Abstract
Why are young women less likely than young men to persist in academic science and engineering? Drawing on 57 in-depth interviews with PhD students and postdoctoral scholars in the United States, we describe how, in academic science and engineering, motherhood is constructed in opposition to professional legitimacy, and as a subject of fear, repudiation, and public controversy. We call this the “specter of motherhood.” This specter disadvantages young women and amplifies anticipatory concerns about combining an academic career with motherhood. By specifying (1) the content of cultural discourses about motherhood in academic workplaces and (2) the processes by which these ideas circulate, produce disadvantage, and inform young, childless scientists and engineers’ career plans, our findings offer novel insight into mechanisms contributing to inequality in academic careers.
Women remain underrepresented in the highest-status and most lucrative professions in the United States (England 2010). One exemplar is academic science and engineering. As of 2017, women constituted only 24 percent of senior, full-time faculty (i.e., associate and full professors) in the natural sciences, and only 15 percent of senior full-time faculty in engineering (Burke 2019). Early career transitions, which often occur before individuals decide to have children, are an important source of this gap. Although women’s representation in STEM (science, technology, engineering, and mathematics) PhD programs has increased in recent years, women PhD recipients are less likely than men to apply for tenure-track faculty positions at research-intensive universities (National Academies Press 2010; Shauman 2017b) and less likely to attain them (Shauman 2017a).
The broad, interdisciplinary literature on this topic offers two central explanations. First, the work-intensive and male-oriented structure of academic science and engineering careers disadvantages women, who are more likely than men to shoulder family responsibilities (Ceci and Williams 2010; Mason, Wolfinger, and Goulden 2013; Williams and Ceci 2012; Xie and Shauman 2003). Second, women are disadvantaged by the highly gendered workplace culture of these fields (Roos and Gatta 2009; Sandler and Hall 1986; Settles et al. 2006; Williams, Alon, and Bornstein 2006). These structural and cultural forces, which prioritize a stereotypically white, male, “ideal” worker, are powerful and mutually reinforcing (Acker 1990; Cech and Blair-Loy 2014; Ecklund and Lincoln 2016).
Although parenthood is an important factor accounting for the relatively higher rates of attrition of women than men out of STEM careers (Cech and Blair-Loy 2019; Mason, Wolfinger, and Goulden 2013), previous studies have not specified (1) the content of cultural narratives about motherhood in academic science and engineering, (2) how such ideas circulate, or (3) how they might play a role in shaping career aspirations before people have children. In this project, we develop these theoretical linkages. Drawing on 57 in-depth interviews with young, childless graduate students and postdoctoral scholars across four research universities, we demonstrate how academic scientists and engineers construct motherhood in opposition to professional legitimacy and as a subject of fear, repudiation, and public controversy. We call this construction the specter of motherhood. We articulate how the specter of motherhood informs young, childless women’s early career experiences and deters some of them from persisting in academic science and engineering. We conclude by discussing implications for organizational change.
Gender and Motherhood in Academic Science and Engineering
One explanation for why women leave elite careers is that workplace practices in these careers are premised on an ideal and abstract, implicitly white, male worker who is continuously available and does not have domestic responsibilities (Acker 1990; Britton 2017; Stone 2007). Overwork expectations, often bolstered by the characterization of work as a moral calling, conflict with prescriptions that women ought to be intensively invested in motherhood (Blair-Loy 2003; Hays 1996). Ideal worker norms and expectations are especially salient in academic science and engineering, where long work hours are prescribed and the tenure clock typically coincides with childbearing years (Bailyn 2003; Cech and Blair-Loy 2014; Ecklund and Lincoln 2016; Williams, Alon, and Bornstein 2006). Intense competition between the demands of career and family motivates some mothers to exit these careers (Bailyn 2003; Cech and Blair-Loy 2019; Mason, Wolfinger, and Goulden 2013; Xie and Shauman 2003). Some young women who are not parents also voice anticipatory concerns about their future ability to balance work and family (De Welde and Laursen 2011), which may factor into their decision to leave (Goulden, Frasch, and Mason 2009; Williams and Ceci 2012).
