. undergraduate and specialized European
educational models. Its courses typically must be taken in sequence
and a larger proportion of the student’s time is required. This leaves
less time for electives, those courses apart form major, distribution or
general education requirements in which students may follow a non-
degree interest or simply take a course that has a reputation for being
interesting, easy or challenging, whatever meets their needs!
Vocational choices can be put off at least until the second year of a
four-year undergraduate career, or even later, unless one is in the
sciences. Even if a science or engineering major is chosen late in the
undergraduate career, courses can be made up in summer school or by
taking an extra year for the degree. Some universities even offer a post-
baccalaureate year program to prepare humanities and social science
majors who have decided after graduation that they wish to go to
medical school, a post-bachelor’s degree program in the U.S. A year of
chemistry, biology, physics and other related courses allows them to
meet the basic requirements for admission.
The U.S. undergraduate model of education, based on courses,
continues on into graduate school. A Ph.D. program typically begins
with a set of courses during the first and second years whose purpose is
to bring everyone up to the same level of basic knowledge in the field.
Now, at this late stage, the U.S. system finally begins to follow the
European model, by evaluating students through an extensive
‘qualifying’ examination, cross-cutting an entire field.
Indeed, students do not necessarily have to prepare for the qualifying
exam, the prerequisite for beginning research for the Ph.D.
dissertation, by taking a set of courses. They may also study on their
own, using reading lists, or more likely, in small groups of fellow
students, so-called study groups, where old exams and problems are
discussed. Again, this organized system of preparation for research is
in contrast to the traditional European model in which a student
tackles a research problem from the outset of the advanced degree
process. There, the problem is often set in advance and candidates are
advertised for in the scientific press.
Although the U.S. secondary and undergraduate education varies
greatly in quality, it is at the graduate level that the U.S. excels.
Research groups of a professor with graduate, undergraduate students
and technicians are the basic building block of U.S. academic science.
Assistant professors in the U.S., who would be junior researchers under
a professor in many European countries, have the responsibility for
raising their own research funds through competitive grants to start
their own group. Success or failure in convincing the research
community to fund their proposal is the prerequisite for attaining a
permanent position in a U.S. research university. However, as we shall
see, women and men experience the various stages and phases of this
system quite differently.
With this system of education in mind, we return to the ‘pipeline’
hypothesis. This optimistic hypothesis has been at least partially
disconfirmed by the mixed experience of the most recent generation of
women in science and engineering. True, a large number of women in
the U.S. major in science and engineering and a significant percentage
of women receive BA degrees. As a result, the proportion of science and
engineering bachelors’ degrees going to women has almost doubled in
three decades, rising from 25% in 1966 to 47% in 1995 (NSF, 1998:
171). But the number of women enrolled in graduate school is still
significantly lower, at 38% in 1995 (ibid.:226–7). And the percentage
who emerge with a Ph.D. in these disciplines is lower still, reaching
only 31% by 1995. Even this figure is misleading, however, since it
conceals sharp contrasts by discipline. Most of the progress is
attributable to the greater presence of women in the life and social
sciences, in contrast to the physical sciences and engineering. Highly
unequal participation is still the norm in many fields.
These contrasts are, not surprisingly, most evident at the highest
academic levels. Starting from what was then a relatively strong base of
16% in the 1960s, women increased their representation among Ph.D.
biologists to 40% by the 1990s (see Table 1.1). From a smaller base of
7%, chemistry has seen a corresponding rise to 27%, while the
geosciences increased more dramatically from 3% to 22% in the same
period. However, although mathematics, physics and engineering
have also seen substantial gains in the presence of women among
doctorates, in none of these fields did the 1995 figure even reach one in
five. Thus, starting from bases of 5% in mathematics, 2% in physics,
and less than 1% in engineering in the 1960s, the proportion of Ph.D.s
going to women has risen to 19%, 12% and 11% in the 1990s.
Table 1.1 Women’s share of science and engineering Ph.D.s, 1966–1997
1960s 1970s 1980s 1990s
Biology 16 21 33 40
Chemistry 7 10 19 27
Geosciences 3 6 16 22
Mathematics 5 10 15 19
Physics 2 4 8 12
All engineering <1 1 6 11
Source: U.S. National Science Foundation, Survey of Earned Doctorates.
Most European countries have shown similar patterns to the U.S. For
example, in the United Kingdom, the starting point was so low in most
fields that, even after some progress, women remain far below parity.
In the late 1980s, female chemists in the U.K. were 35% of
undergraduates, 24% of graduate students, 22% of post-doctoral
researchers, 5.5% of lecturers, 1.5% of senior lecturers, 1% of the
readers and 0% of professors. In U.K. academic science as a whole only
3% of professors and department heads were female, compared with
10% in the U.S. In France, there is a decreasing proportion of women
physicists at the higher levels of government-sponsored research
institutes (CNRS). At the lower levels, 42% of the best-qualified
research assistants are women, perhaps in part reflecting their
disproportionately low (16.8%) representation (Couture-Cherki,
The paucity of women in high-level scientific positions in the U.K. is
exemplified by a footnote identifying the author of a preface to a
volume on the condition of women in science: it notes that Professor
Jackson was the first and only female professor of physics in the United
Kingdom (Haas and Perucci, 1986). She is now deceased, but there were
two female physics professors in British universities in the early 1990s
(Healey, 1992). Nevertheless, the continuing low participation at
higher career levels is a virtually universal cross-national phenomenon
despite a history of improvement at the lower levels. University
College London is a bright spot. The proportion of female professors at
9% is three times the national average. This is due to ‘attention to
problems of family and childcare’. Despite the bleakness of the overall
situation, this instance demonstrates that actions can be taken that
will significantly improve matters.
The expectation that the problem of participation of women and
minorities in the scientific and engineering professions could be solved
with a bit of ‘pump priming’ is a supply side thesis. The supply side
approach is codified in the so-called ‘pipeline’ thesis that recruiting
more women is a sufficient strategy. By encouraging girls to study
science, so the theory goes, participation of women and men in science
will become more equal. Once this is accomplished, it is expected that
one can then wait patiently for the next generation to witness women’s
inevitable rise to leadership positions in science in equal proportions to
male scientists. Such a focus tends to neglect analysis of the ‘demand
side’, especially organizational resistance to change and the
persistence of barriers to entry of women into the scientific and
engineering professions. Although there has been some recent
progress, women continue to be chronically underrepresented in
scientific careers, and their participation declines as one moves higher
up the career ladder (Zuckerman, Cole and Bruer, 1991; National
Research Council, 1993).
Role models
Some proponents of women in science believe that presenting young
women entering the scientific and engineering professions with a
picture of the resistance they will encounter will discourage them from
going on. They believe that introducing young women to successful
role models is the best way to enhance their chances of success.
A recent event hosted by the Section of Women In Science at the
New York Academy of Sciences further illustrates the contradiction of
celebrating the achievement of successful female scientists as an
encouragement to girls to do science, rather than warning them about
(and thus preparing them to meet) the possible obstacles. Several
leading female scientists and engineers including Sheila Widnall, then
Secretary of the Air Force, presented an account of their careers to an
audience primarily composed of secondary school women, pursuing
Westinghouse and other awards. Although one woman mentioned
significant obstacles in her path, such as being turned down for tenure
despite considerable research achievements, the overall tone of the
meeting was upbeat and celebratory. The darker side of the scientific
endeavor for females was played down.
As minorities move up educational and job ladders, it is expected
that the problem of exclusion will be solved. However, a significant
increase in women in academic science is unlikely to be realized
simply by increasing the numbers of women who embark on a
scientific career. Encouraging more women to enter the pipeline is at
best a partial answer if so few are willing or able to come out at the other
end and carry on professional careers in science.
2Women and science:
Athena Bound
Athena, the Greek mythological figure with strong female and male
elements in her identity, personifies the dilemma of the contemporary
female scientist. Contemporary female scientists are expected, and
often expect themselves, to combine a demanding personal and
professional life, without its effects on either. Even as some female
scientists struggle to balance their professional and personal lives,
others continue or are constrained to comply with a traditional ‘male
model’ that rigidly subordinates the personal to the professional.
Women in science comprise a diverse set of persons who, despite a
common gender, do not embrace a collective identity.
