VU Research Portal

The underrepresentation of women in science: differential commitment or the queen

bee syndrom?
Ellemers, N.; van den Heuvel, H.; de Gilder, T.C.; Maass, A.; Bonvini, A.

published in
British Journal of Social Psychology
2004

DOI (link to publisher)

10.1348/0144666042037999

document version
Publisher's PDF, also known as Version of record

Link to publication in VU Research Portal

citation for published version (APA)

Ellemers, N., van den Heuvel, H., de Gilder, T. C., Maass, A., & Bonvini, A. (2004). The underrepresentation of
women in science: differential commitment or the queen bee syndrom? British Journal of Social Psychology,
43(3), 315-338. [Link]

General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners
and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research.
• You may not further distribute the material or use it for any profit-making activity or commercial gain
• You may freely distribute the URL identifying the publication in the public portal

Take down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.

E-mail address:
[Link]@[Link]

Download date: 14. Jan. 2025

 315

British Journal of Social Psychology (2004), 43, 315–338

© 2004 The British Psychological Society
[Link]

The underrepresentation of women in science:

Differential commitment or the queen bee
syndrome?

Naomi Ellemers1*, Henriette van den Heuvel2, Dick de Gilder3,

Anne Maass4 and Alessandra Bonvini4
1
Leiden University, The Netherlands
2
Leiden, The Netherlands
3
Free University Amsterdam, The Netherlands
4
University of Padua, Italy

We examined possible explanations for the underrepresentation of women among

university faculty, in two different national contexts. In the Netherlands, a sample of
doctoral students (N = 132) revealed no gender differences in work commitment or
work satisfaction. Faculty members in the same university (N = 179), however,
perceived female students to be less committed to their work and female faculty
endorsed these gender-stereotypical perceptions most strongly. A second study, in
Italy, replicated and extended these findings. Again, no gender differences were
obtained in the self-descriptions of male and female doctoral students (N = 80), while
especially the female faculty (N = 93) perceived female students as less committed to
their work than male students. Additional measures supported an explanation in
social identity terms, according to which individual upward mobility (i.e. of female
faculty) implies distancing the self from the group stereotype which not only
involves perceiving the self as a non-prototypical group member, but may also elicit
stereotypical views of other in-group members.

‘Women in science: why so few?’ was the title under which, in 1965, Alice Rossi
addressed the relative absence of women who pursued academic careers at the univer-
sity. Today, over 35 years later, we can still pose that same question. The pervasiveness
of this phenomenon is illustrated by data showing that the current proportion of
female faculty in universities across the world almost never exceeds 25% (Lie, Malik, &
Harris, 1994). Importantly, for some time now, about half of the university students
have been women, and in some areas such as psychology they even constitute a clear
majority among undergraduate students (in the Netherlands, Ellemers, 1995; and Great
Britain, Radford & Holdstock, 1995) as well as among those who successfully complete

*Correspondence should be addressed to Naomi Ellemers, Social and Organizational Psychology, Leiden University, P.O. Box
9555, 2300 RB Leiden, The Netherlands (e-mail: Ellemers@[Link]).
316 Naomi Ellemers et al.

their PhDs (in the USA, Sanderson & Dugoni, 1999). Nevertheless, the under-
representation of women becomes worse as their academic position is higher and
this has not improved significantly during the past 40 years. A recent cross-national
comparison (Osborn, 1998) showed that in 13 out of 16 European countries the
percentage of female full professors was 10% or less, and this was only slightly better in
the USA (13%) or Australia (14%).
The present investigation, conducted in two different European countries (the
Netherlands and Italy) constitutes an attempt to investigate two possible factors that
may contribute to the underrepresentation of women among university faculty, namely
differential work commitment of male and female scientists, or biased perceptions of
these levels of work commitment. In each national context on the one hand we
investigated whether male and female doctoral students show signs of differential
work motivation. On the other hand, we assessed whether others in the organization
hold gender stereotypical expectations of the work commitment of male and female
doctoral students. This comparison between self-reports of young scientists at the
beginning of their academic careers and the way they are perceived by more senior
faculty allows us to identify actual gender differences versus potential misperceptions
of work commitment as possible explanations for differential career patterns observed
among male and female academics.

The underrepresentation of women in science

Over the years, several differences have been established in the careers of male and
female academics. For instance, in the Netherlands female post-doctoral researchers
are less likely than men to get a tenure track position (Crum & Bal, 1998), and given
the same qualifications, female lecturers are less likely to be promoted to the senior
level than are their male colleagues (Hawkins & Noordenbos, 1991). Furthermore,
even women who have been extremely successful in the academic system, namely
tenured full professors at the prestigious Massachussetts Institute of Technology (MIT),
turn out systematically to earn less and to receive less laboratory space and research
money than do senior male faculty (Lawler, 1999). Not surprisingly then, women are
more likely than men to leave the university, in order to pursue a career elsewhere
(Portegijs, 1993).
This is generally perceived as a problematic development, not only in view of justice
considerations (denial of equal opportunities to women), but also in terms of the
future viability of university teaching, as well as with respect to optimal use of the
available potential for scientific research (Bosch, Hoving, & Wekker, 1999). Accord-
ingly, in virtually all European Union countries special policy measures have been
taken to address this issue, such as the formulation of target figures for the represen-
tation of women among university faculty, the provision of additional research grants,
or the institution of special chairs for women (see the ETAN-report, 1999, for an
overview), and similar intiatives have been taken in the USA and Canada (Timmerhuis
& Bringmann, 1999). At the same time, there is no clear consensus as to what the main
reasons for the current situation are.

Differential commitment to work

One possible explanation for discrepancies in the careers of men and women focuses
on motivational differences between male and female workers. A rather robust phe-
nomenon across different social contexts is that men and women perform different
 Underrepresentation of women in science 317

social roles (Eagly, 1987). Males predominantly provide the family income through
paid labour and women are primarily responsible for caretaking at home. In fact,
cross-national evidence consistently demonstrates that also when both husband
and wife perform paid labor, when both partners have equal status jobs (Biernat &
Wortman, 1991), and even when the woman is the only breadwinner (Van der Lippe,
1993), women do most of the childcare and household tasks (Van der Lippe, 1993) on
their so-called ‘second shift’ (Hochschild, 1989; Sabbadini & Camporese, 1998).
From this perspective, it has been argued that it is more difficult for women than for
men to display the total availability that is seen as an indicator for their commitment to
science (Etzkowitz, Kemelgor, Neuschatz, & Uzzi, 1992). For instance, a survey among
male and female full professors in the Netherlands (Ellemers, 1993) revealed that,
whereas virtually all male professors have a wife and family, women who have made a
successful career in science are likely to be single and childless. Thus, for women more
than for men, there generally appears to be a trade-off between the dual responsibili-
ties at home and at work, and this has been taken as an indication that paid labour
plays a less central role in the life of women than of men. Accordingly, differential
commitment to their work has been cited as a plausible reason why women are less
successful than men in their careers (Levin & Stephan, 1998).
However, previous research on gender differences in job commitment and/or job
involvement has produced contradictory results: sometimes women appear to be more
committed (Powell, Posner, & Schmidt, 1984), sometimes men are more committed
(Chusmir & Parker, 1992; Powell & Posner, 1989) and sometimes no differences occur
or the results of the different scales contradict each other (Chusmir, 1986; Graddick &
Farr, 1983; Koberg & Chusmir, 1989). Accordingly, in their review Mathieu and Zajac
(1990) conclude that there is no consistent relation between gender and the degree of
commitment, that is to say, when commitment is measured at the organizational level.