Yet concerns about work–family balance do not fully account for women’s propensity either to enter or to leave STEM careers. Work–family plans do not effectively explain gender differences in college major choices (Cech 2016; Morgan, Gelbgiser, and Weeden 2013), nor do they affect young women’s chances of leaving a college engineering major (Cech et al. 2011). Moreover, family formation does not explain women’s relatively higher rates of attrition out of STEM versus non-STEM professional careers (Glass et al. 2013). Indeed, few women graduate students have children (Mason, Wolfinger, and Goulden 2013). Many women leave academic careers on completion of their PhD (National Academies Press 2010), which suggests that many leave before they have children.
Another leading explanation for why women leave elite careers centers on discrimination. When occupational tasks are male-typed, and workplaces male-dominated, cultural beliefs about gender can disadvantage women (Britton 2017; Ridgeway 2011; Turco 2010). In male-typed task settings such as STEM, men are often presumed to be more competent and interested in the subject matter than women (Charles 2011; Correll 2004; Thébaud and Charles 2018). These, often unconscious, beliefs prompt others to view women as less competent and capable and to hold them to higher standards of ability compared with men (Foschi 1996; Moss-Racusin et al. 2012). Women STEM graduate students have reported being ignored, not taken seriously, and feeling they must be more competent than men to be equally respected or rewarded (De Welde and Laursen 2011; Etzkowitz, Kemelgor, and Uzzi 2000; Fox 2001).
These biases are part of a workplace culture that is characterized by few role models and, at times, by overt discrimination and sexual harassment (Milkman, Akinola, and Chugh 2012; Roos and Gatta 2009; Sandler and Hall 1986; Settles et al. 2006; Williams, Alon, and Bornstein 2006). The stereotypically masculine image of the scientist as assertive, competitive, dominant, and strongly identified with work (Bailyn 2003; Cech and Blair-Loy 2014), is reinforced by a “macho” (Hewlitt et al. 2008), “geek” (Varma 2007), and “old boys club” culture (De Welde and Laursen 2011; Rosser 2004). It also reduces the likelihood that women identify with these fields (Murphy, Steele, and Gross 2007; Nosek, Banaji, and Greenwald 2002) or believe they can succeed in them (Cech et al. 2011; Correll 2004).
This masculine image stands in direct conflict with hegemonic cultural conceptions of motherhood. Mothers are expected to be intensive, domestic caregivers and, as such, have difficulty being seen as intensive workers (Blair-Loy 2003; Hays 1996). Mothers are also stereotyped as especially warm and nurturing (Cuddy, Fiske, and Glick 2004), traits that are culturally constructed as being in conflict with ideals of dominance and work commitment.
Because motherhood stereotypes are at odds with the stereotypically masculine image of a scientist, women are often placed in a double bind. Women who are unmarried, do not have children, or who prioritize work over family may be viewed as competent but unnatural and cold compared with male counterparts, whereas mothers who balance work with caregiving are at risk of having their competence and commitment questioned (Benard and Correll 2010; Ecklund and Lincoln 2016; Monosson 2008; Thébaud and Charles 2018; Williams, Alon, and Bornstein 2006).
Whereas cultural biases against mothers have been cited as a reason for dissatisfaction and attrition among science and engineering faculty, they have not figured prominently in accounts of earlier attrition. This is likely because young women often leave before they have children, and presumably before they are at risk of directly experiencing motherhood bias. However, there is some evidence that young, childless women are indirectly affected by such bias. Young, childless women graduate students report being advised to postpone, skip, or hide parenthood, or even a desire for parenthood (De Welde and Laursen 2011; Drago et al. 2006; Ecklund and Lincoln 2016; Etzkowitz, Kemelgor, and Uzzi 2000; Mason, Wolfinger, and Goulden 2013). Furthermore, mere exposure to the idea that scientific workplaces are intolerant of caregiving responsibilities is powerful enough to deter some young women from pursuing or persisting in these careers (Cech and Blair-Loy 2014; Weisgram and Diekman 2017).