Many successful women in scientific and engineering professions
expect to have crossed a threshold into a work life in which gender is
irrelevant. These fortunate few females are taken on as apprentices
and, encouraged by their undergraduate professors, enter graduate
school in the sciences and engineering. There again, they encounter an
opaque competitive system that typically depletes their self-
Those women who complete the Ph.D. face a series of career choices
that often needlessly clash with personal aspirations. As Athena found
in pursuing her adventures as a woman in a higher world dominated by
a male ethos, gender matters.
Alternate competing theses have been suggested to explain the
resistance to women in science. It is not ‘either/or’. Rather than
‘barriers to entry’, visible and invisible impediments to women
pursuing a scientific career, or a ‘glass ceiling’ that places limits on
recognition of achievement, difficulties exist at all stages and phases of
the scientific career line.
Women who have avoided discouraging experiences at an earlier
stage often encounter them later. For example, because women
are often excluded from information and informal channels in graduate
school, they have less access to ‘social capital,’ the network of
relationships and connections, than their male peers. Without
this network of professional and social psychological partners, women
of equal or better ‘human capital’ (their skills and knowledge) are
more likely to drop out of graduate school, and those who receive a
Ph.D. lack the ‘halo effect’ that comes from inclusion in such a
When a relatively small number of women traverse the pipeline to
win a faculty appointment the story is said to have ended successfully.
Yet even at this juncture many highly effective women suddenly find
themselves subtly ostracized while paradoxically expected to be ‘role
models’ during the precarious tenure process. We call all of these
disjunctures aspects of the ‘cascade effect’ in which the steady flow of
energy can be short-circuited at any point, regardless of the level of
The experiences of women scientists begin and end with the
consequences of social exclusion in an activity that necessitates,
perhaps demands, community. All too often the consequences of
social isolation and aloneness have been attributed to inherent deficits
within the women themselves. The argument has been that they lack
the right human capital for physically demanding and mathematically
intensive scientific work, whether by nature’s wisdom which has
divided the gene pool or by self-selection into softer fields that permit
greater attention to family. However, the experience of separateness
and stigma makes more understandable the tendencies for self-blame,
lack of self-confidence, fear of risk-taking and role confusion at the
highest faculty level. These constraints on women arise from the way
that society tracks and awards women and men differently, and are
then manifested and reinforced at the organizational level (universities
and departments) through discriminatory practices, misperceptions,
and social networks that can include or isolate women.
Female scientists sometimes respond to the strictures against them
by adopting a research strategy that emphasizes the careful
construction of extensive data bases in a special field rather than rapid
shift from one ‘hot topic’ to the next, longer but less frequent articles,
and a reluctance to test hypotheses for fear of being shot down. The
barriers to women are such that what appears to be a flawed strategy of
reaction actually represents a creative response to obstacles in their
path. We have found that in science, these strategies are enacted
because the interpersonal networks that promote learning, the
practice of the craft, the knowledge transfer, and ultimately the
psychological freedom to take the risks inherent in innovative and
creative work, are different for men and women. What is paradoxical is
that while women pursue the myth that scientific individualism and
isolation spurs scientific breakthrough, it is in fact a fiction that
undermines their advancements, even as men (and some successful
women) operate within networks of collaborative learning that
advance ideas most competitively (Powell, Koput and Smith-Doerr,
Even as they overcame the obstacles in their path, the most successful
female scientists were constricted by their gender. The careers of Marie
Curie, Lise Meitner, Rosalind Franklin and Rachel Carson provide us
with benchmarks of how much has been achieved during the past
century and how far the distance to equality was in each of their
experiences. Indeed, the entry of women into scientific careers, as
more than an anomaly, is a relatively recent phenomenon.
Just a century ago women were barred from seeking degrees and
advanced training in the sciences in most universities in Europe. In
their youth, during the late nineteenth century, Marie Curie and Lise
Meitner received some of their training in so-called ‘flying
universities’ through courses offered in the living rooms of homes by
sympathetic male academics (Quinn, 1995). Other, less sympathetic,
men believing that women’s nature fitted them mainly for family
and home, accepted female candidates only under exceptional
circumstances, and still others, not at all.
When Lise Meitner emigrated to Germany from Austria to pursue a
scientific career, she received financial support from her family that
made it possible for her to pursue advanced studies. To Max Planck, the
doyen of German physics in the late nineteenth century, Lise Meitner
appeared to be one of those exceptional women and he allowed her into
his advanced courses and, most importantly, his laboratory, a training
experience that an improvised university could not provide (Sime,
During the nineteenth century women could attend German
universities only as unmatriculated auditors. Baden was the first
German state to open its universities to women in 1900. Prussia, where
Lise Meitner aspired to follow her vocation for physics in Berlin,
followed in 1908 and was by no means the last. Perhaps ironically, in
the eighteenth century many laboratories, especially in chemistry, had
been in kitchens in the home and thus more accessible to women’s
participation (Abir-Am and Outram, 1986).
The professionalization of the sciences and their incorporation into
the universities during the nineteenth century placed the increasingly
technologically sophisticated experimental sciences beyond the reach
of most interested women. It was not until the 1970s that female access
to the laboratory bench again reached the level that it had attained in
the eighteenth century, a less institutionalized era in the sciences
when upper-class women, at least, had open access to scientific work
through their family and social connections (Gabor, 1995). Although
women gained formal access to university-level scientific education in
the late nineteenth century, informal barriers have persisted into the
twenty-first century.
Such barriers are not so obvious as the rule that, even when she
attained a research position, restricted Lise Meitner’s presence at the
Chemistry Institute in Berlin to a makeshift basement laboratory.
Despite exclusion from the other laboratories and meeting places of
her erstwhile colleagues, Meitner informally guided the investigations
of male peers such as Otto Hahn through the force of her theoretical
insight, combined with careful experimentation. Hitler’s persecution
finally drove her from her laboratory at virtually the last moment that a
person of Jewish background could openly escape from Nazi Germany.
Nevertheless, through clandestine contacts, she continued to advise
her former colleagues on their research program. Always careful to
soothe the male ego, Meitner negotiated a precarious path in German
science, contributing at the highest level but receiving recognition at a
somewhat lower level than her accomplishments warranted.
Meitner remained an outsider all her life, perhaps most poignantly
during her years in Sweden, which provided a haven from Nazi
persecution. Although she had a post at a research institute, she lacked
access to support staff and research resources. Excluded from the Nobel
Prize for the work she did with Hahn, Meitner received fuller
recognition only late in life in the form of an Institute named jointly for
her and Hahn, several individual scientific awards and a street named
after her in Berlin. Nevertheless, she has perhaps only received full
recompense from Ruth Sime, her excellent biographer (1996).
Despite the difficulties she encountered, Meitner was the key person
in a leading German research center for much of her work life. Nazi
persecution, and the war that marginalized Meitner, ironically brought
another female scientist to the forefront. Until very late in her
professional life, Maria Goeppert Mayer (later a Nobel prizewinner)
pursued an outsider career even more on the margins of U.S. academia
than Meitner’s place in the German research system. Maria Goeppert
grew up in an academic family in Göttingen and, when she showed an
aptitude for physics, had access to leading scientific figures in the
community such as Max Born who became her mentor. Nevertheless,
when she married Joe Mayer, an American chemist, and moved to the
United States in the early 1930s, her Ph.D. and advanced knowledge of
theoretical physics only landed her an unpaid position in the physics
department at her husband’s university.
With his support and encouragement she was able to pursue a
research career at the margins of Johns Hopkins University and then at
the universities of Columbia and Chicago (Gabor, 1995). The war-time
emergency that drew many women into the workforce also opened up a
place for Goeppert Mayer in the Manhattan project, where her previous
research meshed with the needs of the crash-program to develop the
atom bomb. Until 1959, on the eve of receiving the Nobel Prize, when
she left the University of Chicago with her husband to move to the
University of California at San Diego, she held no full-time, fully
remunerated academic position. She wanted nothing more than to be
‘one of the boys,’ fully accepted in scientific conversation.
To a great extent she achieved that goal. In discussions in the early
1950s with Enrico Fermi, the Italian physicist then at the University of
Chicago, he encouraged her to formulate her ideas and set forth a claim
to scientific recognition for her elucidation of the structure of the
nucleus. Although she was granted a full academic position only late in
life, Mayer can be seen as the prototypical traditional woman scientist,
devoted to her work to the virtual exclusion of all other aspects of life.