Different forms of commitment

Even when men and women report similar levels of commitment to the organization,
in view of their different social roles, it is possible that men and women do reliably
differ from each other with respect to the nature of their commitment. In previous
research, Ellemers, De Gilder, and Van den Heuvel (1998) demonstrated that people
may feel committed to different aspects of their work. In addition to affective organiza-
tional commitment, Ellemers et al. distinguished between a form of commitment
directed at individual advancement at work (career-oriented commitment) and a
form of commitment that derives from a devotion to the collaboration with one’s
co-workers (team-oriented commitment).
To the extent that women are generally more communion-oriented than men
(Ashmore, 1981; Williams & Best, 1982), this may also be expressed at work, in the
sense that female workers might be relatively willing to display behaviour that helps
their colleagues or benefits the organization as a whole (organizational citizenship,
see Schnake, 1991; Smith, Organ & Near, 1983). Indeed, it appears that women tend to
be more involved in their work and have a stronger tendency to stay in the organiz-
ation, to the extent that they have the opportunity to contribute to social relations
aspects of the organization (Chusmir & Parker, 1992; Portegijs, 1993). For men, how-
ever, the possibility to provide social contributions does not seem to be related to their
involvement or their tendency to stay. Thus, the relative importance of social work
aspects for women workers seems to suggest that they might be primarily committed
to the collaboration with their colleagues at work. Men, by contrast, are not only
318 Naomi Ellemers et al.

expected to be more individualistic, task-oriented and competitive than women, but

also tend actually to behave in this manner, at work as well as in other contexts
(Anderson & Blanchard, 1982; Bartol, 1974). As a result, male workers may tend to be
primarily committed to the most individualistic aspect of work, i.e. their own career.
If indeed women are primarily team-oriented (while men are mostly career-oriented)
and as a result consistently focus on team goals—possibly at the expense of their own
individual task achievement—this might explain why women are less likely to advance
at work. Thus, it might be that although men and women are equally committed to the
organization as a whole and devote equal amounts of effort to their work, both the
nature of commitment and the goal of their efforts might differ, with the end result that
women are less likely to advance in their career than men are.

Gender stereotypes
Another possibility that has been raised, is that gender stereotypes put female academ-
ics at a disadvantage. Gender stereotyping is an extremely pervasive phenomenon, in
the sense that gender stereotypes influence individual judgments even in situations
where a categorization in terms of gender is clearly irrelevant (Eagly, Makhijani, &
Klonski, 1992). As a result, gender stereotypes have been documented to affect
women’s opportunities for career advancement. For instance, although there is no
overall difference in the performance of women and men who actually hold leadership
positions (see the meta-analysis of Eagly, Makhijani, & Klonski, 1992), the prevailing
image of a successful leader or manager is closer to the male than the female stereo-
type (Rojahn, 1996; Schein, 1973, 1975). Consequently, experimental research has
revealed that such stereotypes may result in a preference for male candidates over
female candidates for promotion, even when judges are aware of the fact that the
applicants are equally qualified for the job (Van Vianen & Willemsen, 1992).
Not only do gender stereotypes affect the way people estimate the potential of men
and women, stereotyping may also result in a biased evaluation of actual achieve-
ments. A successful performance of women (on a typically male task) is not primarily
attributed to their ability but to chance, extraordinary effort or even cheating, whereas
men with the same achievements are simply seen as competent (Deaux, 1976, 1984;
Deaux & Emswiller, 1974; Taynor & Deaux, 1973). There are some indications that
over the years such perceived gender differences have become less pronounced, as a
result of contemporary changes in the social roles of men and women (Diekman &
Eagly, 2000). Nevertheless, recent examinations of possible explanations for differen-
tial successes of men and women in academic careers clearly show that discrimination
of women still occurs when they submit research papers (e.g. Petty, Fleming, &
Fabrigar, 1999), or apply for research grants (Brouns, 1999; Wenneras & Wold, 1997).
Likewise, a recent experimental study among academic psychologists revealed that
male as well as female judges preferred a male over a female job applicant when both
had an identical record (Steinpreis, Anders, & Ritzke, 1999).
When focusing on commitment as a central work-related characteristic, we may
assume that similar processes are likely to play a role. As we have indicated above,
perceived commitment to science is generally seen as an important requirement for
those who pursue an academic career (Etzkowitz et al., 1992). At the same time, the
shared knowledge that women are more likely than men to suffer from their dual
responsibilities at work and at home may lead people to suspect that it is more difficult
for women than for men to display the more than full-time devotion to science that is
expected from scientists. As a result, even in the absence of actual gender differences,
 Underrepresentation of women in science 319

gender stereotypic expectations may lead people to perceive female scientists as

generally less committed to (crucial aspects of) their work than their male colleagues,
which in turn may harm their career progress.

The present investigation

We have shown that women are still clearly underrepresented among university fac-
ulty. Additionally, we have argued that there may be motivational differences between
men and women, and that gender stereotypical expectations may occur. Importantly,
however, either process may explain the differential career success of male and female
academics. Therefore, the present investigation will address the question of whether
male and female scientists actually differ in their levels of work commitment, whether
they are merely perceived by others to be differentially committed, or both.
In order to examine this, we have chosen to focus on male and female scientists at
the beginning of their academic career, as this is a turning point at which the gender
ratio starts to differ substantially (Bosch et al. 1999; Rabasca, 2000), while it is difficult
to determine whether this is because females are actually less motivated than men to
pursue an academic career, or whether this is merely the way they are seen by others,
who may as a result offer less encouragement and support in achieving academic
success. Thus, we assess the (self-reported) commitment of male and female PhD
students and compare this to the way they are perceived by university faculty. Further-
more, we will examine whether differential commitment is evident from other work-
related variables, such as general work satisfaction and (self-reported) work behaviour
and achievements.
We have obtained similar data in two different national settings, the Netherlands and
Italy. Contrary to what is often assumed, in the Netherlands there is considerable
gender inequality at work. Labour participation of women is lower than in any other
European country: only 22% of the female population over 15 years of age is employed
in a full-time job (while this is 34% on average for 15 European countries; Eurostat,
2000, [Link] Likewise, the hourly wages of women as
compared to men (on an hourly basis women earn 77% of what men earn) are lower
than in any other European Union country1. An important statistic for our present
investigation is that there is no country in Europe with a smaller representation of
women among university faculty than the Netherlands, where only 5% of the full
professors and 7% of the associate professors are female (Osborn, 1998). Thus, the
Netherlands can be considered an extreme case, in the sense that it is considerably
more unusual than in other European countries, North America, or Australasia for
women to hold an academic position. The converse situation can be found in Italy.
While the overall labour participation rate of women is average for a European country
(33% of women have full-time employment), in Italy the representation of women
among university faculty is among the highest in Europe, with women constituting
11% of the full professors and 27% of the associate professors (Osborn, 1998; see also
Maass & Casotti, 2000)2. The comparison of results obtained in these two countries
therefore enables us to examine whether there is evidence of a more general process
instead of having to attribute our findings to unique characteristics that are only
relevant in a specific national context.
1
See [Link] (retrieved February 19, 2004).
2
These overall differences are reflected in the specific university populations we studied: at the Free University in
Amsterdam, the Netherlands women constitute 11% of the full and associate professors, at the University of Padua in
Italy, this was 32%.
320 Naomi Ellemers et al.

STUDY 1: THE NETHERLANDS

Method
Participants
The sample consisted of 132 doctoral students (at Time 1; of whom 48 students also
participated at Time 2) and 179 faculty members of the Free University in Amsterdam.
Doctoral students
The participants in this study were 132 doctoral students (85 men and 47 women; the
mean age was 29) who were distributed across 13 different departments (covering the
life sciences, social sciences, humanities, law and medicine) of the Free University in
Amsterdam. Of this first sample, 48 participants were willing to complete a second
questionnaire, approximately 1 year after the first study. At Time 2 the sample
consisted of 27 men and 21 women; their mean age was 30.
Faculty
The participants were 180 faculty members, employed at one of the 13 departments at
the Free University in Amsterdam. They completed one of two versions of the ques-
tionnaire depending on the experimental condition they were randomly assigned to:
one version focused on the opinions about female doctoral students, the other version
asked about male students only. In the accompanying letter it was explained that the
researchers were interested in the views of faculty members about the work commit-
ment of (female or male) doctoral students. The faculty members who participated in
the investigation consisted of 156 men and 23 women (1 respondent did not indicate
his/her gender), representing the full range of academic disciplines. Of these partici-
pants, 88 (17 women) were assistant professors (49%), 27 (2 women) were associate
professors (15%), and 65 (4 women) were full professors (36%). Their mean age was
47 (males: M = 47.5; females: M = 41.7).