These findings suggest that anticipatory cultural expectations about young women’s (in)ability to fulfill ideal worker expectations over the long term may be a source of gender differences in academic career ambitions. Our study builds on these works by detailing (1) the content of cultural discourses about motherhood and (2) the social interactional processes by which these discourses circulate and shape career ambitions before individuals have children. Specifically, we find evidence of a cultural discourse that frames motherhood in opposition to professional legitimacy, and as a subject of fear, repudiation, and public controversy. We theorize this “specter of motherhood”—which can be decoupled from actual motherhood—as a social process that shapes career ambitions and exacerbates gender inequality. Our argument draws inspiration from scholarship documenting how social interactional dynamics and repeated exposure to public narratives can create powerful specters that affect social status and inequality as well as personal decision making (Butler 2011; Miller, Sassler, and Kusi-Appouch 2011; Pascoe 2007).
Extending the Motherhood Penalty
A central focus of motherhood penalty studies has been to theorize how ideal worker norms activate discrimination against mothers. Studies show that mothers are often disadvantaged at work because they are seen as less committed and competent than their counterparts (Correll, Benard, and Paik 2007; Cuddy, Fiske, and Glick 2004). Yet some studies have suggested that women who are not mothers can also be negatively affected by the cultural beliefs that undergird the motherhood penalty (Cech and Blair-Loy 2014; Mason, Wolfinger, and Goulden 2013; Stone 2007; Williams, Alon, and Bornstein 2006).
We extend this work by specifying the cultural processes by which bias stemming from beliefs about motherhood can become relevant for women in the absence of pregnancy or parenthood. In the workplaces we studied, young women of childbearing age—whose childbearing future is generally unknown—receive the message that motherhood is something to be feared, and that they can gain respect by repudiating it and/or hiding any childbearing plans. Extant research tells us relatively little about the specific content of ideas about motherhood in workplaces, how such ideas circulate, or their discursive role. By specifying these cultural processes, we demonstrate how they directly disadvantage young women scientists and shape their aspirations. In doing so, we also build on research on hegemonic gender beliefs (Ridgeway 2011; Ridgeway and Correll 2004) by articulating how dominant narratives are powerful enough to affect our respondents’ experiences and behaviors, even though, in many cases, they did not personally endorse such narratives.
Methods
Research on gender inequality in academic STEM careers has focused largely on the experiences of undergraduates or faculty. We focus our analysis on graduate and postdoctoral years and can therefore identify key social processes contributing to a critical juncture in the “leaky pipeline” (Mason, Wolfinger, and Goulden 2013; National Academies Press 2010; Xie and Shauman 2003). Additionally, unlike other interview studies of early-career scientists and engineers, our sample includes women and men in a range of institutions and disciplines.
Data
We conducted in-depth interviews with 55 PhD students and 2 postdoctoral scholars in natural sciences and engineering departments at four large Research 1 Carnegie–classified universities. Three are private schools in the Northeast; the fourth is a public university on the West Coast. Two of these institutions had received an NSF-ADVANCE “institutional transformation” grant. The four universities are similar in that each offers paid parental leave, ranging from four to 12 weeks, to PhD students.
Interviews were semistructured and started with questions about the interviewee’s field and research focus. We then asked about relationships (with advisors, committee members, other faculty, graduate students, and staff) and experiences, including collaboration, resources, social integration, career goals, and family plans. At the end, we asked about gender-related experiences, though in many cases, gender-related issues had already emerged spontaneously. We concluded by asking questions about characteristics such as age, race/ethnicity, marital and parental status, and class background. We conducted one interview with each interviewee.