Only through far superior work could she be recognized as an equal.
Mayer’s later career coincided with the beginning of the opening up
of academic science to women’s participation, often through pressures
from the Equal Employment Opportunities process. Despite formal,
tenured positions achieved by a growing minority of women, the way
the world of academic science works still marginalizes women.
Nepotism rules that prohibited universities from hiring husbands and
wives were only the most overt of the many social and cultural
restrictions on women’s full participation in academic science.
Nepotism rules are gone but reminders that science is a man’s world
persist even as women strive to make it their own.
In the early post-war era, when a London college’s common rooms
were still sex-segregated, men could take advantage of scientific
women and get away with it by disparaging their femininity. This is
how James Watson treated Rosalind Franklin in his autobiographical
account, The Double Helix. Franklin concentrated on developing a
data base of X-ray crystallography photographs to elucidate the
structure of DNA but was reluctant to specify a structure until she
could be confident of her results.
James Watson and his colleague Francis Crick were more willing to
put forth speculative hypotheses but they needed access to her data to
guide their model building efforts. Watson attempted to wheedle out
the necessary information from her without offering collaboration and
joint publication, the overt coin of the scientific realm. Rosalind
Franklin, the co-discoverer of the chemical composition of DNA,
relatively unacknowledged by her male peers and unavoidably passed
over by the Nobel Prize committee, owing to her untimely death, had
to wait for recognition from her biographer, Ann Sayre.
Rachel Carson, the biologist and author of Silent Spring, was widely
recognized during her lifetime. However, her fame did not derive from
research findings, in the traditional sense, but rather from analytical
and literary accomplishments. Carson drew together and synthesized a
broad body of evidence on the deleterious effects of chemical
production processes and their effluents on the natural environment
and human health. Indeed, Carson’s own research career was stunted
by the social environment of advanced academic science that made it
difficult for a woman to find a Ph.D. advisor and be taken seriously as a
Despite her mother’s unstinting encouragement and the availability
of a female academic scientist (who herself experienced great
difficulties in her research career) as a role model during her
undergraduate years, Carson was precluded from a conventional
research career by the obstacles she encountered as a graduate student
at Johns Hopkins University during the 1920s. Instead, as is still the
case for many women who wish to pursue scientific careers, she found
a job at the outskirts of conventional science, in her case in a
government bureau as a writer of pamphlets on ecology and wildlife.
Collecting the data for her writing projects through field trips and
personal observation as well as from sources among a wide variety of
researchers, provided the basis for her evocative and precise depictions
of The Sea Around Us and other ecological themes that combined
metaphorical insight and scientific acuity (Lear, 1997). Perhaps
ironically, Carson’s career on the periphery of science has become an
exemplar of a new type of scientific career that emphasizes the
relationship between science and society, rather than the traditional
pursuit of research in isolation from its uses (Tobias and Birer, 1998).
Science writing, research management, technology transfer and
science policy analysis are becoming careers in their own right rather
than offshoots of research career lines. As science becomes more
important to the political and economic spheres, the career lines that
embody these intersections become less exceptional and more
important. If traditional practices hold, however, one indicator of the
increasing acceptance of such occupational endeavors will be their
being taken up by an increasing number of men as well as women. If
traditional discriminatory practices persist, the removal of women as
leaders, if not practitioners, of these occupations, is also likely to take
Female scientists often told us, in interviews, about the obstacles that
women encounter as they pursue their scientific callings. Academic
practices, presumed to be meritocratic and gender-free, often work
against women’s professional success. These effects are sometimes
hidden behind a neutral or even positive facade erected on the
publicized achievements of a few exceptional women, some of whom
deny the existence of obstacles in their path. Other women are
unaware that they have been singled out for negative treatment while
still others are all too cognisant but are also wary of challenging unfair
practices for fear of reprisal.
Sex-role stereotyping sometimes colors advisor–advisee relation-
ships. There are hidden obstacles, such as the length of the tenure
process or the expectation that faculty members should move between
schools to broaden academic training, which become apparent when a
family or relationship is considered. Overt processes of discrimination
include the sexual separation of scientific labor, with men seen as
more appropriate to pursue the theoretical aspects of disciplines
(usually mathematical not experimental) and women as more
congruent with the parts of the field related to practice, policy, and the
The 1997 Harvey award lecture at Rockefeller University
unintentionally symbolized some of the continuing gender disparities
in science. The Rockefeller ceremony was a typical scientific honorary
event in many ways. On the podium, the award recipient in black tie,
Leroy Hood, the distinguished molecular biologist and professor of
computer studies at the University of Washington, foresaw the union
of the biological and computer sciences, and set forth the scientific,
technological, commercial and health benefits that would issue from
this marriage of disciplines. Curiously, even though Rockefeller
University has a number of female faculty and graduate students, and
the biological sciences have for some decades attracted a steadily
increasing number of women, Dr Hood’s formally attired cohort of
hosts were all men.
Invidious distinctions, such as differences in timing, even appear in
seemingly positive experiences such as the receipt of rewards. When a
woman receives a prestigious fellowship or award, too often it comes
late in her work life when it does not provide the same career boost as it
would have at an earlier stage.
Cultural traits that are helpful to the conduct of science as well as
those that are discriminatory must be disentangled from their origins
in order to create a gender-neutral scientific role and workplace. The
sexual separation of labor, the association of certain occupational
specialties with one gender or the other strongly persists in most
Perhaps ironically, the gender associated with a particular field may
reverse, suggesting that the association is hardly inevitable. For
example, nursing, a male occupation well into the nineteenth century,
had become a largely female field not long into the twentieth century.
The profession also, along the way, acquired the presumption of
‘natural’ association with the traditional feminine trait of nurturance.
Those males who continued to enter the profession disproportionately
assumed high-level positions, reflecting the continuing association of
traditional male characteristics with leadership (Etzkowitz, 1971).
The state of the economy also affects conditions of entry and retention
of women in science. Barriers to entry in industry and academia fall
most easily under conditions of expansion and prove more intractable
under conditions of recession. In the United States, Finland, and
Portugal, women gained an increased proportion of R&D (research and
development) positions during the post-war expansion of the sciences
(Ruivo, 1987) On the other hand, when the expansionary period ended
in Finland in 1983, it became more difficult for women, relative to
men, to obtain posts in academic science. During such periods of
increased competition, ‘informal discriminatory practices and
attitudes...’ take hold with renewed strength (Luukkonen-Gronow,
1987: 196).
The renewal of discriminatory practices under harsh economic
conditions can best be avoided if enough women have attained
decision-making positions in science and technology workplaces by
the time the downturn occurs. Otherwise, a disproportionate number
of women ‘... will lose their positions . . . unless preventive measures
are devised’ (Ruivo, 1987:390). Even when they retain their positions, a
disproportionate number of women are to be found on the lower rungs
of the job ladder in many scientific and engineering organizations.
Despite often having to put up a brave front in order to gain acceptance
from their male peers, successful women scientists as well as other
female professionals are becoming more willing to acknowledge the
greater burden that they carry as women, and to seek changes in career
structures and work styles. In an era of financial stringency and
increased research competitiveness, change is made more difficult by
pressures to obtain grants and lengthen one’s list of publications. On
the other hand, the struggle for equality is eased somewhat by allies
among younger male scientists seeking some of the same reforms, to
allow a better balance between their personal and professional lives.
Women scientists and academics, individually and collectively, are
taking a more aggressive approach to redressing the imbalances
between male and female participation, especially at the upper reaches
of academia. Several generations of alumni of Radcliffe College are
engaged in an organized effort to get the administration of Harvard
University to increase the extremely low numbers of women with
higher-level academic appointments at the university, including in the
sciences. They have established an escrow fund to encourage donors to
put their gifts on hold until progress is made.
The technical advisor to this effort is Dr Lily Hornig, a physicist and
long-term activist on behalf of women in science. The perpetuation of
gender-linked work roles and the continuing low rate of participation
of women in many scientific disciplines appears to contradict one of
the accepted standards of science: the norm of ‘universalism’, or in
other words, the principle that scientific careers are open to all who
have talent. The norm of universalism, formulated by sociologist
Robert K. Merton, is that the acceptance or rejection of claims should
not be based upon ‘the personal or social attributes of their
protagonists’ (1973 [1942]: 270). It suggests that although science has
traditionally been a male-dominated profession, it is not inherently so.
