For those working in academia, the tenure process is one of the last hurdles in what has been a long series of obstacles and challenges. On the face of it, the process seems straightforward: The candidates’ scholarly work is reviewed by their peers, and tenure is awarded based on the merits of the work. Of course, such statements greatly oversimplify the process and overlook the influences on the process that are outside the candidates’ control.
Some influences that contribute to differences in tenure outcomes include whether the faculty member took a postdoc and whether the faculty member earned a degree outside the US. They can also include demographic characteristics, such as gender, race, and ethnicity. Another factor is whether a faculty member asks for and is granted more time until their tenure review, commonly known as stopping the tenure clock.1
Barriers experienced by faculty members who are women and who are from other minoritized groups have been documented clearly in the literature. Faculty members who are Black encounter biased teaching evaluations, inequitably distributed resources, lower salaries, and unwelcoming environments—in addition, university leaders often undervalue their service work.2,3 Latino faculty members must negotiate workplaces with difficult climates that create isolation and alienation.4 Faculty members who are Asian or Asian American face stereotypes that limit their advancement in academe.5 Women faculty members must deal with less access to resources, isolation,6 and sexual harassment.7 All the barriers create cumulative disadvantage, a situation in which those who are minoritized receive fewer resources than their advantaged colleagues,8 which ultimately leads to a gap in career progress and recognition between white men and people from minoritized groups, including women.
Research, especially qualitative research, shows that the experiences of members of minoritized groups are multidimensional, complex, and unable to be explained neatly. No one person’s experiences are exactly like another’s. But the overwhelming conclusion is that the experiences of faculty members from minoritized groups differ significantly from those of faculty who are white and who are male—and often are much less positive.
Our analysis in this article attempts to document one result of the cumulative disadvantage caused by those multiple barriers. In this article, we use data from a 2021 survey of physics and astronomy faculty members conducted by the American Institute of Physics (the publisher of Physics Today) to describe how gender, race, and ethnicity affect the length of time between receiving a PhD and receiving tenure for faculty members in physics and astronomy departments.
Diversity in physics
It has been long recognized that members of minoritized racial and ethnic groups are poorly represented in physics relative to their representation in US society. The situation is similar in astronomy,9 and other reports have documented the lack of representation of women in physics and astronomy.10
The underrepresentation of people who are Black/African American and Hispanic/Latino in physics starts early on the academic pathway. (In discussing results of the 2021 survey, we use the terms that were presented to the respondents on the questionnaires.) Figure 1 shows that Hispanic/Latino students and Black/African American students are less likely to take physics in high school than white students and Asian/Asian American students. Those differences are largely a function of the socioeconomic status of the high school the students attend, because Black/African American and Hispanic/Latino students are less likely to attend a school that offers physics.11
The proportion of students in different racial and ethnic groups that have taken high school physics in the US. (Adapted from ref. 11.)
The proportion of students in different racial and ethnic groups that have taken high school physics in the US. (Adapted from ref. 11.)
Figure 2 further illustrates the continued underrepresentation of people who are Hispanic/Latino and Black/African American at higher levels in the physics educational system. For the class of 2021, Black/African American students earned just 4% of physics bachelor’s degrees, and Hispanic/Latino students earned about 10% of physics bachelor’s degrees. But the approximate representation of people from those two groups in the US population is 14% and 19%, respectively.12 People from the two groups are also underrepresented among physics PhD graduates. In 2021, less than 2% of PhD graduates were Black/African American, and 6% were Hispanic/Latino. At the faculty level, most people in physics and astronomy departments identified as white. In 2021, about 7% identified as Asian/Asian American, 2% as Black/African American, and 4% as Hispanic/Latino. Among tenured faculty members, the numbers were 8%, 2%, and 4%, respectively.
Underrepresentation in physics. (a) For the class of 2021, Black/African American students earned just 4% of physics bachelor’s degrees, and Hispanic/Latino students earned about 10% of physics bachelor’s degrees. But the approximate representation of people from those two groups in the US population is 14% and 19%, respectively.12 (b) People from the two groups are also underrepresented among physics PhD graduates. In 2021, less than 2% of PhD graduates were Black/African American, and 6% were Hispanic/Latino. (Courtesy of the American Institute of Physics Statistical Research Center.)
