|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 35
PREPUBLICATION COPY
3
Gender Differences in Academic Hiring
This chapter examines this critical entry point into an academic career—and its
components—with a primary focus on differences in hiring outcomes for tenure-track
assistant professor and tenured associate or full professor positions, and how these
differences might be explained. The following research questions are addressed:
• Is gender associated with the probability of individuals applying for S&E
positions in Research Intensive institutions?
• Given that an individual applies for a position, does a woman have the same
probability of being interviewed as a man?
• Given that an individual is interviewed for a position, does a woman have the
same probability of being offered a position as a man?
As the chapter explores the impact of institutional and departmental characteristics, rather
than the individual characteristics of potential applicants and job candidates, another way
to frame the research questions is, what are the characteristics of Research Intensive (RI)
institutions associated with proportionately more applications from women, interviews of
women, and offers to women?
The chapter is divided into five sections. We outline the hiring process with a focus on
three key parts of the hiring process—applications, interviews, and offers. The final two
sections describe faculty perceptions of hiring and institutional policies based on data
from our faculty survey. You can find a review of the relevant literature and research and
what it suggests we should expect to find in our survey data in Appendix 3-1.
THE HIRING PROCESS
The hiring process consists of a series of decisions made sequentially by an
academic department and job applicants. A department is authorized to search to fill a
faculty position. The search may be for a senior faculty member who will be offered a
tenured position; for a tenure-track position, which has the potential to become a tenured
position, but does not provide tenure at the time of hire; or for both. This chapter
separately considers tenure-track positions and tenured positions for which the six
science and engineering departments in RI universities surveyed completed searches in
the period 2002 to 2004. This report does not report on positions off the tenure track
because no data were collected on these openings.
35
OCR for page 36
36 GENDER DIFFERENCES AT CRITICAL TRANSITIONS IN CAREERS
This section briefly outlines the steps in the hiring process as follows:
• the department’s actions in advertising the availability of a position;
• the individual’s decision on whether to apply for the position;
• the department’s choice of individuals to interview and to make the first offer to;
and
• the individual’s choice of whether to accept the offer.
Each of these steps is described below.
Advertising the Position
As part of the process that authorizes a department to fill a faculty position at a
tenured or tenure-track level, the department determines the subfield(s) the individual
will be expected to fill (both in a research and teaching capacity). Tenure-track positions
at the assistant professor level are advertised nationally in journals and at national
conferences. Letters may also be sent to department chairs or faculty in a particular
subfield notifying them of open positions. Efforts are generally made to make the hiring
process for tenure-track positions appear open and equitable. Advertisements note that
the institutions follow Equal Employment Opportunity (EEO) rules and many ads
specifically encourage applications by women and minorities. At this point in the
process, it is very likely male and female candidates are equally aware of most positions.
That is, there is unlikely to be a gender-based information gap.
In addition to national advertising, however, the hiring process for tenure-track
positions also involves recruiting that could result in gender differences in application
rates. For example, word-of-mouth recruiting practices by faculty may generate
differences by gender, intentionally or not, in information about the position available to
potential applicants. Search committees may try to overcome the limitations of
established networks by making special efforts to reach out to increase the number of
women applying for a given position.
The recruitment process for tenured positions may differ from the process for
tenure-track positions in subtle ways. Although the advertising for tenured positions
frequently mirrors the advertising for tenure-track positions, it is also common for a
department to formulate a list of the leading candidates, based on its view of who is doing
the most interesting and important research in that particular subfield, and to ask those on
the list directly if they are interested in applying.
The Decision to Apply
Once a potential applicant is aware of a position, this individual may or may not
choose to apply. In making this decision, a potential applicant may receive advice from
many people, including the person’s mentor, department chair, peers, faculty at various
institutions, family members, or spouse. A variety of factors may be taken into account
in determining whether to apply. These include expectations about the desirability of the
position (salary and benefits, prestige of the department, facilities, or workload); the
PREPUBLICATION COPY
OCR for page 37
GENDER DIFFERENCES IN ACADEMIC HIRING 37
location; and whether a spouse’s or other family member’s needs will be met. An
important factor may also include the encouragement (or lack thereof) that potential
applicants receive from faculty members they consult, particularly their dissertation or
postdoctoral supervisors.
Requests for Campus Visits, Interviews, and Selection
Once applications arrive, decision making reverts to the institution, typically
through an appointed search committee. At this point, the search committee ranks the
applicants and determines whom to invite to campus for interviews or for preliminary
interviews at professional society meetings. Search committees also consider a variety of
factors in determining who they feel are the best candidates, including expectations of
future productivity (e.g., research and grants received), ability to meet teaching needs,
and perceptions of fit. “Fit” is perhaps the most subjective criterion. It is usually thought
of as how well a particular candidate’s area of expertise or methodological approach
works with the department’s current needs or vision for its future strengths and mission.
However, it can also focus attention on a candidate’s demographic background or
personality. Different search committees weigh these factors differently. Top candidates
are invited to interview, which usually includes giving a talk about their research. This
gives the search committee extra information on a few candidates. At the end of this
process, often—but not always—an offer is made to a candidate.
The Decision to Accept or Reject the Offer
The final decision is made by the candidate whether to accept the offer or not.
Again, the candidate weighs many factors in making this decision. These include the
benefits of the position, other employment opportunities, and the candidate’s preferences
(possibly also including the preferences of a spouse or family members).