Questionnaires
The first questionnaire assessed doctoral students’ self-reported levels of commitment
to the organization (the Free University), the department and their own careers.
Additionally, they were asked about their general work attitudes and the time invested
in work-related activities. On the second measurement, which was taken 1 year later,
we repeated the commitment items, to be able to assess the test–retest reliability of our
measures of career-oriented commitment and team-oriented commitment. Addition-
ally, we included two forced-choice dilemmas, in which respondents were asked to
indicate whether they would choose to devote a limited amount of time to their work
or to family obligations/recreational activities. To supplement participants’ reports on
time invested in work-related activities, in the second questionnaire we also asked
them to indicate their actual scientific output (number of publications).
The questionnaire distributed among faculty members was designed to establish
whether differences would occur in the perceived commitment of doctoral students,
depending on whether respondents focused on male or female students as a target
group.

Measures
Commitment
After a brief instruction, 32 statements were presented and participants had to indicate
the extent to which they agreed with each statement on a scale ranging from 1 (not at
 Underrepresentation of women in science 321

all) to 7 (very much). The doctoral students were asked to indicate their agreement
with each statement on this scale. The list of commitment statements was designed to
assess three forms of commitment, i.e. (affective) organizational commitment (8 items,
e.g.: ‘I feel at home at the university’), career-oriented commitment (12 items, e.g.: ‘I
want to move ahead in my job’) and team-oriented commitment (12 items, e.g.: ‘I feel
at home between my colleagues at work’). These three types of items were randomly
ordered. The questionnaire used to measure organizational commitment is a Dutch
translation (see De Gilder, Van den Heuvel, & Ellemers, 1997) of the affective commit-
ment questionnaire developed by Meyer and Allen (1991). The other items were
adapted from various existing questionnaires (e.g. Blau, 1988; Lodahl & Kejner, 1965;
Steers & Braunstein, 1976) with the aim to assess career-oriented and team-oriented
commitment respectively (see also Ellemers, et al. 1998).
This part of the questionnaire was also sent to the faculty members. They were
instructed to think of a typical male/female doctoral student (depending on the exper-
imental condition) and indicate on a 7-point scale (ranging from 1 (not at all) to 7 (very
much)) to what extent a typical male/female doctoral student would agree with each
of the commitment statements.

Work satisfaction
Work satisfaction was measured with a single item (see Wanous, Reichers, & Hudy,
1997), asking participants to indicate how they generally felt towards their work, on a
scale from 0 (very negatively) to 100 (very positively).

Time expenditure
Participants were asked to indicate how they spend their (24) hours each day in a
normal, average week. They were provided with a table in which each row repre-
sented 1 day of the week, and 5 columns represented different classes of activities to
help them complete this part of the questionnaire. In this way they were asked to
specify how many hours per week they spent on (1) work, (2) household tasks and
other chores around the house, (3) spare-time activities, (4) personal care (eating,
getting dressed, etc.), and (5) sleeping.

Demographic and context variables

In the last part of the questionnaire respondents were asked to reveal demographic
information concerning their gender, age, marital status, parenthood, and to specify
their academic status and discipline.

Choice dilemma and research output

At Time 2, we tried to collect some additional data that might reveal possible correlates
of any differences in work commitment among the doctoral students. First, in addition
to the commitment items, a dilemma situation was introduced, in order to have
participants indicate how they would spend their time when forced to choose be-
tween two different activities. The dilemma situation was described as follows: ‘You
have to choose between something at work which is very important and urgent for
you and something outside work which is also very important and urgent for you. This
will take place partly during working hours and partly in your spare time.’ Participants
were asked to indicate on a bipolar scale whether they would be more likely to spend
time on something at work (1) or on some other activity (7).
322 Naomi Ellemers et al.

Finally, to obtain an indicator of academic performance, students were asked to

report their scientific output, in terms of the number of publications they had. In order
to enhance the accuracy of these self-reports, we asked them to specify for each
publication they had prepared since starting on their PhD whether it was an article
or a book chapter, and whether they had only submitted it or whether it had been
accepted for publication.

Results
Commitment
First, a principal components analysis (with varimax rotation) was conducted, to make
sure that the three different forms of commitment also emerge in this particular
professional sample. This resulted in a three-factorial solution, with the items cluster-
ing in the intended way. The rotated factor solution explained 45% of the variance in
the individual items. Further analyses were conducted using the unweighted mean
scores for organizational commitment (α = 0.78), career-oriented commitment (α =
0.87), and team-oriented commitment (α = 0.81). The scales for career-oriented and
team-oriented commitment were also included in the questionnaire at Time 2, in order
to assess test–retest reliability. Both scales proved to assess stable orientations, result-
ing in a correlation between the measures taken at Time 1 and Time 2 of .81 (p < .001)
for career-oriented commitment, and of .73 (p < .001) team-oriented commitment.
We could not establish reliable differences between reported levels of commitment
for male and female doctoral students on any of the three subscales (organizational
commitment: male students: M = 2.99, SD = 1.19; female students: M = 3.08,
SD = 1.28, F(1,126) < 1, ns; career-oriented commitment: male students: M = 3.86,
SD = 1.33; female students: M = 4.01, SD = 1.37, F(1,126) = 1.40, ns; team-oriented
commitment: male students: M = 4.69, SD = 1.11; female students: M = 4.73,
SD = 1.11, F(1,126) < 1, ns) at Time 1. Also at Time 2, male and female doctoral
students indicated they were equally committed to their own career (α = 0.85; males:
M = 3.94; females: M = 4.17; F < 1, ns) and to the team in which they worked
(α = 0.73; males: M = 4.65; females: M = 4.68; F < 1, ns).

Work satisfaction
No gender difference was observed in the mean levels of work satisfaction reported by
male (M = 71.08, SD = 15.28) and female (M = 67.42, SD = 19.90) doctoral students
(F(1,126) = 1.50, ns).

Behavioural indicators

Time expenditure
We examined how student gender affected the number of hours per week spent on
the different activities specified in the questionnaire. This reveals an overall difference
between male and female students, in terms of the number of hours spent on work,
F(1,104) = 4.32, p < .05, and on household tasks, F(1,104) = 8.29, p < .005. Male
respondents generally indicate that they spend more hours per week on work
(M = 51.08) than do female respondents (M = 47.00), whereas female students report a
greater number of hours spent on household activities (M = 15.63) than their male
colleagues (M = 10.94).
 Underrepresentation of women in science 323

Choice dilemma
When asked to indicate how they would decide to spend their time when forced to
choose between two different activities, male respondents appear relatively even-
handed in their preferences (M = 4.14) while women are more inclined to opt for
some other activity instead of work (M = 5.35, p < .002). Correlational analyses
revealed that respondents’ choice for a work-related activity was positively related to
their level of career-oriented commitment, for male (r = .39) as well as female (r = .63)
respondents.

Publications
Although the mean number of publications seems to be lower for female (M = 2.8)
than for male respondents (M = 4.3), no reliable gender difference was established
between the total amount of (in press) publications reported by the male and female
PhD students (t = 1.08, ns). This may be due to the small number of doctoral students
that completed the questionnaire at Time 2.

Faculty: Perceived commitment

First, a principal components analysis (with varimax rotation) was conducted, to
establish whether perceptions of commitment revealed the same three factors that we
distinguished in self-reported levels of commitment. The results of this analysis confirm
that the three intended factors could be distinguished, with the rotated factor solution
accounting for 58% of the variance in the individual items. Further analyses were
conducted using the unweighted mean scores for organizational commitment
(α = 0.68), career-oriented commitment (α = 0.87), and team-oriented commitment
(α = 0.83).
We conducted 2 (faculty gender) × 2 (target gender) ANOVAs to examine any
gender differences in the perceived commitment of male and female doctoral students.
This revealed a main effect of target gender on career-oriented commitment (F(1,164)
= 15.13, p < .001), however this was qualified by an interaction effect between faculty
gender and target gender (F(1,165) = 4.96, p < .05). The relevant means and post hoc
contrasts indicate that female faculty members believe that there is a reliable difference
between commitment levels of male and female doctoral students, while no such
difference in perceived commitment depending on target gender emerges in the
ratings of male faculty. That is, women, but not men, tend to assume that female
doctoral students are less committed to a scientific career than their male colleagues
(see Table 1). A similar interaction and means pattern was obtained for perceived
levels of organizational commitment (F(1,165) = 4.17, p < .05), although the post hoc
contrasts did not reach significance for this variable.