Moreover, by excluding persons of talent, as Merton argues in his
analysis of the scientific profession in Nazi Germany, science is
diminished by a ‘racialist purge’ (Merton, 1973 [1938]: 255). Although
not as immediately striking as the elimination of Jewish scientists
from German universities in the 1930s, the long-term relative
exclusion of women has had a similar hampering effect on the conduct
of science.
An earlier body of research identified as fallacious the notion that
advancing age inevitably inhibited high-quality scientific work
(Merton and Zuckerman, 1976). Unwarranted presumptions that
youth was associated with high scientific achievement had served to
justify extreme work pressures in early career stages. These unduly
heightened expectations for early achievement have had unintended
consequences on women’s participation in science, given their
coincidence with child-bearing years. Nevertheless, the implications
of this earlier research for the structure of scientific careers, and the
leeway for possible restructuring, has yet to be taken fully into
account. We view these issues as ‘critical transitions,’ a series of overt
and covert points in the life course when individuals are either
propelled forward to careers in science or deflected away.
During their early childhood years, boys and girls develop different
gendered images of scientists and what they do. Despite some early
negative perceptions, large numbers of girls express interest in science
and many follow up this interest through coursework and extra-
curricular activities, often with the encouragement of teachers and
parents. When they enter U.S. universities young women are dis-
proportionately removed from science and engineering majors by a
harsh ‘weed-out’ system designed to test the mettle of young males,
well socialized in the norms of competition. Nevertheless, some
women, looking back, report a positive experience of being mentored
as undergraduates.
Despite the increased entry of women into science, opposition to
their full participation continues. Implicitly ‘male’ standards of
behavior permeate scientific time and space, including a belief that a
researcher is most productive when their time is devoted to
investigation to the virtual exclusion of all other aspects of life.
Ironically, the personal qualities required for success in science may
be changing. Sociability, a trait traditionally associated with women,
has also been found to be conducive to success in science, especially as
the individual researcher is supplanted by group research, and
multiple-authored publications become the norm. Perhaps, in the
future, female socialization will become a career advantage in the
scientific and engineering professions.
At present, female social attributes are a disadvantage that is
exacerbated by competitive norms. While scientific training is an
arduous process for all, our research and that of others suggests that
women who aspire to scientific careers face barriers that do not equally
exist for men and that equal success results only from truly heroic
efforts (Abir-Am and Outram, 1987).
A letter to the editor of The New York Times entitled ‘Science is for
Childless Women’ (May 17, 1995) exemplifies the persisting dilemma
of women in science. The writer, Stephanie Dimant, identified herself
as ‘... one of those women who “leaked out of the pipeline”’. She
cited the difficulty of reconciling the hours required of a bench
scientist with the demands of raising a family. In a fast-paced, high-
pressured environment, traditional solutions such as withdrawing
from research for several years to raise a family and returning later were
‘so unrealistic as to be comical.’
In bench science, ‘ . . . no second prizes are awarded, and the
economic situation demands unrelenting writing of grant applications
and publication of results.’ Diment could not think of anyone she
knew who had taken the extended leave option and who later returned
to the academic track. Female scientists who made the decision to
combine an academic career with raising a family typically took only
the briefest time off for having a baby and then spent their limited
maternity leave ‘ . . . with an infant in one hand and a telephone
connected to the lab in the other.’ Nor will there be many protests:
given the stringency of research funding and the paucity of academic
jobs, women do not want to be labelled as ‘lame ducks’.
Nevertheless, given the pressures on women, including those that
force the lower-paid spouse (rarely a man) to assume primary
responsibility for child care, ‘It is not surprising that many eventually
make a heart-wrenching decision to leave bench science to those who
have no children or to those who are fortunate to have that
acknowledged asset, a wife.’ Despite these obstacles, some women
with children attain the highest levels of scientific achievement and
However, the price of success appears to be significantly related to
each woman’s ability to adapt to the highly competitive milieu of
science. Dr Shirley Tilghman, director of a ‘large and wildly successful’
laboratory at the Howard Hughes Institute of Princeton University,
concluded, ‘Maybe it’s because I’ve been in science so long that
competition just seems like life. Maybe I’ve just given up.’ A
competitive sports enthusiast as a child, Dr Tilghman, a Canadian
citizen and a recently elected foreign associate of the U.S. National
Academy of Sciences, was featured in an article in The New York
Times ‘Fighting and Studying the Battle of the Sexes With Mice and
The article discusses Dr Tilghman’s research on genetic imprinting,
her experience as a mother and her concerns about the future of women
in science. She is described as having ‘jury rigged the pieces of her life by
being almost preternaturally organised and focused, as well as
spiritedly fierce in her work’, in contrast to many women who draw
back when criticized. Although she raised her two children as a single
parent for most of their childhood, the article did not detail the child-
care arrangements that made this possible.
Her own female graduate students were highly skeptical of their
ability to follow her example, fearing, like Ms Dimant, that they would
be forced to choose between science and motherhood. The students
told Tilghman, ‘Don’t tell me about your experience. Your experience
has no bearing on me.’ They feared the time pressures of a highly
competitive research funding system as well as the accepted belief that
constant presence in the laboratory is a prerequisite for scientific
success. Is there a one-to-one relationship between time put in and
results achieved? Dr Tilghman attempted to reassure her students that
‘[h]ow one does in science is really dependent on your creativity and
originality, and not how many mini-preps you can do in a 24-hour
period’, but the students were not convinced.
Unsure that this assessment applied to them, the students believed
that the grant environment, now more competitive than their mentor
had faced as a young scientist, inevitably increased the time that had to
be devoted to a scientific career. Although a competitive environment
also affects men, increased time pressures have additional effects on
women. Thus, even this notable success story of a woman’s
achievements at the highest levels illustrates the persisting dilemma
of women in science. This dilemma has its roots in the earliest years of
childhood, and our next chapter focuses on how gender socialization
affects the entry of girls and boys into scientific careers.
This Page Intentionally Left Blank
3Gender, sex and science
The strong effect of culturally defined gender roles persists in science
and other traditionally male professions through the social meanings
attached to gender. Rather than a fluid perspective of human attributes
that can be held by members of either sex, behavioral characteristics
are frequently presumed to be innate and immutably ‘masculine’ or
‘feminine’ in the same way as one’s biology.
The thesis that science is masculine, with ‘masculine’ understood as
a cultural rather than as a biological term, ties issues of women in
science to broader questions of gender roles and how they are culturally
defined and transmitted from birth (Ruskai, 1990; Hyde, 1994). As
Howell (private communication) points out, ‘Sex, which is concrete
and universal, specifies no role whatsoever.’ Rather, it is cultural
prescriptions and proscriptions, delineating which behaviors are
appropriate to one sex and not the other, that creates the ‘psychological
meaning’ of what it is to be male and female.
Thus gender as a concept was created to understand ‘the social
quality of distinctions between the sexes . . . for the explicit purpose of
creating a space in which socially mediated differences can be explored
apart from biological differences’ (Hare-Mustin and Maracek, 1988).
However, the concepts of sex and gender become easily entwined and
socialization becomes confused with biology. Taking this a step
further, ‘... it would be illogical to say that being male or female
would, in itself, make someone a good or bad scientist. Yet this kind of
statement is often made.’ (Howell, private communication). Negative
stereotypes persist. The images of the role of women in science may be
slightly more positive, but they have not been radically reshaped.
As the role of women has shifted to meet both society’s needs and
their own, some mothers and fathers relate to their daughters
differently than in the past. They convey possibilities and expectations
that transcend traditional role designations. Many of the young
women whom we have interviewed over the course of the past decade
not only report that it was their father who encouraged them to attain
the Ph.D. in a science discipline, but credit their male advisors for
sensitivity to gender inequities and their strategic assistance in helping
them move forward. Individuals who encourage an interest in science
need not belong to a particular sex or be a member of the family. What is
essential is either a broad, flexible and encompassing vision of gender
that incorporates non-traditional occupations or, paradoxically, a
definition of gender in which it is viewed as irrelevant to vocational
choice. The following discussion illustrates how far we are from this
The gender roles that children internalize influence which sex will
choose to do science as well as who will have the best chances for
scientific success. Blatant discrimination may be a thing of the past,
but culturally generated gender beliefs play a significant role in leading
children toward or away from an interest in science. Perhaps the most
effective covert barrier to women is the simplistic idea that science is
men’s work and that women cannot make good researchers. The
erroneous view of biological sex and gender as one and the same has led
to the association of the male with the scientific role in western
culture: science, like the Church, has been viewed as a ‘world without
women’ (Noble, 1992). In most of this book we explore the conditions
faced by women already in science. In this chapter we discuss the forces
that work to divert females away from scientific careers from the
earliest years of childhood through adolescence.