Data on Hiring
Data on the hiring process, as described above, are scant. Unfortunately,
nationally representative information is not available. First, there is no national evidence
on applicant behavior. It is not known if male and female S&E doctorates apply to
positions in a similar manner. Second, evidence of how search committees select one
candidate over another is lacking, perhaps because the selection process can be difficult
to quantify. Third, there is little evidence describing the number of individuals who go
through the hiring process. While departments collect information on the number of
applicants who apply for a position and are interviewed, and while gender is often noted
for these individuals, data are rarely made public for rather good reasons, including the
right to privacy of job applicants.1 Further, comparable data on hiring activities at
1
However, some institutions do release their analyses of hiring. An excellent example is the 2003 gender
equity report undertaken at the University of Pennsylvania which presents important data for consideration
and evaluation while maintaining anonymity. See:
PREPUBLICATION COPY
OCR for page 38
38 GENDER DIFFERENCES AT CRITICAL TRANSITIONS IN CAREERS
different universities are not generally available to allow an examination of how
university and departmental search policies and practices affect hiring outcomes.
National statistics such as the NSOPF or the SDR focus on individuals in their current
positions. The SDR asks doctorates about their postgraduate plans and whether they are
interested in a postdoctoral or academic position, but does not follow respondents any
further. As a result, this chapter will draw primarily from this study’s departmental
survey described in Chapter One and in Appendix 1-4.2
The survey asked chairs of the six targeted departments in each of the RI
universities to report whether they had conducted any searches during the 2002-2003 or
2003-2004 academic years. Of the 492 surveyed, 417 responding departments reported a
total of 1,218 searches, ranging between one and 15 searches per department.
Responding departments were asked to identify whether the search was for a tenured or
tenure-track position. In a few instances respondents wrote in “both” (17 out of 1,218),
and to a lesser degree “target of opportunity” (five out of 1,218). A few (40 out of 1,218)
left this question unanswered. Respondents were then asked to provide data on the
number of applicants and interviewees for each advertised position by gender. Finally,
they were asked to identify the gender of the individual who was first offered the position
and the gender of the person who was ultimately hired.3
In general, departments were much more knowledgeable about the later stages of
the hiring process and thus provided more complete data on offers and hires than on
interviews or applicants. The number of cases for which we had complete information on
applicants, interviewees, first offers and hires—all disaggregated by gender—varied
between 534 cases (with complete hire information) and 758 cases (with complete
applicant information). Thus, the number of cases considered in this chapter depends on
the stage of the hiring process. Only tenured and tenure-track cases are considered in the
analysis. For each stage in the hiring process (applications, interviews, offers),
descriptive statistics based on the data collected from the departmental survey are first
presented. Then, the appropriate statistical models are fit in order to understand the
departmental characteristics associated with the percent of females at each stage of the
hiring process.
http://www.upenn.edu/almanac/v50/n16/gender_equity.html. See also the report, University of California:
Some Campuses and Academic Departments Need to Take Additional Steps to Resolve Gender Disparities
among Professors, Report by the California State Auditor, 2001. Available at:
http://www.bsa.ca.gov/pdfs/reports/2000-131.pdf. See also the report by the Commission on the Status of
Women at Columbia University: Advancement of Women Through the Academic Ranks of The Columbia
University Graduate School of Arts and Sciences: Where are the Leaks in the Pipeline? Available at:
http://www.columbia.edu/cu/senate/annual_reports/01-02/Pipeline2a_as_dist.doc.pdf.
2
The committee acknowledges that the p-values for all of the data presented are unadjusted and the fact
that many of the data presented are interconnected.
3
A limitation of the survey was that it did not ask for the gender of every candidate offered a particular
position.
PREPUBLICATION COPY
OCR for page 39
GENDER DIFFERENCES IN ACADEMIC HIRING 39
APPLICATIONS FOR FACULTY POSITIONS
A necessary precondition for hiring a female faculty member is to have women
who are interested in applying for the position. The survey data clearly show that some
departments are more successful than others in attracting female applicants.4 Moreover,
our data show that there are still a number of positions for which no women apply.
Through out this report, we will present summary statistics, such as the following
ones, that state that for some characteristic of interest current values for men and for
women. These statistics do not reflect the survey weights5 and are not treated for the
different degrees of nonresponse that depended on the characteristic examined.
Therefore, these statistics are NOT appropriate estimates of any national characteristics
for men and women, but instead are quick impressions of the data collected, which are
often the beginning of a more meaningful analysis that is conditional on the disciplinary
area.6
Descriptive Data
While women are increasingly receiving Ph.D.s in S&E, they are still greatly
outnumbered by men in terms of applications for RI positions. For tenure-track jobs, the
median number of applications a department receives is 52 applications from men and
eight applications from women—or about seven applications from men for every
application from a woman. For tenured positions, the median number of applications a
department receives is 40 applications from men and eight from women, for a ratio of 5
to 1.7 Figure 3-1(a) presents a histogram of the percentage of female applicants for all
positions; Figure 3-1(b) presents the same information for tenured positions; and Figure
3-1(c) presents this information for tenure-track positions.
4
Note that this analysis implies nothing about the quality of applicants. Some people apply for jobs for
which they are not a very good fit. The committee did not assess whether male and female applicants
would behave any differently in this regard.