Discussion
Building on our previous work (Ellemers et al., 1998) a first objective of this investi-
gation was to establish whether gender differences in work motivation would emerge
more clearly when distinguishing between specific forms of commitment, rather than
relying on a general measure of affective organizational commitment, as is usually the
case in research on this topic. In line with results obtained among other professional
samples (Ellemers et al., 1998) we were able to assess career-oriented commitment
and team-oriented commitment as relatively stable constructs, separate from general
324 Naomi Ellemers et al.

organizational commitment. Furthermore, in view of the higher mean commitment

levels reported, it would seem that—for this sample of academic professionals—these
specific forms of commitment may be more relevant measures of work motivation than
the widely used measure of affective organizational commitment.
However, contrary to what might be expected, at both points in time, male and
female doctoral students reported similar levels of team-oriented as well as career-
oriented commitment. We could observe no difference in the average extent to which
male and female doctoral students feel committed to various aspects of their work.
While a similar distinction between the three forms of commitment emerged in
faculty ratings of the doctoral students, the nature of their perceptions differed from
the doctoral students’ self-reports. In line with prevailing stereotypes, overall, women
are seen as less committed to their careers than men (even though target gender was
manipulated between participants). In order to explain this finding, we might look at
the results of other measures. In line with our general argument and the results of
previous investigations, the self-reported time expenditure of the students as well as
participants’ responses to the choice dilemma attest to the fact that—also among this
sample of young academic professionals—women seem to suffer more from the dual
tasks at home and at work than men do. Indeed, in the trade-off between work and
home-related responsibilities, the female respondents indicate that they spend more
hours on household tasks and less hours on work than do their male counterparts.
Perhaps this difference in time expenditure constitutes the observational basis that
causes the faculty to believe that female students are less committed to the university
and to their academic career. However, to the extent that this perceived difference
in commitment to work might be a reason for faculty to be more encouraging and
offer more career opportunities to male than female PhDs, it is important to establish
whether the difference in time allocated to work and home responsibilities actually
results in a difference in academic performance. Now the only indicator of this that we
included in the present study is the number of (in press) publications reported by the
students. Here we see no significant difference between male and female students.
However, the mean level of publications seems higher for male than female respond-
ents, and the fact that this did not emerge as a reliable difference might indicate a lack
 Underrepresentation of women in science 325

of statistical power, due to the relatively small number of students that participated at
Time 2 and completed this measure. Therefore, additional data are necessary to
establish whether or not there is an actual gender difference in the academic
performance of PhD students.
It is striking that female faculty instead of male faculty most clearly differentiates
according to gender in their perceptions of doctoral students.3 While there is a per-
ceived gender difference in career-oriented commitment in the eyes of the female
faculty, the male faculty perceives male and female students as equally committed to
their careers. A similar pattern emerges with respect to the perceived commitment of
male and female doctoral students to the university, suggesting that only female faculty
assumes that female doctoral students are less committed to their work than their male
counterparts. Why?
It is a common belief that males are more inclined than females to endorse gender
stereotypes (e.g. Brenner, Tomkiewicz, & Schein, 1989). Indeed this has been taken as
a point of departure to explain the underrepresentation of women at higher levels in
organizations from the fact that men traditionally have been more likely than women
to make decisions about hiring and promotion of personnel (e.g. Levin & Stephan,
1998). Nevertheless, the literature also shows that sometimes women are more likely
than men to rate other women negatively (Mathison, 1986). Previous attempts to
explain why it is that women seem sometimes more biased and sometimes less biased
than men have argued that this may have to do with differences in personal attitudes
(Cooper, 1997), or differences in the social roles they occupy (Eagly, 1987; see also
Pratto, Stallworth, Sidanius, & Siers, 1997). Indeed, it has been demonstrated that
women are less likely to support policy measures that would be beneficial for women
as a group, to the extent that they hold a competitive attitude towards other women
(Cowan, Neighbors, DeLa Moreaux, & Behnke, 1998).
The term queen bee syndrome was coined to describe the general phenomenon
that in particular women who have been individually successful in male-dominated
environments are likely to oppose the women’s movement (Staines, Tavris, &
Jayaratne, 1974). It has been argued that women who have risen to the managerial
level in a male-dominated work organization may oppose such more systemic changes
as they should be motivated to maintain the organizational culture in which they have
fared so well (Gibson & Cordova, 1999). Such behaviour may be understood in a more
theoretical sense, from a social identity point of view, i.e. to the extent that women in
male-dominated work environments are different from women in general, they can be
seen as pursuing some form of individual mobility (Tajfel, 1978; Tajfel & Turner, 1979).
A variety of cognitive, affective and behavioural responses has been associated with
the use of individual mobility strategies (Branscombe & Ellemers, 1998), all having to
do with a emphasis on intra-group differences rather than similarities, which enables
individual group members to contrast the self with the rest of the group. Indeed, in a
previous study among full professors at different universitities in the Netherlands,

3
In seeming contrast with our findings, Brown and Smith (1989) observed that female faculty tended to view the
performance of their female colleagues more favourably than that of their male colleagues. However, the work of Brown
and Smith differs from our study in several important ways. In their study, the female faculty favourably judged their
female colleagues while the target of their ratings (a) like themselves had already established a successful academic career
(whereas we asked faculty to rate PhD students), (b) actually performed better than the male faculty on objective
indicators (while we had no reliable evidence of performance differences), and (c) comprised a group the judges were likely
to identify with (while we have reason to assume that our female faculty disidentified with the female target group).
326 Naomi Ellemers et al.

Ellemers (1993) obtained more masculine (and less feminine) self-descriptions from
female professors than from their male colleagues, indicating a perception of self as a
non-prototypical (gender) group member. Similar processes may then account for our
current finding that female faculty members are relatively inclined to view other
members of their gender group in gender-stereotypical terms.
Although this pattern of results can be explained from a social identity perspective,
and seems to fit in with previous related findings, it is based on observations among a
relatively small number of female faculty. Furthermore, our argument is based on the
assumption that the greater tendency of female faculty to hold gender-stereotypical
perceptions of PhD students has to do with their own careers and the context in
which these were made. Therefore, we have conducted a second study, in order to
examine whether these findings can be replicated with a different sample, and to
obtain some additional data that may shed light on the validity of our theoretical
account of these findings.

STUDY 2: ITALY
A second study was conducted to examine more closely the most striking finding
obtained in Study 1, namely that female faculty is more likely than male faculty to
endorse gender stereotypical perceptions of doctoral students. First, we assess
whether this is unique to the Dutch national context or whether it represents a more
general phenomenon, that can also be observed in another national context. Secondly,
we aim to investigate more specifically whether our post hoc explanation of the
pattern observed in Study 1 is valid.
The social identity analysis of our findings revolves around the assumption that
people’s own career efforts and the way they relate to their group are a crucial
determinant of the way they view others around them. Empirical support for this
position may be derived from a study by Tougas, Brown, Beaton, and St Pierre (1999),
who focused on women who were frustrated in their attempts to advance at work, and
showed that those who fail to gain access to male-dominated jobs are inclined to
endorse egalitarian views and favour affirmative action programmes at work. We look
at the other side of the coin and argue that women who have been individually
successful in a traditionally male field of work would likely see themselves as different
from ‘regular’ women. As a result, they are inclined to maintain stereotypical views of
men and women, while they perceive themselves as non-prototypical group members.
Converse evidence that people’s own experiences in gaining their current position
determine the way they view other group members can be obtained from the work of
Fajak and Haslam (1998). In their studies, the experimenter assigned male and female
research participants to a certain level in a simulated organization. Under these con-
ditions, i.e. when they did not have to demonstrate that they qualified for a powerful
position despite their gender, female research participants tended to favour other
women in promotion decisions to redress the underrepresentation of women at higher
organizational levels.
Thus previous work shows that there are circumstances under which women will
perceive themselves as representative of their female colleagues and try to help them
to advance at work. Notably, this is the case when one’s own attempts at upward
mobility have been unsuccessful (as in the study by Tougas et al., 1999), or when
achievement of a powerful position was not contingent on one’s gender identity (as
was the case in the study by Fajak & Haslam, 1998). By contrast, our hypothesis refers
 Underrepresentation of women in science 327