We interviewed respondents in geology, physics, horticulture, neurosciences, astronomy, chemistry, biological sciences, and engineering (materials, aerospace, mechanical, and electrical). Departments varied in gender ratio and the degree of mismatch between faculty and graduate student gender ratio. The percentage women faculty ranged from 6 to 38 and the percentage women graduate students was 14–53. Within departments, we used a random number generator to randomly select an even mix of women and men. E-mail requests for interviews did not mention gender. If someone declined, we randomly selected another person of the same gender from the same department. A few interviews were obtained through referral chains.
Table 1 shows respondents’ gender, race/ethnicity, aggregate academic field, age, and marital status. No respondents had children.
Characteristics of Sample Participants
NOTE: We could not hear the year of birth of one respondent.
The two authors conducted about three-quarters of the interviews, either together or separately, while we were graduate students or postdoctoral scholars. A graduate research assistant conducted the remaining interviews. Our common ground as graduate students and postdocs elicited candid and rich responses. Nevertheless, as social scientists, we were outsiders to the natural sciences and engineering.
All three interviewers are white women from the United States who speak English as our first language, which likely influenced the degree to which respondents felt they could open up to us and the relative ease with which we recruited white women to participate compared with members of other groups. Although our sample roughly mirrors the racial and ethnic composition of the departments we recruited from, the scarcity of women of color in these fields demonstrates a need for future studies to engage in purposive sampling on race, ethnicity, and nation, in addition to gender and to employ a more diverse team of researchers.
Analytic Strategy
All interviews were transcribed, and then coded and analyzed using ATLAS.ti and Excel. We relied on an emergent approach, similar to flexible coding or modified grounded theory, using open and theoretical coding (Deterding and Waters 2018; Glaser 1978). We used open coding to uncover major themes and theoretical coding to focus on sources, consequences, and content of themes. Themes emerged similarly across diverse institutional and disciplinary contexts.
We also assessed patterns in our data by coding and analyzing characteristics that prior research suggests would be related to gender and attrition (e.g., academic discipline, marital status, age, parental education). We counted, stratified by gender of respondent, whether the respondent was planning on leaving academia, staying, or was undecided, cross-classified by a subset of the more than 50 respondent characteristics.
Results
The Content and Circulation of the Specter of Motherhood
Our data show that dominant narratives in academic science and engineering frame motherhood, but not fatherhood, in opposition to legitimacy as a scientist or engineer. Fumie’s advisor, who is a father, said, “There’s more to life than babies . . . you should have a passion for science that should be driving you more than . . . family.” Erika’s advisor told her, “I hope you don’t have a kid during grad school” because “[I don’t] know how any woman would graduate when they have a kid.” When Erika pointed out that her lab recently graduated a student with a child, he told her that is different because that student was a man. Similarly, a male professor, who is a father, told Jean “I don’t understand why women complain . . . you just have to decide you get a family or a career in chemistry, one or the other and just accept it.”
This logic was bolstered by the assumption that motherhood negatively affects perceptions of commitment and competence (Correll, Benard, and Paik 2007). For instance, participants expressed a common understanding that mothers are viewed as less hireable and that their work is held to a higher standard. When we asked Nik whether his expectation to get married might affect his career, he responded, without any prompting about gender: “No. But if I was a woman, I would . . . because the general feeling is you get married and you have a baby. And if you’re going to go be an academic . . . it’s going to make people less likely to want to hire you . . . I think it sucks and it’s stupid. But it’s the way it is.” Although Nik personally thinks that mothers being viewed as less hireable is “stupid,” he believes that parenthood poses a credible threat to women’s, but not men’s, careers. Julia similarly expresses distaste for discrimination, even as she articulates her belief that mothers are held to higher standards. She described how a faculty member who is “brilliant” with an “amazing publication record”: . . . got this comment back pointing out that [her record] hasn’t been as robust as before she had a kid . . . people who . . . know the field really well, were like, that’s BS . . . if that had been a male faculty member [who had had a child] that’s a perfectly acceptable publication rate . . . we have to prove ourselves more . . . I think about these things a lot.