5
Recall that the committee’s survey was stratified in order to collect similar numbers of respondents in
each of the six disciplinary areas, and therefore respondents from different disciplines have different survey
weights.
6
These estimates would be useful as national estimates only in situations in which the disciplines are
relatively homogeneous with respect to a given characteristic and the non-response which occurred was
such that nonrespondents did not differ in their characteristics from respondents..
7
These figures are medians. The median was used because the data are skewed; there are a few positions
that had hundreds of applicants. The mean number of applications for tenure-track jobs was 85
applications from men and 17 from women. The mean number of applications for tenured jobs was 78
from men and 17 from women.
PREPUBLICATION COPY
OCR for page 40
40 GENDER DIFFERENCES AT CRITICAL TRANSITIONS IN CAREERS
FIGURE 3-1(a) Proportion of females among applicants to all tenured and tenure-track
positions.
NOTE: Final version will have intervals presented.
FIGURE 3-1(b) Proportion of females among applicants to all tenured positions.
NOTE: Final version will have intervals presented.
PREPUBLICATION COPY
OCR for page 41
GENDER DIFFERENCES IN ACADEMIC HIRING 41
FIGURE 3-1(c) Proportion of females among applicants to all tenure-track positions.
FIGURE 3-1(a-c) Histogram of the distribution of female applicants for tenure and
tenure-track positions in Research Intensive universities.
NOTE: The total number of tenure and tenure-track positions was 758.
NOTE: Final version will have intervals presented.
SOURCE: Survey of departments carried out by the Committee on Gender Differences in
Careers of Science, Engineering, and Mathematics Faculty.
Overall, departments received from one to 800 applications for their advertised
tenure-track positions (n = 630), and from one to 500 applications for tenured positions (n
= 128). Departments recorded only one applicant for 17 (3 percent) tenure-track
positions and nine (8 percent) tenured positions. The survey results showed that three
men and two women were hired through “target of opportunity” positions where the
position by intention was offered to only one candidate, though the rank at hire was not
known. Table 3-1 shows the number of cases with complete applicant information by
discipline and type of position (tenured or tenure-track). Note that the number of cases
across discipline and type of position combinations is roughly similar, so no discipline
contributes an inordinate proportion of the data to the analyses that follow.
PREPUBLICATION COPY
OCR for page 42
42 GENDER DIFFERENCES AT CRITICAL TRANSITIONS IN CAREERS
Table 3-1: Number of Tenured and Tenure-Track Positions with Complete Information
About the Gender of Applicants, by Discipline.
Discipline Tenured Tenure-Track
Biology 24 (15) 118 (43)
Chemistry 19 (16) 128 (47)
Civil Engineering 13 (9) 73 (33)
Electrical
Engineering 14 (9) 75 (27)
Mathematics 31 (16) 98 (37)
Physics 27 (14) 134 (47)
NOTE: Numbers in parentheses are the numbers of separate departments offering those
positions.
SOURCE: Survey of departments carried out by the Committee on Gender Differences
in Careers of Science, Engineering, and Mathematics Faculty.
.
Another finding is that for job openings for which only individuals of one gender
applied, that gender was more likely to be male. For tenure-track positions, there were
only nine openings for which no men applied (only women applied), and eight of these
were cases in which only one woman applied. On the other hand, there were no female
applicants (only men applied) for 32 tenure-track positions, or about 6 percent of
available positions, with only nine of these positions having a single applicant. Similar
findings were seen for tenured positions. For two positions, no men applied. These were
the two cases in which there was only one applicant. Conversely, no women applied to
16 tenured jobs, or 16.5 percent of the positions; only seven of these were single-
applicant positions. This finding may lend credence to the anecdotal argument
sometimes propounded by chairs or search committees that no women applied for
particular advertised positions (Brennan, 1996; see especially page 9).
Considering the data by discipline, in the instance of tenure-track positions, most
of the cases (29 of 32) in which only men applied occurred in physics or the engineering
fields. For tenured positions, 10 of the 16 cases occurred in chemistry (6) and physics
(4). This may reflect the fact that engineering and physics have a lower percentage of
female doctorates or that woman engineers and physicists are more likely to prefer
employment outside of major research universities.
Finally, how do the percentages of women applicants relate to the proportion of
women in the doctoral pool from which departments are drawing? One might expect the
proportion of female applicants to be similar to the percentage of doctorates awarded to
women in S&E across each of the disciplines. Table 3-2 suggests that this relationship is
more complex. In the table, the second column shows percentages of doctorates awarded
to women in the period 1999 to 2003 by doctoral-granting institutions, while the third
column shows percentages of Ph.D.s awarded to women by the subset of RI institutions.8
8
For a discussion of how to define the “pool of qualified candidates,” see NAS, NAE and IOM (2007).
PREPUBLICATION COPY
OCR for page 43
GENDER DIFFERENCES IN ACADEMIC HIRING 43
Data on the proportion of women among all applicants for tenure-track jobs by discipline
are presented in column four.
TABLE 3-2 Percentage of Women in the Doctoral Pool and Distribution of the
Percentage of Women Among Job Applicants for Tenure-Track Positions, by Discipline.