to women who have been individually successful in a male dominated setting. In this
sense, it is relevant to a class of situations that differs fundamentally from the contexts
examined in previous research. Hence, we may still expect to find evidence of the
queen bee syndrome, in the sense that the very strategies individual women employ to
achieve career success despite their gender are likely to cause them to discriminate
against other women.
As we have indicated previously, in Italy women currently are relatively well repre-
sented on the labour market in general and among university faculty in particular.
Thus, in Italy, the older generation of female academics, who made their career when
this was still exceptional for women, is most likely to emphasize the ways in which
they are different from other women (a similar argument has been proposed by
Etzkowitz, Kemelgor, Neuschatz, Uzzi, & Alonzo, 1994). However, there is reason to
assume that younger women have encountered less career difficulties and hence have
a different view of themselves as well as their female colleagues. Based on our assump-
tion that one’s own individual mobility experiences are a crucial determinant of the
way one views others, we therefore surmise that younger faculty members are less
likely to differentiate themselves from other women, while the older generation, who
arguably have experienced more difficulties in achieving individual career success,
should be most likely to replicate the pattern of discrimination against other women
that we obtained in the Netherlands.
To the extent that biases against female PhD students stem from the process of
setting the self apart from women as a group, we should find that those women on the
faculty who endorse gender stereotypical views are most inclined to perceive them-
selves as non-prototypical group members. As a result, they should describe them-
selves in relatively masculine terms. To examine the validity of this argument, in the
second study Bem’s (1974) Sex Role Inventory was added to the faculty questionnaire.
This enables us to examine whether gender stereotyping among male and female
faculty is indeed related to the content of their own gender identity.

Method
Participants
The sample consisted of 80 doctoral students and 93 faculty members from different
departments at the University of Padua.

Doctoral students
The participants were 39 male and 41 female doctoral students at Padua University
their mean age was 28. They worked in 26 different departments across the university
and were therefore roughly representative of the doctoral student population of the
University of Padua.

Faculty members
The participants in this study were 52 male and 41 female faculty members (mean age
45) who were working in 14 different departments of the university. Due to the
inferior number of women among university faculty, female faculty members were
contacted first and, where possible, an equal number of male faculty of the corre-
sponding rank in the same department were recruited subsequently. In order to allow
comparisons across rank and age groups, a roughly equal number of assistant
328 Naomi Ellemers et al.

(ricercatori: 30%), associate (associati: 34%) and full professors (professori di prima
fascia: 36%) were included in the study.

Questionnaires
The questionnaires were virtually identical to those used in Study 1 conducted in the
Netherlands except for the inclusion of the Bem Sex Role Inventory (BSRI) for the
faculty sample, to assess whether they perceived the self predominantly in masculine
or feminine terms.

Doctoral students
The questionnaire of the doctoral students first comprised the scale measuring career-
oriented commitment organizational commitment, and team-oriented commitment.
Subsequently, the students were asked to complete the measure of work satisfaction,
the measure of time expenditure specifying different activities, and were also asked to
indicate their number of publications (which was only assessed at Time 2 in Study 1).

Faculty members
The questionnaire was similar to the one used in Study 1. Each faculty member was
randomly asked either to rate typical female doctoral students or typical male doctoral
students. Thus, the questionnaire consisted of a scale assessing perceived levels of
career-oriented, organizational and team-oriented commitment among doctoral stu-
dents (either male of female, depending on the experimental condition). Additionally,
in this study the questionnaire included the BSRI to assess the extent to which faculty
members perceived themselves in masculine of feminine terms (content of gender
identity). The BSRI was administered in its short form consisting of 40 items half of
which representing typically masculine (e.g. assertive), half typically feminine traits
(e.g. sensitive). Participants had to describe themselves using a 7-point scale (from 1 =
not at all applicable to 7 = extremely applicable). First, two scores were derived from
this measure: a masculinity score and a femininity score. Then, in order to estimate the
relative prevalence of masculine vs. feminine traits, the difference between the average
ratings on masculine and feminine traits was calculated.

Results
Doctoral students
Self-reported commitment
The commitment scales had satisfactory internal consistency. Cronbach’s alpha was
0.85 for career-oriented commitment, 0.75 for organizational commitment, and 0.80
for team-oriented commitment. Statistical comparisons between the levels of commit-
ment reported by male and female students only revealed a marginally significant
difference for team-oriented commitment, with women being more committed to the
colleagues in their department (M = 4.66, SD = 1.01) than men (M = 4.21, SD = 1.05,
F(1,78) = 3.76, p < .06). No gender difference was observed for self-reported commit-
ment to one’s career (females: M = 4.14, SD = 1.05; males: M = 3.92, SD = 1.07,
F(1,78)< 1, ns), or to the university organization (females: M = 3.79, SD = 1.01; males:
M = 3.87, SD = 1.05, F(1,78) < 1, ns).

Work satisfaction
Again, males (M = 72.34, SD = 19.28) and females (M = 73.61, SD = 18.81) reported to
be equally satisfied with their work (F(1,77) < 1, ns).
 Underrepresentation of women in science 329

Time expenditure and scientific output

In this study, male and female doctoral students reported similar time expenditure on
different activities. Although the mean number of (self-reported) hours spent at work
seems slightly larger for males (M = 49.31) than for females (M = 46.59), this difference
is not significant. Additionally, and consistent with the findings in the Netherlands,
male and female doctoral students showed a very similar publication record. The
average number of publications in international journals was practically identical
(males: M = 1.51, females: M = 1.56), as was the number of book chapters (males:
M = .20, females: M = .19), and the number of conference papers (males: M = 1.84,
females: M = 1.77). The only difference emerged for national journals (males: M = 1.41,
females: M = .46, p < .05) that, however, tend to have rather low impact and recogni-
tion. Taken together and in line with the results regarding self-reported commitment,
these indicators provide very little evidence for a reliable gender differences in
work-related effort.

Faculty members
Perceived career-oriented commitment of doctoral students
In order to see whether male and female faculty members perceived male and female
doctoral students as differentially committed to their careers, a 2 (faculty gender) × 2
(target gender) between-participants ANOVA was run. The results revealed a main
effect for target gender, F(1,85) = 6.60, p < .01, indicating that male students were
perceived as more committed to their careers (M = 5.19) than female students
(M = 4.65). Although there was no significant interaction with faculty gender (F(1,89)
= 1.27, ns), separate comparisons for males and females revealed the predicted pattern
that was also observed in Study 1. Female faculty members perceived male doctoral
students (M = 5.41) as reliably more dedicated to their careers than female students
(M = 4.62), t(41) = 2.15, p < .05, while male faculty members did not rate male and
female students as significantly different in this respect (M = 5.00 for males and
M = 4.67 for females, ns, see Table 2).

Intergenerational differences in gender stereotyping

We have argued that to the extent that gender stereotyping is caused by one’s own
career experiences, there might be a difference between older and younger gener-
ations of female faculty. The faculty sample was therefore split into two age groups at
330 Naomi Ellemers et al.

the median (above or below 47 years of age) to investigate whether gender stereotyp-
ing is indeed less pronounced among the younger generation. Generation was there-
fore added as an independent variable, resulting in a 2 (faculty gender) × 2 (generation)
× 2 (target gender) between-participants ANOVA. In addition to the main effect of
target gender described above, this analysis revealed a main effect for generation,
F(1,84) = 17.23, p < .001, indicating that older faculty members (M = 5.35) generally
perceived doctoral students (regardless of target gender) as more committed than did
their younger colleagues (M = 4.52), as well as a marginally significant main effect of
faculty gender (F(1, 84) = 3.48, p < .07), with female faculty seeing students as more
committed to their career (M = 5.12) than male faculty (M = 4.75). More relevant to
our argument is the interplay between these effects of target gender, faculty generation
and faculty gender, even though the three-way interaction was not significant. As can
be seen in Table 3, the only subgroup that perceived male doctoral students as
significantly more committed to their careers than female students were the older
female faculty. Male faculty of both age groups and younger female faculty did not
differentiate in their ratings of male and female students.