Underrepresentation in physics. (a) For the class of 2021, Black/African American students earned just 4% of physics bachelor’s degrees, and Hispanic/Latino students earned about 10% of physics bachelor’s degrees. But the approximate representation of people from those two groups in the US population is 14% and 19%, respectively.12 (b) People from the two groups are also underrepresented among physics PhD graduates. In 2021, less than 2% of PhD graduates were Black/African American, and 6% were Hispanic/Latino. (Courtesy of the American Institute of Physics Statistical Research Center.)
Although the representation of Asian/Asian American people in physics is low, it is not lower than their representation in the US population, which is 6.3%. For that reason, Asian/Asian American people are not considered underrepresented in physics. But a group’s representation compared with the general population says nothing about whether people in that group experience inequality in their educational or professional careers. To document one aspect of that inequality, we analyzed whether faculty members who are Asian/Asian American, Black/African American, or Hispanic/Latino take longer on average to receive tenure than white faculty members.
Faculty survey
In the spring semester of 2021, we sent a questionnaire to a representative sample of approximately 5500 faculty members in US degree-granting physics and astronomy departments. The approximately 1800 respondents included adjuncts and faculty members who were tenured, not yet tenured, and not on the tenure track.
Our goal was to focus on early-career academics. Because some faculty members may have entered academia after a career working in another sector—and it likely took those people longer to earn tenure—we removed from our analysis faculty members who took more than 10 years to become an assistant professor after earning their highest degree. We found no statistically significant difference in race and ethnicity or in gender between those who took more than 10 years to reach assistant professor and those who did not.
In our analyses, we include 448 respondents who had received tenure and used the data in an ordinary least-squares regression model.13 On the basis of the literature, we hypothesized that faculty members who are women14 or who are members of other minoritized groups would take longer to receive tenure.
We calculated time to tenure as the difference between when tenure was conferred and when the respondent was awarded their highest degree. The median time to tenure was 9.3 years. Half of the respondents received tenure between 7.3 and 11.7 years after receiving their highest degrees. Figure 3 depicts the time to tenure by race and ethnicity. The median time for faculty members who identify as Asian/Asian American or as Black/African American was more than 10 years.
Faculty members who are Asian/Asian American and Black/African American have longer median times to tenure than faculty members who are white. The data come from a 2021 faculty survey administered by the American Institute of Physics. The dark purple boxes indicate the middle 50% of the data, and the median is indicated by the yellow lines. The light purple boxes extend to the 10th and 90th percentiles.
Faculty members who are Asian/Asian American and Black/African American have longer median times to tenure than faculty members who are white. The data come from a 2021 faculty survey administered by the American Institute of Physics. The dark purple boxes indicate the middle 50% of the data, and the median is indicated by the yellow lines. The light purple boxes extend to the 10th and 90th percentiles.
Because many factors can affect the length of time it takes faculty members to receive tenure after earning their highest degree, we developed a multivariate model that includes explanatory factors for which we had measures in the survey. They include data about the institution and department where the faculty member is employed, including what type of school it is (public or private); whether it is a historically Black college or university; and what the highest physics degree it offers is. The model also considers personal factors, such as whether the highest degree was earned in the US or abroad; whether the faculty member took a postdoctoral appointment, and if so, how long it was; whether the faculty member’s tenure clock was stopped; and what demographic characteristics the respondent possesses, including gender identity and race/ethnicity.
Although the questionnaire allowed respondents to indicate more than one race/ethnicity, a respondent’s data may only be included once in a regression model. For that reason, faculty members who indicated that they belonged to more than one minoritized racial and ethnic group were counted as a member of the smaller group. In the analysis, the “another race/ethnicity” category included faculty members who identified as Indigenous, because so few identified that way in the survey.
Time to tenure
Some of the model results, which are summarized in figure 4, are not surprising. For example, faculty members who stopped or extended their tenure clock took about a year longer to receive tenure than those who did not. It also is not surprising that faculty members who took a postdoctoral appointment took longer to receive tenure than those who did not. For every year spent in a postdoc, however, the time to tenure increased by less than a year. That indicates a good return on the time invested in a postdoc.