1999-2003 1999-2003 Mean Proportion of
All Doctoral- Research I Female Applicants
Granting Institutions for Tenure-Track
Positions (%)a
Discipline Institutions (%) Only (%)
Biology 45 45 26 (8, 25, 50)
Chemistry 32 32 18 (6, 15, 39)
Civil Engineering 18 18 16 (0, 10, 100)
Electrical
Engineering 12 12 11 (0, 10, 22)
Mathematics 27 25 20 (9, 20, 34)
Physics 15 14 13 (0, 10, 27)
NOTE: In parentheses, we show the 5th percentile, the median, and the 95th percentile
(computed over all tenure-track positions in each discipline) of the proportion of females
among applicants. Only those tenure-track positions with complete information about the
gender of candidates were included in these calculations (as in Table 3-1).
a
Mean proportion of female applicants computed as the average (over all tenure-track
positions) of the proportion of females in the applicant pool.
SOURCE: Ph.D. data are from NSF. WebCASP distribution of the percentage of female
applicants was computed using the same data used to construct Table 3-1.
In examining Table 3-2, it is important to note that while the second and third
columns reflect averages over individuals, the last column relates to the percentage of
women averaged over job openings. Thus, the values are not strictly comparable. An
individual can apply to more than one job and may be counted multiple times as an
applicant. If women are more likely to apply to multiple jobs than men, then the
proportion of women among applicants is overestimated. Conversely, if women only
apply to a few positions while men apply to many, then the average proportion of women
applicants (and the rest of the distribution of the percentage of female applicants) is
underestimated.
Table 3-2 shows that the proportions of applications from women are consistently
lower than the percentage of Ph.D.s awarded to women. There are, however, substantial
differences among the disciplines in how much lower. In electrical engineering,
mathematics, and physics, the percentage of women applying for faculty positions is only
modestly lower than the percentage of women receiving Ph.D.s. However, in the fields
with the largest representation of women with Ph.D.s—biology and chemistry—the
percentage of Ph.D.s awarded to women exceeds the percentage of applications from
women by a large amount. This finding should be further explored. Possible
explanations that might be tested in follow-on research include:
PREPUBLICATION COPY
OCR for page 44
44 GENDER DIFFERENCES AT CRITICAL TRANSITIONS IN CAREERS
• Biology and chemistry women doctorates prefer occupations outside of
Research Intensive institutions relative to men (for example, in higher
education, but in liberal arts colleges; in education as K-12 teachers; or in
industry or government);
• As the percentage of doctorates awarded to women increases, departments may
make fewer special efforts to encourage women to apply for faculty positions;
or
• Women Ph.D.s in biology and chemistry apply for fewer jobs than women in
other fields relative to men.
The first hypothesis may also, to a greater or lesser extent, hold for the smaller disparities
found in civil engineering, electrical engineering, mathematics, and physics.
Another study examining the proportion of women in Ph.D. pools relative to the
proportion of female faculty also found mixed results (NAS, NAE and IOM, 2007).
Examining data for faculty who were tenure-track or tenured in 2003 compared with
earlier averages of doctorates revealed that in engineering, chemistry, and the physical
sciences, there was a smaller proportion of women in the Ph.D. pool than in assistant
professor positions, while in the life sciences, computer sciences, and mathematics, the
proportion of women in the pool of doctorates was larger. Comparing the doctoral pool
to associate professors in engineering and life sciences, the proportion of women in the
pool exceeded the proportion of women associate professors. In computer science,
chemistry, the physical sciences, and mathematics, there was a greater proportion of
women associate professors. Considering full professors, the proportion of women full
professors in most fields was smaller than the proportion of women in the relevant
doctoral pool.
Statistical Analysis
Having summarized earlier in this chapter the literature on the factors potentially
associated with the percentage of applicants who are women, we now investigate whether
the data on hiring collected in our surveys support the hypotheses put forth by earlier
investigators. In our applicant models, the following institutional, departmental, and
position-level variables measured in our survey were used as explanatory variables:
discipline, type of position (tenured, tenure-track), whether the institution is private or
public, the prestige level of the department advertising the position, the proportion of
females in the search committee, the number of family-friendly policies advertised by the
institution, whether the search committee chair is a male or a female, the proportion of
female faculty in the department, and the size of the metropolitan area in which the
institution is located.
We first investigated whether any of these factors are associated with the
probability that no females apply to a position.9 To do so, we first created a binary
The vast majority of both tenure-track (94 percent) and tenured (83.5 percent) positions
9
had at least one female applicant.
PREPUBLICATION COPY
OCR for page 45
GENDER DIFFERENCES IN ACADEMIC HIRING 45
variable with the value 0 if there were no female applicants and the value 1 if at least one
woman applied to the position. We excluded for this analysis those positions identified as
target of opportunity and open rank positions. We fitted a logistic regression model to the
binary outcome variable and included as predictors in the model the institutional,
departmental, and position-level variables listed above, as well as two-way interactions
between discipline and the other predictors to investigate whether any of the potential
effects of predictors is discipline-dependent. To account for possible correlations within
positions advertised by the same institution, we implemented the method of generalized
estimating equations (GEE) to compute standard errors for all parameter estimates that
account for possible correlations across positions in the same institution.