Gender identity
Based on social identity theory and previous findings with regard to female professors
in the Netherlands (Ellemers, 1993), we argued that—particularly in the early days—
female faculty, in order to be successful in a male-dominated work environment, had to
have a particularly masculine self-perception. This involves emphasizing differences
rather than similarities with other women, which might explain the bias of female
faculty against female graduate students. In order to assess whether further empirical
evidence could be obtained for this line of reasoning, we examined the content of
the gender identity among our faculty sample (operationalized as self-attributed male
characteristics minus self-attributed female characteristics). First, in contrast to obser-
vations among other samples of men and women (e.g. college students, Holt & Ellis,
1998), we obtained no reliable overall difference in gender identity content between
male and female faculty (F(1,88) = 1.64, ns). Even though the interaction between
faculty gender and generation was not significant (probably due to limited statistical
power), in view of our predictions and previous observations, we investigated the
extent to which this lack of difference in gender identity could be traced to the older
 Underrepresentation of women in science 331

and younger generations of men and women in our faculty sample. The pattern that
emerges (see Table 4) shows that only the older generation of female faculty primarily
describe themselves in masculine terms, i.e. their self-descriptions are not reliably
different from those of their male colleagues. However, among the younger generation
we find a more prototypical pattern, with women clearly holding a less masculine
self-description than men. While this observation deviates from previous research
showing that older women are generally more clearly sex-typed than younger women
(Moore & Rosenthal, 1980; Twenge, 1997), it is consistent with our argument that the
career difficulties encountered by the older generation of female faculty should result
in a perception of the self as a non-prototypical group member. Indeed, only in this
subgroup of our female faculty sample that reports a relatively masculine gender
identity do we find evidence of stereotyping against other women.

Discussion
As was the case in the first study, the self-reports of the doctoral students mainly reveal
similarities rather than differences between males and females. In this sample, male
and female doctoral students do not reliably differ in self-reported career-oriented
commitment, organizational commitment, time spent at work, or scientific output. The
only difference that emerged indicates that female doctoral students are more commit-
ted than their male counterparts in that women report somewhat higher levels of
team-oriented commitment.
In their ratings of the doctoral students, male faculty reflect these self-reports most
accurately: they do not perceive male and female students as differentially committed
to their work at the university. However, as was the case in the Dutch sample, female
faculty do differentiate between doctoral students according to their gender, by rating
female students as less career-oriented than male students. The inclusion of some
additional measures in this study allows us to shed some more light on the reasons why
this might be the case.
A closer look at the data revealed that the bias against female doctoral students is
limited to the older generation, i.e. to female faculty with an age above the median (47
years). These women (born between 1921 and 1949) decided to pursue their own
careers at a time when this was still exceptional for women. Consistent with our
argument that making a successful career under these circumstances involves setting
the self apart from one’s gender group, there is indeed evidence that these faculty
members regard themselves as non-prototypical women. Whereas younger colleagues
describe themselves as reliably less masculine than their male comparison group (see
also Twenge, 1997), the self-descriptions of older female faculty are just as masculine
332 Naomi Ellemers et al.

as those of their male colleagues. In other words, these women have adopted a
masculine self-stereotype that emphasizes how they differ from more prototypical
members of their gender group (see also Moore & Rosenthal, 1980).
Thus, these data are consistent with our theoretical argument that when it is
exceptional for women to pursue an academic career, those who are successful in
doing so perceive themselves as non-prototypical members of their gender group.
Indeed the parallel effects we have observed with respect to age of female faculty,
masculinity of self-description and perceived commitment in female students suggest
that this heightened awareness of differences (rather than similarities) between the self
and the gender in-group not only involves adopting a masculine self-image, but can also
imply that other women are contrasted with this self-view, and as a result are per-
ceived in gender stereotypical terms. While, due to the relatively small number of older
female faculty in our sample we cannot establish statistically reliable evidence of
mediation, the pattern we observe is intriguing, is consistent with theoretical notions
presented here and clearly deserves further research.

GENERAL DISCUSSION
This investigation set out to examine possible reasons why women continue to be
underrepresented in science. Two alternative hypotheses were advanced in the intro-
duction: (1) due to their dual responsibilities at home and at work women might be
less committed to their work, hence they invest less effort and are less successful than
their male colleagues, or (2) due to gender stereotyping women may be perceived as
less committed and hence seem less suitable for a scientific career to those who have
to select and encourage young academics. Furthermore, when focusing on commit-
ment as an indicator of one’s devotion to work, we distinguished different forms of
commitment and argued that actual or perceived gender differences in career-oriented
commitment might be most detrimental to the academic careers of women. We exam-
ined these hypotheses with data collected in two different national contexts, which
differ from each other in terms of the representation of women in the general labour
force, as well as in the university.
In neither of the two studies did we find evidence that female doctoral students are
less committed to their work than their male colleagues. Male and female students
generally reported being equally committed to different work aspects. The only excep-
tion to this pattern was found in the Italian sample, where female students reported
somewhat greater team-oriented commitment than male students. While this is consist-
ent with the overall notion that women are generally more community oriented than
men, this in itself does not explain a gender difference in career success, as there is no
evidence that as a result female students are less committed to their careers or to the
university than male students.
In the Dutch sample (Study 1), female doctoral students indicate that they spend less
time on work and more time on household tasks than do male doctoral students.
However, when we look at the resulting scientific performance, there is no clear
evidence of a gender difference in the output indicators that are available. This is
consistent with findings from other studies, examining the productivity of male and
female researchers. First, it has been demonstrated that there is no one-to-one relation
between the amount of time available for work and scientific output. On the contrary:
there is evidence that female scientists who are married and have children produce
just as much or are even more productive than single females (ETAN-report, 1999).
 Underrepresentation of women in science 333

Furthermore, when scientific output is not only measured in terms of number of

publications but is corrected for journal impact, gender differences in productivity
tend to disappear (ETAN-report, 1999; Maass & Casotti, 2000; Van Vianen, Ottens, &
Van Schie, 1997).
While the self-reports in both samples reveal few, if any, differences in indicators of
work motivation and scientific performance among male and female doctoral students,
faculty members in the Dutch as well as the Italian sample perceived female students
as less committed to their careers than male students. Thus, taken together, the results
of these studies offer more evidence for a stereotyping explanation for the fact that so
few women have been successful in an academic career, than for an account based on
differential work commitment of male and female researchers. Moreover, unlike what
is often assumed, it seems that men are not the main culprit for the occurrence of
gender stereotypical views among university faculty. By contrast, it turns out that
female faculty members are most inclined to hold stereotypical views of the work
commitment of male and female doctoral students.
As we have argued when discussing the results from the Dutch sample (where the
underrepresentation of women among university faculty is extreme compared to
other countries), survival of women in a male-dominated work environment entails a
form of invidual mobility, in the sense that they have to prove to themselves and
others that they are unlike other women in order to be successful in an academic
career. When discussing the findings of Study 1 we referred to a previous study
(Ellemers, 1993) revealing that, overall, female full professors in the Netherlands have a
particularly masculine self-image, which would be consistent with our theoretical
argument.
In the Italian sample we replicate this pattern and have additional data that further
corroborates our argument, as we found that the older generation of female faculty, for
whom pursuing an academic career was most exceptional, both describe themselves
as non-prototypical group members and hold stereotypical views of other women at
the university. While the small number of observations among this particular group of
research participants does not allow us to demonstrate mediation, parallel effects
emerge on different variables, in the sense that older female faculty are the ones who
(1) show biased judgments of female doctoral students, and (2) report a masculine
self-image. By contrast, for the younger generation of women in Italy the pursuit of an
academic career is less extraordinary, as is evident from the relevant statistics. Hence,
they have less reason to accentuate the differences between themselves and other
women. As a result, they continue to describe themselves in relatively feminine terms
and hold unbiased judgments of other women in the university.
Thus, stereotyping of women rather than differential work commitment emerges as
a plausible reason that women have more difficulty than men to be successful in an
academic career and—because of their own precarious position—women are more
likely than men to engage in gender stereotyping in this context. Importantly, when
women hold biased evaluations of other women, this is usually not recognized as a
form of gender discrimination (Baron, Burgess, & Kao, 1991; see also Petty et al.,
1999), but taken at face value instead. This is even more likely to be the case when the
merit of individual men and women is being assessed (as is often the case in academic
settings), so that the detection of any group-based prejudice is impaired (see also
Schmitt, Ellemers, & Branscombe, 2003). Thus, when (biased) judgments of senior
women about the potential of their junior colleagues are not questioned, other women
may suffer from this. Indeed recent research (Barreto & Ellemers, in press) reveals that
334 Naomi Ellemers et al.