Multiple factors—including institutional (orange), personal (blue), and demographic characteristics (purple)—could affect how long it takes a faculty member to receive tenure. In our multivariate analysis of faculty survey data, we compared each characteristic in the first column with the reference characteristic in the second column. The green check marks in the third column show which factors had an effect on time to tenure.
Multiple factors—including institutional (orange), personal (blue), and demographic characteristics (purple)—could affect how long it takes a faculty member to receive tenure. In our multivariate analysis of faculty survey data, we compared each characteristic in the first column with the reference characteristic in the second column. The green check marks in the third column show which factors had an effect on time to tenure.
Our hypothesis that women take longer to receive tenure than men was not supported. In addition, we found no statistically significant difference in time to tenure between faculty members who identify as white and faculty members who identify as Hispanic/Latino, although barriers in the tenure process have been found in qualitative research.15 The intersecting identities of some faculty members can lead to multiple barriers for them, which could increase time to tenure. When we included intersectionality of gender and race/ethnicity in our model, however, the result was not statistically significant. The lack of statistical significance is likely because of the small number of people in the intersectional groups that were studied, which led to a large standard error in the analysis.
We did, however, find evidence of differences in time to tenure for faculty members who are Black/African American and Asian/Asian American compared with white faculty members. When we control for all the variables in the model, Black/African American faculty members on average received tenure 1.5 years later, and Asian/Asian American faculty members 0.8 years later, than similar white colleagues. The differences are statistically significant.
Mounting evidence
Consistent with research on the barriers experienced by faculty members from minoritized racial and ethnic groups, our results provide quantitative evidence that faculty members who are Asian/Asian American and those who are Black/African American have longer time-to-tenure experiences than faculty members who are white. That may result from the barriers experienced by faculty members who are minoritized. The stereotyping, bias, and unwelcoming environments documented for faculty members who are Black/African American and who are Asian/Asian American may result in cumulative disadvantages, which lead to a longer time to tenure. For example, when faculty members work in unwelcoming environments and are implicitly categorized by stereotypical traits, they may be less likely to be offered opportunities to collaborate. Fewer collaborations may lead to fewer publications and fewer successful grant applications, cumulating in longer time to tenure.
Unfortunately, our quantitative analysis can go no further in specifying the exact causes of the differences in time to tenure. There could be several reasons a faculty member receives tenure later than colleagues. For example, some faculty members could make a request or be encouraged to submit their materials later without a formal “stop the clock” process. The latter “encouragement” may be well-intentioned but could be a result of unconscious biases about the abilities of faculty who are members of minoritized racial and ethnic groups.
Some institutions, for example, automatically give longer time periods to faculty members who are moving from a less prestigious institution or who take parental leave. Faculty members may not think of those delays as stopping the tenure clock, which would have been accounted for in the model. Another consideration is that faculty members from minoritized racial and ethnic groups may have a more difficult time finding a welcoming institution. They may be more likely, therefore, to change institutions before coming up for tenure, which delays their total time to tenure. None of those factors were measured in our survey.
Ideally, faculty members receive tenure and remain with a department for several decades or more. But as our analysis shows, the process may be far from ideal, since barriers lengthen some people’s tenure process. That makes it even more important to continue examining the experiences of faculty members who have not yet received tenure. (For some personal perspectives on being denied tenure, see the article by Toni Feder on page 44.)
A wealth of qualitative research has identified many of the barriers faced by faculty members from minoritized groups. Our results show that researchers need to continue to collect evidence in order to identify the mechanisms by which the barriers translate into a longer tenure process. Once the mechanisms are identified, they can be dismantled. Scientists routinely address hard problems. It is time to include among them the inequitable experiences of faculty members from minoritized groups so that access to tenure is not hampered by bias or barriers.
We are grateful to the physics and astronomy faculty members who completed the questionnaire. We gratefully acknowledge the contributions of John Tyler, who manages the faculty survey data set; Arlene Modeste Knowles and Jovonni Spinner, who reviewed the piece; and an anonymous reviewer, who provided helpful comments.
References
Rachel Ivie is a senior research fellow at the American Institute of Physics in College Park, Maryland. Susan White directs AIP’s Statistical Research Center.