We found the probability that at least one female would apply to a position is
associated with the set of discipline indicators (p = 0.03), type of position (p < 0.0001),
type of institution (p = 0.08), prestige of the institution (p = 0.04), and the number of
family-friendly policies in effect at the institution (p = 0.001). No other factor was
statistically associated with the probability of at least one female applicant. Results can
be more easily understood by looking at the adjusted means of the differences in the
probability of no female applicant across levels of some of the statistically significant
factors. These adjusted means are the means computed after “adjusting for” or
“accounting for” all other effects in the model. Technical details and the tables are given
in Appendix 3-1 (CD to confirm).
We then focused on all positions and modeled the number of female applicants as
a function of the same independent variables listed above. To do so, we fitted a Poisson
regression model to the number of female applicants and used total number of applicants
as an exposure variable. Possible correlation across positions advertised by the same
institution was accounted for when computing standard errors of parameter estimates via
the method of generalized estimating equations method. Again, we only included
positions that were advertised as tenured or tenure-track.
As expected, we found statistically significant differences across disciplines in the
proportion of females in the applicant pool. Biology, chemistry, and mathematics had
significantly higher proportions of female applicants than did all other disciplines across
all types of institutions and positions. The proportion of female applicants in civil
engineering, physics, and electrical engineering was significantly lower. The type of
position was not substantially associated with the proportion of females in the applicant
pool. The proportion of females among applicants to tenured positions was similar to the
proportion of females among applicants to tenure-track positions.
It has been speculated that the appearance of a woman-friendly environment
attracts female applicants. Our results confirm this view. The proportion of females in the
search committee and whether a woman chaired the committee were both significantly
and positively associated with the proportion of women in the applicant pool (p = 0.01
and p = 0.02, respectively). For every 1 percent increase in the proportion of females in
the search committee, we can anticipate an increase of about 0.7 percent in the proportion
of women in the applicant pool. In contrast, the number of family-friendly policies
advertised by the institution did not appear to be associated with the proportion of female
applicants. Other factors including type of institution (public or private), prestige of the
institution, and location of the institution had no association with the proportion of
women in the applicant pool.
PREPUBLICATION COPY
OCR for page 54
54 GENDER DIFFERENCES AT CRITICAL TRANSITIONS IN CAREERS
HIRES
Explaining hires made is more difficult, as the decision to hire involves the
department, which makes the offer, and the applicant, who accepts. The committee’s
departmental survey does not have information on characteristics of those ultimately
hired, beyond their gender. However, the committee’s faculty survey did ask faculty
some questions about reasons for accepting the position offered to them. Answers to
these questions are explored in the next section of this chapter.
Table 3-7 presents data on the gender of the individual receiving the first offer
and the gender of the faculty member ultimately hired for tenure-track positions.
TABLE 3-7 Percent of Candidates of Each Gender Who Received the First Offer and
Gender of Candidate Who Eventually Accepted Each Tenure-Track Position.
Person Hired Was a
Position was offered to Female Male
Female 70 (107) 30 (46)
Male 5 (19) 95 (362)
NOTES: Number of cases is given in parentheses.
Table 3-7 is based on the subset of the positions used to construct Table 3-6 for which the
gender of the person who accepted the position was known. We do not know from these
data whether the person who accepted the position is the same person who received the
first offer, even in those cases in which the gender is the same.
SOURCE: Survey of departments carried out by the Committee on Gender Differences in
Careers of Science, Engineering, and Mathematics Faculty
In 95 percent of the cases in which a man was the first choice for a position, a
man was ultimately hired in that position. Compare this to the case for women, where
only 70 percent of cases in which a woman was first offered a position was a woman
ultimately hired. In 30 percent of the cases in which women were offered first, a man
ultimately ended up in the position.10
10
Note, however, that we do not know if the person first offered and the person hired are the same person,
where the genders are the same. Nor do we know how many offers were made before someone was
eventually hired. Since men outnumber women in the offers made, one would expect that the proportion of
times women turn down an offer, resulting in a man being ultimately hired, should be higher than the
proportion of times that men turn down an offer, resulting in a woman ultimately being hired.
PREPUBLICATION COPY
OCR for page 55
GENDER DIFFERENCES IN ACADEMIC HIRING 55
Table 3-8 presents data on the gender of the individual receiving the first offer
and the gender of the faculty member ultimately hired for tenured positions.
TABLE 3-8 First Offer and Person Hired for Tenured Jobs, Percent by Gender.
Person Hired Was a
Position was offered to Female Male
Female 77 (20) 23 (6)
Male 0 (0) 100 (67)
NOTES: Number of cases is given in parentheses.
Table 3-8 is based on the subset of the positions used to construct Table 3-6 for which the
gender of the person who accepted the position was known. We do not know from these
data whether the person who accepted the position is the same person who received the
first offer, even in those cases in which the gender is the same. Number of cases is given
in parentheses.
SOURCE: Survey of departments carried out by the Committee on Gender Differences in
Careers of Science, Engineering, and Mathematics Faculty
In all cases in which a man was offered the position first, a man was ultimately
hired. Only 77 percent of the cases in which a woman was offered the position first was
a woman ultimately hired. In 23 percent of the cases in which a woman was offered the
position first, a man was ultimately hired, again suggesting that if the woman who is first
offered the position does not accept, there is a substantial chance the job will go to a
man.11
We do not have information in our survey data to permit investigating this
difference further. One plausible explanation is that many women who are offered
positions are the only woman interviewed for that position. If the only woman
interviewed is offered the position and turns it down (for whatever reason), that position
will inevitably be filled by a man. In fact, only one woman was interviewed for 205 (38
percent) of the tenure-track and 23 (24 percent) of the tenured openings for which more
than one person was interviewed. While there are many reasons why a person might
turn down a job offer, in this particular instance, it is possible women, who are
interviewed at disproportionally higher rates, also receive more offers than men and have
to turn some of them down.