Differences between boys and girls appear at an early age as part of
the social creation of the ‘self’. As classically formulated by
philosopher George Herbert Mead (1934), the child learns to ‘take the
role of the other’ in play and other social relations. The self is thus
constituted through a reflexive interplay of mirroring events. In the
words of classical sociologist Charles Horton Cooley, the self is a
‘looking glass self.’ Much of our fate thus depends upon what other
people think of us and how we respond to them. Children are
influenced by the appraisals of others and respond according to those
appraisals. If the information received is restrictive, whether based on
race, sex or any other variable, there will be loss of human potential.
Too frequently experiences for females are constricted through a
process in which gender differences are recast into gender stereotypes.
Messages from those close to the child are especially influential in
initially shaping a self-concept. As the influence of the child’s family
decreases, peers, authority figures, and culture as interpreted through
the media, perpetuate the transmission of ideals of masculinity and
femininity (Ortmeyer, 1988). The curiosity of infants and young
children creates for them energy and excitement as they interact and
are drawn to novelty in their environment.
However, if experiences are foreclosed and the child’s world
becomes constrained by what is seen to be appropriate to that sex, the
child not only tends to abandon socially unacceptable interests, but
comes to fearfully avoid that which is unfamiliar (Schachtel, 1959).
Such a self-limiting process is exemplified in the historically high
proportion of girls who lose interest in how the natural world works.
When socialization impedes the individual’s fulfillment of his or her
potentiality, society as well as the individual loses.
Given the forces that push girls and boys apart, is there an inevitable
dichotomy between the female and the male; the female and the
scientist? A sample of fifty-three children from Southeastern
Montessori School, ages two to six, were interviewed to analyze the
emergence of gender differences in the perception of the scientist in
early childhood. The middle-class school composition promised the
optimal probability for a child’s acquaintance with scientists and/or
representations of science. A table contained four photographs (from
the covers of Chemical and Engineering News) of male and female
scientists. The interviewer asked each child to tell her about the
pictures: ‘Who do you think these people are?’ ‘What are they doing?’
‘How do you feel about them?’
Preliminary data that we have collected suggests that sex-typing
persists and appears to become more evident the older the child. For
some boys, science was seen as an activity that males, but not females,
should take seriously. A typical response was that of a four-year-old
boy who said, ‘ . . . only boys should make science.’ The strength of the
male identification with technology was also indicated by a boy who
referred to a picture of a woman at a computer as ‘he’. Yet in several
instances rigid classifications by sex appeared to be less fixed as some of
the children were able to identify both sexes with the role of the
scientist. A four-year-old boy recognized a female scientist in the
pictures and described her work thus, ‘That one looks like a doctor.’
‘She’s working.’ ‘Something in a science. She’s looking. Doing gravity.
Making things fly. Someone who makes things we never saw before.
With machines.’
In addition to discerning gender differences among very young
children in their image of the scientist, the objective of this investi-
gation was to identify discrepancies between their perception of the
role of scientist and the child’s view of themself. Boys were more likely
to see themselves like the scientist, engaged in ‘serious’ behavior.
Boys, in general, were more negative in their views of women
scientists than girls. Moreover, the older boys in the sample (ages five
to six) were increasingly less likely to see girls as possible future
scientists. One said, ‘My sister Amanda wouldn’t like to do this; she’s
really into Barbie dolls.’ When, as part of the survey, the children were
asked to draw scientists, more of the girls who drew women scientists
did so in their second drawings. This suggests that even where the
image of a woman scientist is held, it may be considered ‘not quite
right’ and be presented only after the first, more acceptable picture of
the male scientist has been recorded.
These perceptions and self-concepts illustrate the notion of the
construction of gender schema (Bem, 1983), a highly selective process
comprising ‘a sprawling network of associations’ in which information
is taken in and organized according to the sex-differentiated values of
the culture. A schema functions as a cognitive structure, serving to
anticipate and make sense of new information coming in based on pre-
existing perceptions. For instance, boys and girls at age two had
‘concepts’ of persons by occupation only if they had previous exposure
to people holding the particular role. None of the two-year-olds were
able to identify ‘scientists’, but those youngsters whose parents were
expecting babies understood the notion of ‘doctor’ and thus applied
this familiar concept based on the white lab coat. In other words, the
young child’s perceptions are not always dictated by a concrete
situation. In this same way, the child learns to link attributes with
their own sex or the sex of others. The perception of gender does not
depend on the actual situation, but rather on organizing information
that makes sense of novelty.
Two-year-olds without expected siblings, and therefore less
exposure to doctor visits, were consistent with Piagetian develop-
mental theory. When a child said, ‘I don’t know this person’, it
indicated that there were no available mental concepts whatsoever
which reflected the attributes in the photograph. Therefore the child
could make no interpretation of the photograph.
However, children above the age of three could identify scientific
and medical occupational roles and had begun to link occupations with
sex based on their knowledge of their family and the outside world. A
three-year-old girl said, ‘That’s a man doctor,’ while a three-year-old
boy identified another picture as ‘a big girl doctor with a cigarette.’
A six-year-old boy said, ‘My daddy’s a builder; my mom’s a scientist,
but she’s a student.’ On the other hand, although a three-year-old girl
recognized and characterized the activity, she did not attribute the
activity to the woman in the photographs, only the man: ‘Someone is
working hard. He’s a scientist because he is doing science stuff.’ By age
three to six, not only were most children in the sample familiar enough
with the concept of the scientist to correctly identify the pictures, but
they had begun to generalize sentiments and meanings across
situations, demonstrating how at this early age information is encoded
and organized according to cultural definitions of what is masculine
and feminine (Bem, 1983).
In responding to the pictures, girls tended to see doctors while boys
saw scientists. Showing the power of popular culture on gender images,
one six-year-old male associated the images in the pictures taken from
Chemical and Engineering News with similar images seen on
television: ‘They’re called scientist[s]. I know that because of a cartoon
that does [the] same thing he does.’ Possibly boys are more frequently
exposed to science in the media, games and toys, whereas girls
apprehend the physician’s role through their own increasing
encounters with women doctors. When queried, girls more often
identified the scientists as doctors, possibly because more
obstetricians and pediatricians are now frequently women, chosen by
other women rather than male doctors.
Young girls then not only learn about doctors in relation to female
reproduction, but may experience this specialty as gender-neutral or
women-friendly. Certainly it supports Bem’s suggestion that gender
schematic processing is dependent upon the social context and is a
‘learned phenomenon and, hence, neither inevitable or unmodifiable.’
Indeed, some boys draw similar androgynous conclusions especially
when their mother holds a non-traditional occupational role. Thus, a
three-year-old boy said, ‘My mom’s a doctor. [The person in the picture
is] a doctor because he has a thing on his coat.’
Despite the persistence of sex-typing, there were indicators of
change. For instance, a four-year-old girl demonstrated a working sense
of the disciplinary order and division of labor in science. She said,
‘Doctors fix people. A scientist checks things. I only want to be a
veterinarian’, and a four-year-old boy used clothing, not sex, as the
identifying marker: ‘Scientists! The clothes look like scientists.’
Lastly, a six-year-old boy could identify researchers with a purely
operational definition of the scientific method, notably gender-free,
‘These are scientists. They’re working really hard with experimenting
to see if something does it or not. That’s figuring out what do.’
Adults who provide a neutralizing message serve to counteract
stereotypical notions of gender pervasive in the larger society. For
example, a six-year-old female at Southeastern Montessori School told
her interviewer ‘I would like to do science. My mom gives me kits some-
times. A scientist is somebody who creates things. I like computers.’
Although science was still linked to notions of gender by a number of
the youngsters, some shift in traditional gendered associations of the
scientific role can be inferred from the responses of others.