In the following sections, we detail how these ideas circulate and construct motherhood as a specter. First, because it is widely assumed to be a career liability, motherhood is interpreted as something to fear. Second, students and faculty devalue motherhood by repudiating and hiding it. Third, by repeatedly framing motherhood as a public and controversial issue, participants and their colleagues fuel the idea that motherhood is problematic. Though these themes were not evident in every interview, they were very prevalent in our data. At least two themes emerged in 47 (83 percent) of the interviews (67 percent of men and 90 percent of women), and in many cases, all three were present.
Fearing motherhood
Most of our women respondents expressed a sense of fear or negative emotions when discussing motherhood. Words such as “scary,” “frightening,” “worry,” “struggle,” and “stressed” routinely came up when we asked women to discuss their thoughts on combining a career in academic science with family, whereas these words were never used when we asked men the same question. Alice said, “It’s still scary—you know, combining a faculty job with a family is just so much . . . It’s scary.” Emily stated “Basically, I’m stressed and hoping for the best.” Jun’s response was dire: “Ahh [motherhood] would completely derail [my career].” Lauren unpacked her fear by explaining how she anticipates motherhood would amplify her current experiences of sexism: I don’t think it’s easy and I don’t think it’s friendly and . . . right now [if] I don’t like the way [a male colleague] talks to women, I can just walk away . . . but when somebody knows that you have a child, more people become that person . . . once you do have kids . . . they put all these assumptions on you and I think that must be really scary . . . trying to apply for faculty jobs knowing that people are judging you in these ways, that they do just naturally, not even thinking about consciously.
Notably, many respondents pushed back against these “assumptions,” even as they acknowledged how they were affected by them. Christina, who thinks people “probably” view women in science who intend to become mothers as less competent, said, “That’s definitely a fear I have, and I feel it shouldn’t be that way.” Hannah was similarly critical: “Academia, specifically, I think [engineers] make [balancing work and family] difficult on purpose . . . ridiculously difficult for no reason whatsoever.” Erin explained, “It’s really aggravating when I hear ‘women just don’t choose to do it’ and it’s like, yeah, no one forced those women to leave this program, but there are so many pressures that specifically apply to [mothers] . . . The system is screwed up . . . obviously, I think the system needs to be changed.”
In contrast, most of the men did not bring up work–family balance, for themselves, until they were asked about it, and then stated that they had either not thought about how parenthood might affect their career at all or had thought about it very little. Some men, especially those in relationships with other academics, acknowledged that it would be “hard” or “difficult.” But these acknowledgements were never conveyed with a sense of alarm or fear. When asked “What are your thoughts on balancing being a scientist with family life?” Stuart replied, “Very tough—I think that it would be—I guess I haven’t thought about that too much.” Similarly, Andy said “I mean definitely it’s from what I can gather it is challenging . . . I don’t know. My current advisor seems to balance it alright I guess. He has a baby-sitter. I don’t think he sleeps much but anyway so I don’t think too much about it honestly.”
Women’s fears were often informed by their perceptions of faculty. When asked, most participants indicated that they knew of at least one woman faculty member with a child in their department. However, only 16 participants (25 percent of the women, and 33 percent of the men) were able to describe a mother in academia who they felt “successfully” combined career with family. Miranda, whose department has only one woman faculty member with a child, explained, “It seems like all of the professors here . . . don’t have much of a family life. And even if they do, they have a stay at home [spouse]. . . . This is one of the biggest things I struggle with because I have no role models here.” Erin described a similar concern: “One thing . . . the girls and I talk about at these wine nights is staying in science and having an academic position and having kids but actually having a good relationship with those kids. Because we talk about, do you know any women, a tenured female professor who has kids who has a healthy relationship with them? We’re like no.” Such sentiments were common: Most mothers who had successful careers were perceived by students to be failing at motherhood since it “seemed” like they were unable to live up to intensive mothering expectations. The minority of participants who did name a woman faculty member who they felt successfully combined work and family used such terms as “amazing,” “badass,” “very pragmatic,” “extremely ambitious,” “high energy,” and “cutthroat” to describe them. Jenny described a “hard core” woman who had significant caregiving support from her partner. Similarly, Betsy explained how: I know of people who are professors or who go on to very high-powered, demanding jobs who have kids . . . Those [mothers] just have a lot of energy that I possibly, I don’t—I think I just don’t have that kind of energy—to work as much as some professors do and still like have a good life outside of work. So it seems like it can be done but it’s very difficult.”