FACULTY PERSPECTIVE ON HIRING
Turning to the faculty survey, the committee asked faculty who were either
tenure-track or tenured and had been hired after 1996 what were their “main
considerations in deciding to work for your current institution.” Respondents could
check up to 15 choices (the 15th and final choice was Other). For each selection,
respondents could check yes or no. These data were coded for analysis as follows: If a
11
. Footnote 76 applies here as well.
PREPUBLICATION COPY
OCR for page 56
56 GENDER DIFFERENCES AT CRITICAL TRANSITIONS IN CAREERS
respondent selected “yes” or “no” for some choices but left others unchecked, the
unchecked choices were recoded as no. A chi-square (χ2) test was conducted on each of
the 14 substantive selections against gender to investigate whether women and men
weighed factors differently when deciding to accept an offer for a position. The
responses are presented in Appendix 3-8 and are summarized in Figure 3-2 below. The
effect of gender was statistically significant only in the case of family-related reasons. As
might have been anticipated, women were more likely to weigh family-related factors
more heavily than men when deciding whether to accept an offer, but the difference is not
substantial.
Percent
0 10 20 30 40 50 60
Collegiality of faculty
Reputation of department or university
Opportunities for research collaboration
Desire to build or lead a new program or area of
research
Job location
Quality of research facilities
Start-up package
Access to research facilities
Family-related reasons
Promotion opportunities
Funding opportunities
Pay
Benefits
This was the only offer I received
Male Female
PREPUBLICATION COPY
OCR for page 57
GENDER DIFFERENCES IN ACADEMIC HIRING 57
FIGURE 3-2 Main considerations for selecting current position (percent saying “yes, this
was a factor”), by gender (see Appendix 3-8).
SOURCE: Faculty Survey carried out by the Committee on Gender Differences in the
Careers of Science, Engineering, and Mathematics Faculty.
INSTITUTIONAL POLICIES FOR INCREASING THE DIVERSITY
OF APPLICANT POOLS
Our findings suggest that once females apply to a position at a Research Intensive
institution, the chances that they will be invited to an interview and be offered a position
are disproportionately high for many of the disciplines we surveyed. Yet the proportion
of females in faculty positions continues to be low despite increasing numbers of women
receiving doctorates in the sciences and engineering. In this light, and given that the
percentage of women applying for positions is apparently lower than the percentage of
women receiving Ph.D.s in the six target disciplines, it appears that the only strategy to
increase female representation in the faculty ranks is to increase the proportion of women
in the applicant pool.
The NRC’s To Recruit and Advance: Women Students and Faculty in Science and
Engineering (2006) identified institutional characteristics, culture, and policies that may
have an impact on the proportion of females who choose to apply to academic positions
in science and engineering. Some of these include:
• Increased institutional efforts in signaling the importance of a gender-diverse
faculty. This might be accomplished by increasing the frequency of positive
declarative institutional statements, by establishing a committee on women, by
exercising close oversight over the hiring process, or by devoting additional
resources to hiring women.
• Modified and expanded faculty recruiting programs. Consider, for example,
creating special faculty lines earmarked for female or minority candidates,
ensuring search committees are diverse, encouraging intervention by deans when
applicant or interview pools lack diversity, and systematically assessing past
hiring efforts.
• Improved institutional policies and practices. These might include inserting some
flexibility into the tenure clock, providing child care facilities on campus,
establishing policies for faculty leave for family or personal reasons, significantly
stepping up efforts to accommodate dual career couples, and continuing to offer
training at all levels to combat harassment and discrimination and to raise the
awareness of all campus citizens.12
12
However, analysis presented in this chapter does not find an effect of the number of family- friendly
policies on the proportion of female applicants. The impact of such policies on applications may bear
further study.
PREPUBLICATION COPY
OCR for page 58
58 GENDER DIFFERENCES AT CRITICAL TRANSITIONS IN CAREERS
• Improved position of candidates through career advising, networking, and
enhancing qualifications.
While all of the strategies above might have an impact on the proportion of women in
applicant pools, it appears that only the last two might actually encourage more women to
choose academia for their professional activity. The issue is not whether women
applicants are treated fairly in the interviewing and hiring process; by several indications,
they are. Where progress can still be made is in attracting more women to academia by
encouraging more of them to apply for faculty positions at RI institutions. It seems that
refocusing resources to develop strategies to encourage female graduate students to
pursue a career in academia has the potential for enormous impact.
Written policies and handbooks for faculty searches frequently note specific steps
that can be taken to improve the diversity of applicant pools. These include:
• Defining searches broadly to encourage a more diverse applicant pool;
• Posting the job advertisement in a wide range of outlets;
• Contacting professional associations that represent women (e.g., the Caucus for
Women in Statistics, Society for Women Engineers, Association of Women in
Science, etc.); and
• Evaluating the applicant pool during the search to determine if sufficient numbers
of women are applying.