The notion of clear-cut sex differences in the way newborns behave has
not been borne out. In repeated studies girls could not be distinguished
from boys without seeing physical differences. In those few studies
where subtle behavioral differences were discerned, they were deemed
moderate at best and were observed to diminish, disappear or be
heightened based on the social context. The majority of studies have
found that for parents, the sex of a newborn is a ‘central organizer, a
potent description of who the newborn baby is’ (Tronick and Adamson,
1980). From infancy, boys and girls receive divergent messages from
adults. Both Block (1984) and Hoffman (1977), in their studies of child-
rearing practices, found parents encouraged their sons to actively
explore the physical world, emphasized achievement, competition,
and self-reliance, and felt it was important they try new things. In
contrast, daughters were expected to be ‘kind, unselfish, attractive,
loving and well mannered [and grew] up in a more structured and
directive world than males’ (Block, 1984).
With few exceptions, these studies reflect how adult reactions to
babies based on presumed sex perpetuate cultural beliefs about
masculinity and femininity. Beginning in infancy, adults speak to and
touch girls and boys differently. Sex-typed toys are offered by both
parents with ‘boy toys’ providing more active physical manipulation
and feedback from the physical world. In general, boys are given more
freedom and less supervision, while girls are interrupted more
frequently by their parents, particularly fathers.
Researchers alternated between dressing the same baby in pink and
calling her Beth or dressing her in blue and calling her Adam. The
adults who played with ‘her’ or ‘him’ noted that Beth was ‘feminine and
sweet’ and Adam was ‘sturdy and vigorous’ (Will, Self and Datan,
1976). In a similar study, subjects were shown a videotape of a nine-
month-old baby playing with a jack-in-the-box. When the baby was
said to be a girl, her response was called fear; when a boy, anger (Condry
and Condry, 1976) When a researcher told subjects that her new baby
was a girl they responded with coos about ‘how sweet she is’. When
others were told that the baby was a boy, they said he was ‘big and
strapping’. In yet another study, parents were given a fourteen-month-
old with whom to play. When designated a boy, ‘he’ was encouraged in
active play with typically masculine toys. As a girl, ‘she’ received more
nurturance and cuddling. In each instance, parental attitudes were
projected onto the baby, depending upon the sex label, once again
demonstrating the thesis of sociologist W.I. Thomas that ‘a social
situation is real if it is real in its consequences.’
These unconscious parental behaviors create an underlying ‘gender
awareness’ during early childhood in which the world becomes
categorized into mutually exclusive classifications (Condry, 1984).
Gender roles come to be perceived as ‘all or nothing’ categories leading
to prescriptions that scientists are men and secretaries are women. A
female child may therefore believe that she cannot be a scientist even
if she would like to because she is of the wrong sex (Kohlberg, 1966).
This sex-typing process frequently continues with all authority
relationships as the child’s social world expands. However, the
broadening of experience can sometimes provide the possibility for
new influences which serve to enhance a child’s self-concept when
early familial experiences may have been rigid and stifling.
Among the many forces working against women’s participation in
science is the masculine image of the scientific role that frequently has
taken hold by primary school. This is often followed by neglect and/or
discouragement of girls from doing mathematics in secondary school
in concert with parental (and particularly maternal) perceptions of
mathematics as difficult and non-essential for their daughters (Eccles
and Jacobs, 1986). Moreover, conformity to stereotypes is frequently,
but subtly, encouraged by educators and other authority figures. By
puberty, these cumulative cultural messages are reinforced by the
powerful need for peer approval and acceptance.
Unless the child is exposed to a wider range of possibilities, they may
come to see gender-determined life choices as mutually exclusive and
exhaustive. It is notable that both brighter girls and boys have come
from family environments which were responsive to all aspects of the
child’s personality. Optimal performance was reported when children
of each sex were encouraged to take the role of the other. Girls were free
to actively explore and were ‘encouraged to fend for themselves’ while
boys had received ongoing ‘maternal warmth and protection’
(Maccoby, 1966). However, once in school, many female children who
have had a wholesome beginning in which they were relatively free to
explore all aspects of themselves may now experience an erosion of the
self based on stereotyped demands of teachers.
Enlightened parents may feel helpless in counteracting ubiquitous
sexual stereotyping once their daughter enters school and other social
situations outside the home. The recent dismay of dynamic classroom
teachers who had allowed their interactions with students to be
videotaped underscores the unconscious pull of stereotypical sex-
typing behaviors even by educators who thought they were self-aware.
As had been found in earlier studies, the teachers observed themselves
calling on boys more frequently, providing extended conversation,
information and help. On the other hand, because girls were less vocal
and more cooperative, teachers were less likely to notice them or
reward their talents, appreciating the girls’ compliance in large
classroom settings which they had to control.
A longitudinal study, over a 25-year period, found that girls were
eight times less likely to call out comments, but when they did were
reminded to raise their hands. In contrast, teachers responded to the
typically rowdier and more assertive behavior of boys. Thus highly
intelligent young girls often give up their own assertiveness and risk-
taking behavior in order to earn their teacher’s acceptance, fulfilling
the social virtues of selflessness and cooperation (Sadker and Sadker,
By treating boys and girls differently, teachers encouraged
‘the exploratory, autonomous, independent mathematical skills
associated with males . . .’ and discouraged them in females (Birns,
1976). Similarly, teachers gave extra attention to boys who chose to
play at more complex tasks, but did not reward girls for the same
behavior (Fagot, 1978). In addition, girls received few rewards for highly
active behavior, whereas boys gained the attention of their teachers
and also the admiration of their peers. While compliance may provide
some rewards, it does so at a cost to development. Paradoxically,
behaving like the boys can bring with it severe penalties, including
adult reprimand and peer ostracism.
Significantly, teacher attention favors those attributes considered as
male. Boys frequently succeed in gaining attention by using negative
and inappropriate behavior, while less aggressive but more appropriate
expressive bids by girls are often ignored (Block, 1984). Thus many
teachers unconsciously reward compliance and cooperation from girls,
while encouraging or condoning a highly competitive style of
interacting for boys. At its most virulent, competitive ‘putting others
down’ can become a pervasive part of personal interactions within the
classroom. Not surprisingly, it is these very kinds of behavior on the
part of adult male peers that have been identified by female Ph.D.
candidates as disturbing and alienating.
Since boys have generally been previously exposed to manipulative
toys, such as construction sets and models, they enter science classes
with more confidence based on these earlier experiences. Moreover,
the teacher’s interactions with students influence children’s
perception of their own ability to do science. Not only are interactions
more frequent with boys, but science experiments are often segregated
into all-girl and all-boy groups with the boys receiving more attention,
or if in integrated groups, the girls often watch the boys do the
experiments (Wellesley College Center, 1992).
A learning environment that emphasizes experimentation, self-
motivated exploration and inquiry is often an unfamiliar experience
for girls, given ‘the more structured, supervised, prescribed and
proscribed world of girlhood as compared to boyhood’ (Block, 1984).
Thus rather than having a lack of interest in science, girls may tend to
avoid the lack of structure in science laboratories where their anxiety
would be higher than that of boys. In contrast to this kind of early
experience, our studies indicate that women Ph.D. candidates
frequently identify a particular high school science teacher who was
attuned and responsive to their interest and inherent competencies,
ultimately assisting in the development of independent exploration.
On a deeper level, these socializing processes may also account for
later differences in the cognitive strategies used by girls and boys. In
their more highly structured play and learning environments, girls use
‘assimilative strategies’ for adapting (that is, they tend to fit new
information or experiences into their pre-existing cognitive ways of
understanding) and are discouraged from engaging in more anxiety-
provoking innovative efforts (Block, 1984). In contrast, where boys
have been encouraged to explore ‘a less predictable world . . . success in
inventive ad hoc solutions would be expected to benefit boys’ self-
confidence’ and the freedom to take risks. Along these lines, women
graduate students in our studies, comparing themselves to male peers,
reported feeling ‘less able to take risks.’ However, their prior
educational successes and continued positive movement through the
pipeline appear to reflect the importance of women’s programs, as well
as past and current mentors who created and continue to provide
learning and social experiences free of gender-laden constraints.
Thus, gender differences become significant when masculine and
feminine are defined in terms of narrow cultural norms, some of which
are peculiar to American society. These norms set up a supposed
contradiction between the traditional notion of ‘scientific values’ and
what are considered feminine characteristics. Once they have
identified themselves as males or females, girls and boys then want to
adopt the behaviors consistent with their newly discovered status.