women suffer most when they encounter gender sexism by other women, as their
difficulty in recognizing this as a form of gender discrimination impairs the use of
adequate coping responses.
This may lead to the provocative conclusion that recent measures intended to
prevent biases against women may help perpetuate them. Specifically, involving senior
female scientists in supervision and review procedures may harm rather than help
the cause it is intended to serve, as chances are that this eventually results in the
provision of less rather than more encouragement and opportunities to young female
scientists. Thus, paradoxically, in this way the very people who have successfully
escaped their plight as members of a disadvantaged group may be the ones who
legitimate and perpetuate the status quo in terms of overall social relations (see also
Ellemers, 2001).
On a more positive note, however, the younger generation of faculty in the Italian
sample seems to suffer less from such biased judgments. This suggests that there might
be a critical point at which it becomes easier for women to make a career while
maintaining their femininity. Thus, when it no longer seems necesssary to distance
oneself from other gender group members in order to prove one can be successful at
work, this may prevent gender stereotypes from affecting the career opportunities of
women at the university.

References
Anderson, L. R., & Blanchard, P. N. (1982). Sex differences in task and socio-emotional behavior.
Basic and Applied Social Psychology, 3, 109–139.
Ashmore, H. C. (1981). Sex stereotypes and implicit personality theory. In D. L. Hamilton (Ed.),
Cognitive processes in stereotyping and intergroup behavior (pp. 37–82). Hillsdale, NJ:
Erlbaum.
Baron, R. S., Burgess, M. L., & Kao, C. F. (1991). Detecting and labeling prejudice: Do female
perpetrators go undetected? Personality and Social Psychology Bulletin, 17(2), 115–123.
Barreto, M., & Ellemers, N. (in press). The perils of political correctness: Responses of men and
women to old-fashioned and modern sexism. Social Psychology Quarterly.
Bartol, K. M. (1974). Male and female leaders and small work groups. East Lansing, MI:
Michigan State University.
Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting and
Clinical Psychology, 42, 155–162.
Biernat, M., & Wortman, C. B. (1991). Sharing of home responsibilities between professionally
employed women and their husbands. Journal of Personality and Social Psychology, 60,
844–860.
Blau, G. J. (1988). Further exploring the meaning and measurement of career commitment.
Journal of Vocational Behavior, 32, 284–297.
Bosch, M., Hoving, I., & Wekker, G. (1999). In het hart van de wetenschap: Naar total
E-quality en diversiteit in de universiteit [At the heart of science: Towards total E-quality
and diversity at the university]. AWT-achtergrondstudie nr. 15. Den Haag: SDU.
Branscombe, N. R., & Ellemers, N. (1998). Coping with group-based discrimination: Individualis-
tic versus group-level strategies. In J. K. Swim & C. Stangor (Eds.), Prejudice: The target’s
perspective. (pp. 243–266) New York: Academic Press.
Brenner, O.C., Tomkiewicz, J., & Schein, V.E. (1989). The relationship between sex role stereo-
types and requisite management characteristics revisited. Academy of Management
Journal, 32, 662–669.
Brouns, M.L.M. (1999). De kwaliteit van het oordeel: De rol van sekse in de toekenning van
KNAW/NWO subsidies [The quality of judgment: The role of gender in the awarding of
 Underrepresentation of women in science 335

grants by the Dutch Royal Academy of Sciences and the National Science Foundation].
Utrecht: Nederlands Genootschap Vrouwenstudies.
Brown, R., & Smith, A. (1989). Perceptions of and by minority groups: The case of women in
academia. European Journal of Social Psychology, 19, 61–75.
Chusmir, L. H. (1986). Gender differences in variables affecting job commitment among men
and women. Journal of Personality and Social Psychology, 126, 87–94.
Chusmir, L. H., & Parker, B. (1992). Success strivings and their relationship to affective work
behaviors: Gender differences. Journal of Social Psychology, 132, 87–99.
Cooper, V. W. (1997). Homophily or the queen bee syndrome: Female evaluation of female
leadership. Small Group Research, 28, 483–499.
Cowan, G., Neighbors, C., DeLa Moreaux, J., & Behnke, C. (1998). Women’s hostility toward
women. Psychology of Women Quarterly, 22, 267–284.
Crum, B., & Bal, J. (1998). Werk- en loopbaanpositie van postdocs [Employment and career
prospects of postdoctoral researchers]. Leiden: Research voor Beleid.
Deaux, K. (1976). The behavior of women and men. Monterey, CA: Brooks/Cole.
Deaux, K. (1984). From individual differences to social categories: Analysis of a decade’s
research on gender. American Psychologist, 39, 105–116.
Deaux, K., & Emswiller, T. (1974). Explanations of successful performance on sex-linked
tasks: What is skill for the male is luck for the female. Journal of Personality and Social
Psychology, 29, 80–85.
Diekman, A. B., & Eagly, A. H. (2000). Stereotypes as dynamic constructs: Women and
men of the past, present, and future. Personality and Social Psychology Bulletin, 26,
1171–1188.
De Gilder, D., Van den Heuvel, H., & Ellemers, N. (1997). Het drie-componentenmodel van
commitment. Gedrag en Organisatie, 10, 95–106.
Eagly, A. H. (1987). Sex differences in social behavior: A social-role interpretation. Hillsdale,
NJ: Erlbaum.
Eagly, A. H., Makhijani, M. G., & Klonski, B. G. (1992). Gender and the evaluations of leaders: A
meta-analysis. Psychological Bulletin, 111, 3–22.
Ellemers, N. (1993). Sociale identiteit en sekse: Het dilemma van succesvolle vrouwen [Social
identity and gender: The dilemma of successful women]. Tijdschrift voor Vrouwenstudies,
14, 322–336.
Ellemers, N. (1995). Psychologie: Tussen vrouwenstudie en mannenvak [Psychology: Between
women’s major subject and men’s profession]. Nederlands Tijdschrift voor Psychologie en
haar Grensgebieden, 50, 224–230.
Ellemers, N. (2001). Individual upward mobility and the perceived legitimacy of intergroup
relations. In J. T. Jost, & B. Major (Eds.), The psychology of legitimacy. Cambridge: Cam-
bridge University Press.
Ellemers, N., De Gilder, D., & Van den Heuvel, H. (1998). Career-oriented versus team-oriented
commitment and behavior at work. Journal of Applied Psychology, 83, 683–692.
ETAN-report (1999). Women and science: Mobilising women to enrich European research.
Brussels: European Commission/European Technology Assessment Network.
Etzkowitz, H., Kemelgor, C., Neuschatz, M., & Uzzi, B. (1992). Athena unbound: Barriers to
women in academic science and engineering. Science and Public Policy, 19, 157–179.
Etzkowitz, H., Kemelgor, C., Neuschatz, M., Uzzi, B., & Alonzo, J. (1994). The paradox of critical
mass for women in science. Science, 266, 51–54.
Fajak, A., & Haslam, S.A. (1998). Gender solidarity in hierarchical organizations. British Journal
of Social Psychology, 37, 73–94.
Gibson, D. E., & Cordova, D. I. (1999). Women’s and men’s role models: The importance of
exemplars. In A. J. Murress, F. J. Crosby, & R. J. Ely (Eds.), Mentoring dilemmas (pp.
121–142). Mahwah, NJ: Erlbaum.
Graddick, M. M., & Farr, J. L. (1983). Professionals in scientific disciplines – sex-related differ-
ences in working life commitments. Journal of Applied Psychology, 68, 641–645.
336 Naomi Ellemers et al.