These exceptions underscore the strength of the notion that, in this context, being a mother is at odds with career success: These women were perceived to be extraordinary individuals who possessed special traits—such as more savviness or extra energy—or who had special circumstances—such as a supportive partner—that enabled them to overcome the challenges of being a mother in academia. As such, these examples contributed to, rather than challenged, the idea that combining motherhood with academia is unattainable for most women.
Repudiating and hiding motherhood
Many faculty members and graduate students repudiated—that is, they refused to accept or associate themselves with—motherhood, and especially notions of intensive motherhood, in their day-to-day interactions. This occurred in part because hemogenic ideals of intensive mothering are in conflict with academic ideals of overwork. For example, many of our respondents reported hearing faculty members speak in demeaning or discouraging terms about motherhood or reject motherhood for themselves or others. Those who were unable to fully repudiate motherhood minimized their association with it by hiding it or by distancing themselves from their own parenthood and caregiving responsibilities. All of this contributed to the belief that motherhood holds power to undermine one’s professional legitimacy.
Erin recounted how, at a panel on gender issues in STEM, one professor’s “gist was that having children is sort of narcissistic. And she’s above that . . . like, simpletons want to have kids.” This characterization of motherhood demeans it, constructs it as a low-status identity, and also repudiates it. Anne explained how her advisor, who is also a woman, discourages motherhood: “She’s not that fond of the idea of other people having children while they’re in grad school. So she kind of casually discourages that any time she gets a female student so I feel like that’s a topic that I would try to avoid discussing with her until it was really an issue. . . . [She thinks] if you had a baby then that would just be like the end of productivity.” These conceptualizations of motherhood as both demeaning (e.g., for “simpletons” or “narcissists”) and in opposition to productivity contributed to our participants’ sense of motherhood as something to reject or downplay.
Faculty’s performances of their own parenthood also sought to minimize their association with motherhood and caregiving. Faculty talked very little about their children, and when they did, they deemphasized caregiving commitments. Grace explained that she only found out that her advisor had a child when she brought the child to a department event, which Grace interpreted as “awkward” and “odd” because she had been working with her for a long time. Layla described the cultural taboo against discussing family in detail: People just never talk about their families. . . . I think there’s this potential for not believing that people are as focused on their careers [if they discuss family]. I mean the woman I worked with at [university] had two kids, but she never talked about her family. . . . I never saw the baby and I have no idea what its name was . . . like pretty extreme. . . . I don’t know how many of the other female faculty in my department have kids . . . we just don’t talk about it. I feel like it’s kind of taboo.
Like Grace, Layla interprets faculty members’ performances of motherhood as out of step with life outside of academia (“extreme”). She also articulates how revealing any sort of family commitment can signal a lack of career commitment in this context.
Male faculty also distanced themselves from, and disparaged, caregiving. Chris recounted how his advisor is “very raw . . . he refers to his afternoon of taking care of his kids as ‘I’m screwed, I have to go play daddy.’ . . . I think he does it . . . as something to be cool or to give excuses why he’s not there.” It appears that it is challenging for Chris’s advisor to balance work and family, but he downplays his identity as a father to make himself seem like more of a committed scientist. Chris’s observation underscores how avoiding the stigma of caregiving applies to both men and women. Yet because hegemonic beliefs assume that women will take primary responsibility for caregiving, this stigma ends up, in effect, reinforcing the stigma of motherhood.