This process of socialization results in children forming a conception
of maleness and femaleness, revolving about such highly visible traits
as hair style, dress, and occupation. They then use these gender images
to organize their behavior and to cultivate attitudes and actions
associated with being a boy or a girl (Kohlberg, 1966)
Although a young girl may have attended the same classes as male
peers, by the time a young woman enters high school, she will not have
had the same educational experience (Eccles and Jacobs,1986). The
cultural message that has predominated has been that active exploration,
the capacity to be competitive, and the opportunity to handle machinery,
play with chemistry sets and operate a computer are exclusively male
activities. Not only are boys pictured prominently on the packaging of
most science and building toys, but even the box of a chemistry set shows
a girl looking on while a boy conducts the experiment. An eighth-grader
described a dream in which she saw herself working for a scientist who
did the experiments while she was left to write the paper. The youngster
ended with, ‘That’s the way it is, right? . . . That’s how we’ll end up, the
girls.’ It is not that most girls will have been directly told that they ‘can’t’
do what boys can. Indeed, most will be encouraged to ‘fulfill their
potential.’ Nevertheless, in various ways many receive veiled messages
of discouragement and denigration (Orenstein, 1994).
One of the primary tasks of adolescence is the further consolidation of
identity. Peers replace adults in importance, and social acceptance has
primacy. The cumulative subtle and covert messages regarding
expectations and perceptions of females eventually influence the sense
of one’s place in the world, feelings of self-worth, and possibilities for
the future. At a time when peer acceptance is crucial, conformity to
stereotypical social roles is heightened. The anticipation of rejection
by male and female peers in competitive activities may make such
activities too threatening for most young women to undertake,
particularly if such ‘masculine’ behavior contradicts the attitudes of
the popular culture and the family (Block, 1984; Eccles and Jacobs,
1986). Therefore, adolescence is a logical time for the hidden meanings
of gender roles to solidify further and become enacted in school
performance and life choices unless the social milieu strongly
encourages an inclusive ethos free of assumptions about gender.
By the time young women enter college many are not sufficiently
qualified to major in those hard science areas that require a strong
mathematical background, having avoided advanced classes in high
school. Based on the research available and the fact that girls
demonstrate equal mathematical capacity before adolescence, we can
regard this ‘giving up’ of a potential skill as part of the legacy of the days
when women’s attempts at mastering mathematics met with
indifference or overt rejection.
The controversy over differences between males’ and females’
mathematical skills, in concert with the issue of ‘math anxiety’, is of
particular importance in illustrating how gender-appropriate roles
affect later competencies. Nevertheless, an analysis of the results from
numerous studies found no differences between males and females of
any age in ability to understand mathematical concepts (Hyde, 1994).
Gender differences in science achievement do not appear until the
eighth grade; thereafter, strong gender differences in career orientation
emerge, with half as many girls as boys showing interest in
mathematics and science careers (Catsambis, 1994).
Thus, females appear to have the same aptitude for mathematics as
males, but begin to lose interest and take only the minimum
requirement in high school. Based on class grades, girls and boys are
similar in mathematical and scientific ability until about tenth grade
when girls decline to take elective mathematics courses. It is then that
sex differences in problem-solving abilities begin to emerge. The
question is not that of inherent ability, but one of why girls drop out of
mathematics courses in high school and college (Hyde, 1994).
A 2-year longitudinal study of seventh- through ninth-graders, and
their mathematics teachers and parents, argued that sex differences in
attitudes toward mathematics, as well as achievement, are due to
‘math anxiety’ (Eccles and Jacobs, 1986). The level of anxiety correlates
with the gender-stereotyped beliefs of parents (and particularly
mothers) and the values placed on mathematics by the family.
Students’ attitudes and plans to continue taking mathematics courses
were substantially influenced by parental perception that mathe-
matics was difficult and of little value for their daughters. Thus rather
than grades and performance having direct bearing on girls’ self-
confidence in mathematics, beliefs about their competence and desire
to pursue interests and goals appeared to be strongly influenced by
the parent’s response to their daughter’s grades. In short, prior
performance, even when stellar, was secondary to the response of
significant others. While there is a correlation between teachers’
attitudes and the student’s beliefs, the impact of teachers was not as
strong as the influence of the parents.
There is conflicting evidence over whether the support of the school
or the family is more significant to the minority of young women who
do express an interest in science. Alice Rossi (1965) noted that a young
girl with high intelligence and scientific interests must come from a
very special family situation and must be a far rarer person than the
young boy of high intelligence and scientific interests. On the other
hand, if she reaches adolescence with the same intellectual
inclination, it is often despite her early family or social experiences
rather than because of them. This may reflect why women, when
questioned in college about the background of their science interests,
frequently point to particularly important teachers they had, often as
early as the third or fourth grade, who provided them with new
channels of communication and new expectations. In contrast,
graduate students in our studies frequently cited both parents, and
most often the father, as highly encouraging. Fathers have been found
to be particularly supportive of their daughters’ mathematical abilities
(Eccles and Jacobs, 1986).
The gap in gender differences in standardized testing and
achievement has narrowed since the early 1980s (Berryman, 1983;
Orenstein, 1994; Hyde 1994). Nevertheless, girls’ historically lower
mathematics scores are typically attributed to biological
characteristics, even a so-called ‘math gene.’ Yet how can biological
differences explain the 50% reduction in this score gap between boys
and girls by 1994 and current near-parity? Girls’ supposedly lesser
‘spatial abilities’ have also improved with increasing exposure to
spatial tasks. If mathematical skills were biologically determined they
would presumably be impervious to such rapid and dramatic shifts.
Another seeming anomaly is the fact that non-Caucasian girls
outperform boys in the highest-level mathematics classes in Hawaii
(Orenstein, 1994). Since most people construct their perceptions of the
world largely in accordance with cultural prescriptions that they take
for granted, ethnic variation within the larger structure speaks to the
influence of subsets, including family attitudes. During the past few
decades there has been increased awareness of the way girls are treated
in mathematics and science classes. Indeed, given the coincidence in
the timing of the change in testing outcomes, even a modest shift in
social attitudes might well be an indirect cause of improved awareness
in how teachers relate to girls.
Nevertheless, it has been argued, perhaps most prominently in an
article in the journal Science, that innate, biological male superiority
was the best explanation for sex differences in standardized testing,
based on the premise that girls and boys received identical training
(Benbow and Stanley, 1980). In contrast to the fathers, mothers’
confidence in their daughters’ mathematical aptitude declined further
in response to the Science article. Given the prestige of the journal and
its prominence within the scientific community, the influence of the
article in adding to pre-existing gender bias in the sciences could be
high, but is still unknown.
Benbow and Stanley’s paper certainly enabled advocates of
‘meritocracy’ to draw the conclusion that females are, in fact,
‘incompetent’ at science and competent at other things. Along these
lines, our own findings indicate that any inference of ‘difference’,
including variation in socialization, opens the floodgates for negative
interpretation of what it means to be female. For instance, graduate
women’s programs, created to mitigate social isolation by building
networks, are frequently interpreted as indicating that women have
special needs. The possibility of different biological influences does
not have to imply that behavior is ‘predetermined’. Instead, biological
propensities may be ‘manifested in behavior in diverse and complex
ways, as organisms are shaped by . . . the environment in which they
must function’ (Block, 1984).
Our concern is how socialization, based on stereotyped sexual
division, appears to restrict the possibilities for girls and women and
has a destructive impact on the sense of self for both sexes. The
consequences of female socialization have a concomitant deleterious
effect for young males as well, in the demand that they ‘maintain the
image of the aggressive, detached, active male’ (Gould, 1978).
Neurological differences between right and left side brain
development, or distinctive identifications and attachment between
mother and daughter, have provided insight into the more verbal,
relational and nurturing characteristics of females (Chodorow, 1978;
Miller, 1976; Gilligan and Brown, 1990). However, qualities of
maleness and femaleness are not rigid and impermeable. Boys and men
have rich capacities for empathy, nurturance and attunement in
relationships just as girls and women have aggressive, active, and
competitive capacities.