Hawkins, A., & Noordenbos, G. (1991). The ‘Matthew effect’ in the relative position of women
in academe. Universiteit en Hogeschool, 38, 124–135.
Hochschild, A. (1989). The second shift. New York: Avon Books.
Holt, C. L., & Ellis, J. B. (1998). Assessing the current validity of the Bem Sex Role Inventory. Sex
Roles, 39, 929–941.
Koberg, C. S., & Chusmir, L. H. (1989). Relationship between sex-role conflict and work-
related variables – gender and hierarchical differences. Journal of Social Psychology, 129,
779–791.
Lawler, A. (1999). Tenured women battle to make it less lonely at the top. Science, 286,
1272–1278.
Levin, S. G., & Stephan, P. E. (1998). Gender differences in the rewards to publishing in
academe: Science in the 1970s. Sex Roles, 11/12, 1049–1064.
Lie, S. S., Malik, L., & Harris, D. (Eds.) (1994). The gender gap in higher education; World
yearbook of education. London: Taylor & Francis.
Lodahl, T. M., & Kejner, M. (1965). The definition and measurement of job involvement. Journal
of Applied Psychology, 49, 24–33.
Lyness, K. S., & Thompson, D. E. (2000). Climbing the corporate ladder: Do female and male
executives follow the same route? Journal of Applied Psychology, 85, 86–101.
Maass, A., & Casotti, P. (2000). Gender gaps in EAESP: Numerical distribution and scientific
productivity of women and men. European Bulletin of Social Psychology, 12, 14–31.
Massenburg, Graddick M. M., & Farr, J. L. (1983). Professionals in scientific disciplines:
Sex-related differences in working life commitments. Journal of Applied Psychology, 68,
641–645.
Mathieu, J. E., & Zajac, D. (1990). A review and meta-analysis of the antecedents, correlates, and
consequences of organizational commitment. Psychological Bulletin, 108, 171–194.
Mathison, D. L. (1986). Sex differences in the perception of assertiveness among female man-
agers. Journal of Social Psychology, 126, 599–606.
Meyer, J. P., & Allen, N. J. (1991). A three-component conceptualization of organizational
commitment. Human Resource Management Review, 1, 61–89.
Moore, S. M., & Rosenthal, D. A. (1980). Gender, generation, and self-esteem. Australian
Psychologist, 15, 467–477.
Osborn, M. (1998). The situation of women in scientific research in European Member States.
Resources of Higher Education Institutions 1995/1996, Cheltenham: HESA Services.
Petty, R. E., Fleming, M. A., & Fabrigar, L. R. (1999). The review process at PSPB: Correlates of
interreviewer agreement and manuscript acceptance. Personality and Social Psychology
Bulletin, 25, 188–203.
Portegijs, W. (1993). Jammer dat u gaat. Het snelle vertrek van vrouwelijke wetenschappers:
Een lek in het emancipatiebeleid [It’s a pity that you are leaving. The rapid turnover of
female scientists: A leakage in positive action policy]. Utrecht: Uitgeverij Jan van Arkel.
Powell, G. N., & Posner, B. Z. (1989). Commitment to career versus family/home life: Effects of
sex, sex-role identity, and family status. Psychological Reports, 64, 695–698.
Powell, G. N., Posner, B. Z., & Schmidt, W. H. (1984). Sex effects on managerial value systems.
Human Relations, 37, 909–921.
Pratto, F., Stallworth, L. M., Sidanius, J., & Siers, B. (1997). The gender gap in occupational role
attainment: A social dominance approach. Journal of Personality and Social Psychology,
72(1), 37–53.
Rabasca, L. (2000). In search of equality. Monitor on Psychology, November, 30–31.
Radford, J., & Holdstock, L. (1995). Gender differences in higher education aims between
computing and psychology students. Research in Science and Technological Education, 13,
163–176.
Rojahn, K. (1996). Gender in the context of leadership. Unpublished PhD thesis, University of
Amsterdam.
Rossi, A. (1965). Women in science: Why so few? Science, 148, 1196–1203.
 Underrepresentation of women in science 337

Sabbadini, L. L., & Camporese, R. (1998). Italian experience in analysis of ‘different times’ and
methodological aspects in measurements of unpaid work: The European pilot survey on
time use. International research and training institute for the advancement of women,
Working Paper No. 7. Presented at the Conference of European Statisticians, Geneva, April
1998.
Sanderson, A., & Dugoni, B. (1999). Summary report 1997: Doctorate recipients from United
States universities. Chicago, IL: National Opinion Research Center.
Schein, V. E. (1973). The relationship between sex role stereotypes and requisite management
characteristics. Journal of Applied Psychology, 57, 95–100.
Schein, V. E. (1975). Relationships between sex role stereotypes and requisite management
characteristics among female managers. Journal of Applied Psychology, 60, 340–344.
Schmitt, M. T., Ellemers, N., & Branscombe, N. (2003). Perceiving and responding to gender
discrimination at work. In S. A. Haslam, D. Van Knippenberg, M. J. Platow, & N. Ellemers
(Eds.), Social identity at work: Developing theory for organizational practice (pp. 277–
292). Philadelphia, PA: Psychology Press.
Schnake, M. (1991). Organizational citizenship: A review, proposed model, and research agenda.
Human Relations, 44, 735–759.
Smith, A. C., Organ, D., & Near, J. P. (1983). Organizational citizenship behavior: Its nature and
antecedents. Journal of Applied Psychology, 68, 653–663.
Staines, G., Tavris, C., & Jayaratne, T. E. (1974). The queen bee syndrome. Psychology Today, 7,
55–60.
Steers, R. M. (1977). Antecedents and outcomes of organizational commitment. Administrative
Science Quarterly, 22, 46–56.
Steers, R. M., & Braunstein, D. N. (1976). A behaviorally-based measure of manifest needs in
work settings. Journal of Vocational Behavior, 9, 251–266.
Steinpreis, R. E., Anders, K. A., & Ritzke, D. (1999). The impact of gender on the review of the
curricula vitae of job applicants and tenure candidates: A national empirical study. Sex Roles,
41, 509–528.
Tajfel, H. (1978). Differentiation between social groups: Studies in the social psychology of
intergroup relations. London: Academic Press.
Tajfel, H., & Turner, J. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S.
Worchel (Eds.), The social psychology of intergroup relations (pp. 33–47). Monterey, CA:
Brooks Solidus Cole.
Taynor, J., & Deaux, K. (1973). When women are more deserving than men: Equity, attribution,
and perceived sex differences. Journal of Personality and Social Psychology, 28, 360–367.
Timmerhuis, V. C. M., & Bringmann, E. S. (1999). Vrouwen in de wetenschap: Nederland in
internationaal perspectief [Women in science: the Netherlands in an international frame
of reference]. Den Haag: AWT/NWO.
Tougas, F., Brown, R., Beaton, A. M., & St Pierre, L. (1999). Neosexism among women: The role
of personally experienced social mobility attempts. Personality and Social Psychology Bull-
etin, 25, 1487–1497.
Twenge, J. M. (1997). Changes in masculine and feminine traits over time: A meta-analysis. Sex
Roles, 36, 305–325.
Van der Lippe, T. (1993). Arbeidsverdeling tussen mannen en vrouwen [Division of labour
between men and women]. Amsterdam: Thesis.
Van Vianen, A. E. M., Ottens, M., & Van Schie, E. C. M. (1997). Wetenschappelijke productie van
vrouwen en mannen in de psychologie. Nederlands Tijdschrift voor de Psychologie [Scien-
tific productivity of men and women in psychology], 52, 192–198.
Van Vianen, A. E., & Willemsen, T. M. (1992). The employment interview: The role of sex
stereotypes in the evaluation of male and female job applicants in the Netherlands. Journal
of Applied Social Psychology, 22, 471–491.
Wanous, J. P., Reichers, A. E., & Hudy, M. J. (1997). Overall job satisfaction: How good are single
measures? Journal of Applied Psychology, 82, 247–252.
338 Naomi Ellemers et al.

Wenneras, C., & Wold, A. (1997). Nepotism and sexism in peer-review. Nature, 387, 341–343.
Williams, J. E., & Best, D. L. (1982). Measuring sex stereotypes: A thirty-nation study. Beverly
Hills, CA: Sage.

Received 18 October 2001; revised version received 17 February 2003