Given this cultural context, it is not surprising that most students felt compelled to dissociate themselves from motherhood as much as possible by hiding any plans to have children, or by explicitly making it known that they did not want children. For instance, despite having access to supportive policies, Erika felt that she had to hide her pregnancy from her advisor because he had previously told her that “no woman could graduate if they had a kid.” Ann described a scenario about her friend Nicole, whose advisor denied her paid leave owed to her under university policy. Needing a letter of reference, Nicole did not pursue the issue. This story affected Ann, who then decided to be “pretty forward” that she was not planning on having children—an intentional strategy of repudiation she used to let her advisor know he “didn’t have to worry about it.”
Taken together, our data suggest that academic scientists and engineers repudiate and hide motherhood in an effort to bolster their legitimacy. That is, the frequent and taken-for-granted rejection of, and distancing from, motherhood by both faculty and graduate students underscores the extent to which motherhood is believed to detract from one’s value as a scientist or engineer. The more our women participants felt they could not discuss family plans and obligations, but rather had to reject and hide them, the more these plans seemed to pose an insurmountable obstacle to career success and the more motherhood operated as a “specter.”
Framing motherhood as public and controversial
Faculty and students discursively construct motherhood as public and controversial with frequent discussions. Such discussions of motherhood effectively remind young women that they could be discredited as scientists or engineers if they become a mother or disclose plans to do so.
Respondents frequently reported hearing conversations about women’s, but never men’s, difficulties balancing work and family on panels and professional listservs, from faculty and graduate students, in discussion groups, in popular and academic articles, and from family and acquaintances. Charlotte feels like she has to think about future work–family balance because she is “surrounded by it all the time” and “it always comes up when there’s a discussion group.” Violet related, “The [school newspaper] sent out some article a couple of weeks ago . . . the ‘Baby Burden’ or something for women in academia and how women who have children are punished . . . in terms of their career.” Jon described how “we talk a lot about women and science . . . people say that women are told that they have to focus and plan and then they try and plan out the whole work/life balance and they realize that academic jobs are ridiculous.” Arianna described a discussion group: “once or twice a semester young faculty come who are women who are doing all these very good things. . .they try to reassure you, but even if you have the reassurances you can see that they’re all saying, oh, my god, it’s so crazy.” Although these discussions are often intended to support women or to understand the problem, their frequency and tone (e.g., motherhood is a “burden” or “crazy”) often, paradoxically, ends up undermining young women’s confidence in their ability to combine a career with family.
The idea that motherhood in academia is controversial and fair game for public discussion further works to devalue motherhood. Lyn related: “[A faculty member] already had a toddler and then while she was getting tenure she was pregnant again, which is a little bit controversial . . . some people were saying that that was unwise . . . Just ’cause she’s giving her tenure talk and her belly is out to here.” The characterization of pregnancy during the tenure process as “controversial” conveys the notion that motherhood should not be combined with career progress, or that, at the very least, it should be hidden from view. Along these lines, Michael perceived that women take tenure-track positions and “then all of a sudden [have] a giant belly and they go on maternity leave. . . . It’s taking advantage, maybe, of the system.” Like Lyn, Michael found it normative to discuss and evaluate women’s choices and “bellies.” He noted that if a male colleague were starting a family, “I probably wouldn’t have the same bias . . . I’m probably a little sexist.”
These were not isolated cases. It was commonly understood that women’s family plans are worthy of public comment. Alexandra described how a graduate student in her lab asked “weird personal questions” about a fellow student’s pregnancy. Layla described how faculty weighed in on student choices: . . . [another graduate student] who had a baby and she’s pregnant again now—she says like my advisor won’t congratulate her for being pregnant or whatever like maybe a normal person might. . . . I know when the girl had just had her baby, my advisor was like okay, it’s been so many number of weeks, you need to teach your kid how to drink from a bottle. . . . So I think when it comes to being encouraging and helping people get back to work she can do that, but she probably strongly prefers you to not have the baby.