Considering gender along lines of difference or non-difference
presents numerous paradoxes with which women in the scientific
community currently struggle. As mentioned above, when differences
are acknowledged, as in graduate women’s programs, females are
negatively construed as the same (in some way needy or deficient) and
not viewed as individuals. Thus, by focusing on difference, this
approach minimizes similarities between males and females while
obfuscating institutional sexism. On the other hand, adherence to a no-
difference model ‘makes man the referent . . . women must aspire to be
as good as men’ (Hare-Mustin and Maracek, 1988). For women
scientists in less hospitable milieus, the no-difference model creates
another paradox in which women must be either ‘better than’ or ‘just
like’ men in order to prove they are equal.
By adolescence, gender socialization has affected career plans. The
achievement orientation of the scientist depends on competitive
success. Yet for females, competitive success is often accompanied by
great emotional costs based on family attitudes and their early
experiences in the classroom.
In many ways, women are unable to choose to do science; society has
already chosen who will do science through its construction of gender
roles. There is considerable evidence of the relationship between
the adolescents’ notions of gender-appropriateness and recruitment
to scientific careers (Eiduson and Beckman, 1973; NSF, 1988). The
image of the scientist as eccentric, non-conformist, and lacking in
emotional capacity suggests that the potential recruit must have
certain types of personality characteristics and live a particular
If the caricature of the scientific personality and lifestyle does not
mesh with the student’s interests, beliefs, and values, she or he is
unlikely to become committed to being a scientist. For women, the
requirements of a college major or pre-college mathematical
preparation are not the only factors when making a career choice
(Barnett, 1978). Rather, women avoid majors in science and
engineering, in part, because they are socially ascribed as ‘men’s jobs.’
Other studies corroborate these findings (Gerson, 1985; Berryman,
An early study by anthropologist Margaret Mead and Rhoda Metraux
(1957), conducted during the 1950s, identified a negative image of the
scientist among high school students in the U.S. and found that girls,
especially, viewed a scientific career as an inappropriate form of work
for themselves. Girls rejected science as being concerned with things
rather than people. Moreover, they viewed science as a highly
demanding career that would take them away from their future
husbands and children, an issue which continues to trouble and
impede women scientists today.
More recently, in Norway, strong clashes were found between girls’
values and priorities and their perception of what it means to be a
scientist (Sjoberg, 1988). Indeed, a series of U.S. studies, from the 1950s
through the 1980s, showed that both boys and girls identify the typical
scientist as a man (LaFollette, 1988). On the other hand, a study of
elementary and middle school children in Taiwan found that although
older students were more influenced by stereotypical images
representing scientists as men, female students were three times more
likely than male students to draw female scientists (She, 1995).
A study of the image of the scientist among older primary school
students in Ireland found that girls, but not boys, drew pictures of
female scientists, suggesting that even if the boys did not see science as
an appropriate career for women, some girls, at least, could envision
the possibility (O Maoldomhnaigh and Hunt, 1988) .
Given these disparate and possibly contradictory findings across
cultures, what is not yet known is which aspects of gender remain fixed
and which are more flexible and amenable to change as individuals
mature, particularly as they pertain to the image of the scientist.
American girls’ performance in mathematics and science is still
negatively affected by traditional gender beliefs. In other countries,
particularly in Asia, boys and girls perform equally on mathematics
tests. David Dunn of the University of Texas at Dallas notes, ‘We tend,
both in our family lives and in grade school and high school, to counsel
girls away from math and science.’
By the time young women attend high school and college, they are
frequently viewed as inappropriate persons to become scientists and
engineers. Girls are often given the impression that they will face ‘ . . .
intolerable obstacles, conflicts and handicaps’ (Moulton, 1972), an
understanding that all too accurately reflects the traditional
organization of science.
4Selective access
Social practices that work against women’s participation in science are
often embedded in a seemingly gender-neutral competitive selection
system. In this chapter we discuss how the normal workings of the U.S.
higher educational system push women out rather than recruiting
them into science and engineering careers. We contrast the workings of
the unofficial ‘weed-out’ system in undergraduate education at large
universities with a ‘reverse weed-out’ system at small colleges that
must recruit students to their science courses in order to maintain
their majors.
The weed-out system
In large universities at the bachelor’s or first degree level, women often
encounter a ‘weed-out’ system of courses based upon a competitive
model that is designed to eliminate unwanted numbers of prospective
students. This system has even worse effects on women than it does on
men. Its encoded meanings, obscure to young women whose education
was grounded in a different system of values, produce feelings of
rejection, discouragement, and lowered self-confidence (Seymour,
A fortunate few women, after surviving this perilous journey, are
recruited into a smaller scale, supportive version of the graduate
research apprenticeship model. These women had no difficulties
academically as undergraduates, in fact they were usually at the top in
their classes and worked closely with their professors who were often
important researchers. This perhaps explains why virtually all of the
students interviewed in the graduate school samples reported positive
and successful experiences in undergraduate school. Once in graduate
school, many women recall their college experience as having been a
nurturing environment that typically provided them with a mentor (as
advisor, professor, lab director, etc.) who encouraged them to aim for
the Ph.D. Ironically, once in graduate school, women often encounter a
second weed-out system, a harsher, more discouraging, version of the
research model they experienced as undergraduates. Their self-
confidence, so precariously acquired in college, is once again deflated.
Most women who choose to major in science at university have had a
positive high school experience which was one of the factors that
encouraged them to continue. Thus, at each level the system removes
disproportionately large numbers of women from the science career
pipeline while providing a positive experience to a much smaller
number, most of whom are fated to have a discouraging experience at
the next level of their training. It may be said that the system applies to
men as well, but as we shall see, the same strictures affect women
worse than men, given the cultural differences between most women
and men.
The weed-out system sifts large intake classes for intrinsic interest,
talent, and fortitude, while, at the same time, drastically reducing the
classes to a size that departments can handle in the upper division
regardless of variations in the caliber of particular student cohorts.
‘Weed-out’ is a long-established tradition in a number of academic
disciplines, but it is dominant in all science, mathematics and
engineering (SME) majors. It has a semi-legitimate, legendary status
and is part of what gives SME majors their image of hardness. It is thus
an important feature in students’ informal prestige ranking systems,
both for individuals and for majors, disciplines, or sub-specialties.
Weed-out systems are similar to the ‘hazing’ practices of military
academies and fraternities. Although these practices seem archaic
they persist because they serve important functions that are difficult to
achieve by other means. ‘Weed-out’ strategies are perceived as a test for
both ability and character and are the main mechanism by which SME
disciplines seek to find the most able and interested students of all who
enter their introductory SME classes. The system operates in its most
stringent form in larger, less elite universities, and although it still
exists in elite universities and small colleges, its impact is moderated
by countervailing forces. These forces include, in the one case, the
likely higher class background of the students, and in the other case,
the ability of programs to accommodate a higher proportion of students
at the upper level.
The core of college education is the course, a set of class meetings
held two or three times a week during a semester of fifteen or sixteen
weeks, punctuated and/or concluded by examinations testing
students’ knowledge. The official purpose of a course is to impart
knowledge to students, traditionally by lecture or recitation, more
recently through laboratory practice or class discussion. Based upon
the oral transmission of knowledge and originating with the founding
of universities in the medieval period, before the invention of the
printing press, the course of lectures has been a quintessential element
of the academic structure.
In addition to its educational purpose, the course has traditionally
had a role of evaluation, as the examinations attached to it show. Some
colleges have tried to separate education from evaluation by
scheduling examinations after blocks of courses, for example in a
‘junior examination’. For the most part the examination has remained
a part of the course, also serving as a sorting mechanism to place
students into different categories. The highest category traditionally
has been the few students most worthy of personal attention from the
master: those most likely to have the abilities and inclination to
become masters themselves.
As universities became training institutions for distinct professions
the selection mechanism took on other functions as well. If there was a
surplus of students interested in a profession, excess numbers could be
selected out by raising the standards and eliminating the unwanted
Selection mechanisms can also accomplish more covert purposes,
even some that may not be acknowledged consciously by persons
running the system. For example, one covert goal may be to eliminate
persons who are not in the image of those already in the profession.
Selection can take place on a seemingly meritocratic basis by
organizing the process according to cultural criteria that fit and
therefore select for members of one group but are incompatible with,
and therefore deselect, members of the unwanted group. Thus, the
normal operation of the academic system will insure that reproduction
of the profession occurs in a way that selects for people with similar
social, cultural and economic characteristics to those already in the
profession. Those eliminated will have little grounds for protest since
the selection has seemingly been made according to universalistic
The weed-out process acts as a post hoc selection system which