Departments reported a variety of actions in response to our survey question,
“What steps (if any) has your department or institution taken to increase the gender
diversity of your candidate pool?”. This was an open-ended question and the most
frequent responses are shown in Table 3-9. Four hundred seventeen departments
responded. Departments wrote in with answers ranging from zero to six steps and citing
anywhere from having none to 15 policies in place. Targeted or special advertising was
the most frequently cited action, followed by general advertising. These were followed
by recruiting at conferences, contacting women directly, and using personal contacts and
assistance from on-campus diversity offices.
In addition, for most departments the total number of steps taken is not large. As
shown in Table 3-10, 23 percent reported taking no specific action, and 43 percent
reported taking just one. Only slightly more than 10 percent reported taking three or
more steps.
PREPUBLICATION COPY
OCR for page 59
GENDER DIFFERENCES IN ACADEMIC HIRING 59
TABLE 3-9 Steps Taken to Increase the Gender Diversity of the
Candidate Pool
Step Number of
Departments
Reporting
Targeted or special advertising 80
Other 71
General advertising 58
Recruiting at conferences, contacting women 47
directly, using personal contacts
Help from diversity/EEO office or 47
coordinator
Contacting colleagues and other universities 42
Special language used in advertising 34
Special consideration to females (e.g., 34
making extra effort to interview females)
Informal networks 25
Grants or special funds for hiring women 19
Target of opportunity 19
Use of special databases or directories 18
Having a diverse search committee 17
Broadening searches 11
NOTE: Many of the 417 departments provided multiple answers to the open-ended
survey question, and 71 departments reported they have taken steps other than those
listed in the table.
SOURCE: Survey of departments carried out by the Committee on Gender Differences in
Careers of Science, Engineering, and Mathematics Faculty.
PREPUBLICATION COPY
OCR for page 60
60 GENDER DIFFERENCES AT CRITICAL TRANSITIONS IN CAREERS
TABLE 3-10 Number of Policy Steps Taken by Departments
Number of Number of Steps
Departments Reported Taken
96 (23) 0
178 (43) 1
98 (24) 2
34 (8) 3
10 (2) 4
0 (0) 5
1 (0) 6
NOTES: Numbers in parentheses are the percent of all responding departments; 417
departments responded. Of these, 98 (24 percent) took two policy steps to increase the
gender diversity of the candidate pool.
SOURCE: Survey of departments carried out by the Committee on Gender Differences in
Careers of Science, Engineering, and Mathematics Faculty.
SUMMARY OF FINDINGS
The analyses in this chapter reveal a number of important findings about the
application, recruitment, interview, and hiring process.
Applications
Finding 3-1: Women account for about 17 percent of applications for both tenure-
track and tenured positions in the departments surveyed.
There was wide variation by field and by department in the number and proportion of
female applicants for faculty positions. In general, the higher the percentage of women in
the Ph.D. pool, the higher the percentage of women applying for each position in that
field, though the fields with lower percentages of women in the Ph.D. pool had a higher
propensity for those women to apply. The proportion of applicant pools that included at
least one woman was substantially higher than would be expected by chance. However,
there were no female applicants (only men applied) for 32 (6 percent) of the available
tenure-track positions and 16 (16.5 percent) of the tenured positions.
Finding 3-2: There are statistically significant differences in the percentage of
women in the tenure-track and the tenured applicant pools across the six disciplines
surveyed. Biology, chemistry, and mathematics had significantly higher proportions of
female applicants than did all other disciplines. The proportion of female applicants in
civil engineering, physics, and electrical engineering was significantly lower. The
proportion of females among applicants to tenured positions was similar to the proportion
of females among applicants to tenure-track positions.
PREPUBLICATION COPY
OCR for page 61
GENDER DIFFERENCES IN ACADEMIC HIRING 61
Finding 3-3: In all six disciplines, the proportion of applications from women for
tenure-track positions was lower than the percentage of Ph.D.s awarded to women.
There were substantial differences among the disciplines. In civil engineering, electrical
engineering, mathematics and physics, the percentage of women applying for faculty
positions was only modestly lower than the percentage of women receiving Ph.D.’s.
However, in the fields with the largest representation of women with Ph.D.s—biology
and chemistry—the percentage of Ph.D.s awarded to women exceeded the percentage of
applications from women by a large amount (Table 3-2).
Finding 3-4: The median number of applications a department received for tenure-
track jobs was 52 applications from men and eight applications from women—or
about seven applications from men for every application from a woman. For
tenured positions, the median number of applications a department received was 40
applications from men and eight from women, for a ratio of 5 to 1. (Figure 3-1)
Finding 3-5: For job openings where only individuals of one gender applied, the
gender was more likely to be male. There were no female applicants (only men
applied) for 32 tenure-track positions or about 6 percent of available positions. Similar
findings were seen for tenured positions. No women applied to 16 tenured jobs—or 16.5
percent of the positions. Most of the cases (29 of 32) where only men applied occurred in
physics or the engineering fields.
Finding 3-6: Five factors were associated with the probability that at least one
female would apply to a position, including: (1) the type of position (p < 0.0001); (2)
the number of family-friendly policies in effect at the institution (p = 0.001); (3) a set
of discipline indicators (p = 0.03); (4) prestige of the institution (p = 0.04), and (5)
type of institution (approaches significance p = 0.08). No other factor was statistically
associated with the probability of at least one female applicant.
Recruitment
Finding 3-7: Most institutional and departmental strategies proposed for increasing
the proportion of women in the applicant pool were not strong predictors of the
percentage of women applying. Most steps (such as targeted advertising and
recruiting at conferences) were done in isolation, with almost two-thirds of the
departments in our sample reporting that they took either no steps or one step
designed to increase the gender diversity of the applicant pool. (Tables 3-9 and 3-10)
Finding 3-8: The proportion of females on the search committee and whether a
woman chaired the committee were both significantly and positively associated with
the proportion of women in the applicant pool (p = 0.01 and p = 0.02 respectively).
PREPUBLICATION COPY
OCR for page 62
62 GENDER DIFFERENCES AT CRITICAL TRANSITIONS IN CAREERS
Interviews
Finding 3-9: Across all the positions—tenure-track or tenured—an average of four
men and one woman were interviewed for any particular position. Our survey data
allowed us to examine the actual behavior of departments for the 545 tenure-track and 97
tenured openings for which we have gender data for applicants, interviewees, offers, and
ultimate hires.
Finding 3-10: The proportion of women who were interviewed for tenure-track or
tenured positions was higher than the percentage of women who apply. For each of
the six disciplines in this study the mean percentage of females interviewed for tenure-
track and tenured positions exceeded the mean percentage of female applicants. For
example, the female applicant pool for tenure-track positions in electrical engineering
was 11 percent, and the corresponding interview pool was 19 percent. (Table 3-3)
Finding 3-11: Although the proportion of females in interview pools across the six
disciplines exceeded the proportion of females in applicant pools, no women were
interviewed for 28 percent (155 positions) of the tenure-track and 42 percent (42
positions) of the tenured jobs. These figures are substantially higher than those for
the men. However, the proportion of male applicants was much higher than the
proportion of female applicants and part of this number was comprised of cases in which
there were no female applicants. In 23 percent of the tenure-track job openings (124
positions), at least one woman applied, yet no women were interviewed. In 25 percent of
the tenured jobs (23 positions), at least 1 woman applied, but no women were
interviewed. No men were interviewed for 3 percent (18 positions) of tenure-track, and
in one-half of those cases, there were no preceding male applicants; and for 4 percent (4
positions) of tenured jobs, and in one-half of those cases, there were no preceding male
applicants.
Finding 3-12: For tenure-track positions, the percentage of actual interview pools in
which only men were interviewed (no women) was smaller than would have been
expected based on applications and interviews for the positions surveyed for each of
the six disciplines. For tenured positions, this was the case for three of the disciplines
surveyed. Put another way, the percentage of actual interview pools in these disciplines
including women was larger than would have been expected. For tenure-track positions,
there were significant differences in electrical engineering (35 percent actual all-male
interview pools versus 56 percent probability of all-male pools) and mathematics (13
percent actual pools versus 33 percent probable pools).
For tenured positions, there were significant differences, again, in electrical
engineering (42 percent actual all-male interview pools versus 62 percent probability of
all-male pools); mathematics (39 percent actual versus 44 percent probable); and physics
(32 percent actual versus 35 percent probable). This was not the case for the remaining
disciplines, including biology (25 percent actual all-male interview pools versus 18
percent probability of all-male pools); civil engineering (46 percent actual pools versus
35 percent probable pools); and chemistry, which had the greatest difference (50 percent
actual pools versus 24 percent probable pools). (Table 3-4)
PREPUBLICATION COPY
OCR for page 63
GENDER DIFFERENCES IN ACADEMIC HIRING 63
Job Offers
Finding 3-13: The proportion of women who received the first job offer was higher
than the percentage who were invited to interview. Women received the first offer in
29 percent of the tenure-track and 31 tenured positions surveyed. Tenure-track women in
all of these disciplines received a greater proportion of first offers than their proportion in
the interview pool. For example, women were 21 percent of the interview pool for tenure-
track electrical engineering positions and received 32 percent of the first offers. This
finding is also true for tenured positions, with the notable exception of biology, where the
interview pool was 33 percent female and women received 22 percent of the first offers.
(Tables 3-5 and 3-6)
Finding 3-14: In 95 percent of the tenure-track and 100 percent of the tenured
positions where a man was the first choice for a position, a man was ultimately
hired. In contrast, in cases where a woman was the first choice, a woman was
ultimately hired in only 70 percent of the tenure-track and 77 percent of the tenured
positions. When faculty surveyed were asked what factors were considered for selecting
their current position, the effect of gender was statistically significantly for only one
factor, and that was “family-related reasons” (Figure 3-2; Tables 3-7 and 3-8).
As several of these findings suggest, many women fare well in the hiring process
at Research Intensive institutions. If women apply for positions at RI institutions, they
have better-than-expected chances of being interviewed and receiving offers compared to
male job candidates. The likelihood of receiving an interview and ultimately an offer was
particularly high, relative to application rates, in fields where women were less well
represented, such as engineering and physics. These findings suggest that many
departments at Research Intensive institutions, both public and private, are making an
effort to increase the numbers and proportions of female faculty in the sciences,
engineering, and mathematics. At the same time, women continue to be underrepresented
in the applicant pool relative to their representation among the pool of recent Ph.D.s.
The next chapter examines more fully the day-to-day lives of academics once they
are hired, considering whether there are disparities by gender in the areas of faculty
workload, institutional resources, and perceptions of departmental climate.
PREPUBLICATION COPY
OCR for page 64