Here the advisor’s interactions and comments, and Layla’s interpretation of them, underscore the controversial and devalued nature of motherhood: The advisor does not react to motherhood in a “normal” way, and when she does engage, she encourages the student to eschew caregiving responsibilities as quickly as possible. Notably, there are no instances in our data in which men’s fertility decisions were commented on or policed.
Above, we described the discursive element of repeatedly framing motherhood to make it seem public and controversial. Women students are surrounded by others’ concerns that they will not be able to combine motherhood with a career. The dialogue about mismatch between motherhood and career is pervasive and amplifies the notion that motherhood is, and should be, worrisome for women scientists. Even when respondents could see that it is possible to do both—say, to get tenure and have children—or, even if a woman herself does not want children, workplace beliefs that construct motherhood as a career threat, and therefore something to fear, were salient in the minds of most participants. Next, we describe how this specter of motherhood works to disadvantage potential mothers.
The Specter of Motherhood and Disadvantage
We find that the specter of motherhood creates disadvantages for young women without children. First, potential motherhood is routinely used to question their legitimacy in the profession. Graduate students joked to Shelby “about how women applicants get preferably interviewed, but then they end up dropping out anyway to have babies.” Christine was asked “Why are you getting a degree when you’re just going to leave the field and have children?” Megan described a professor as “very demeaning towards women,” who told her, “you finally decided to show up, you finally have time” after she missed class for a conference. Megan notes he would never have said that to a male graduate student. That this professor holds a young woman without children to higher expectations for class attendance than her male peers shows that, in this cultural context, the professional commitment of women without children is in question, similar to the way that mothers’ commitment has been shown to be (Correll, Benard, and Paik 2007).
Second, the more that women repudiate and hide their identity as potential mothers, the more they earn respect. Most women who believed they might have children in the future hid their plans from their advisors out of a concern that, if they did not hide their plans, then they would be perceived as less competent or committed. When asked what topics she might discuss with her advisor, Betsy explained: “If it were something [like] ‘I’m having a child’ . . . I would feel uncomfortable about how he’d receive that because of the ‘women always fail’ thing.” Betsy anticipates that she would be especially disadvantaged by the stereotype that women lack ability in science (i.e., “women always fail”) if she made her potential motherhood salient. Furthermore, women who did not want children, and who communicated that, believed they were given more opportunities than childless women whose plans were ambiguous. Mina related: My advisor affords me some really special opportunities . . . partially because I always put forth . . . that I was completely driven from a career perspective . . . that I was not restricted in any way by [plans for] having kids . . . I’m just not sure he affords the same caliber of opportunities to other women.
That Mina gains respect by being explicit about not wanting children suggests that young, childless women are at risk of having their abilities questioned unless they reject motherhood. Mina avoids disadvantage but in doing so, inadvertently reinforces the belief that motherhood is problematic. This dynamic echoes the work of Rhoton (2011), who showed that STEM faculty perpetuate gender inequality by distancing themselves from other women and by disparaging feminine practices and traits.
Third, the specter of motherhood can be decoupled from actual motherhood, which sometimes produces counterintuitive disadvantages. Whereas many interviewees related stories of mothers being seen as less committed and competent, Layla related how: they . . . [hired] this woman who . . . was breastfeeding between interviews. I think people were impressed by that. . . . If you are able to show that it’s not slowing you down . . . [motherhood] could be a positive or a neutral thing, but . . . if you’re talking about it in the abstract it would be . . . “you’re distracted from your research.”
When faced with a mother who was given the opportunity to demonstrate her ability, colleagues were impressed. Whereas some mothers may be able to compensate for motherhood bias through workplace behaviors, the potential for motherhood “in the abstract” activates doubts that are difficult to fend off with behavioral demonstrations. Ryan described this dynamic: People definitely think of [mothers] differently . . . it’s funny . . . it can be both a positive and a negative. If it’s a woman who’s already established then people definitely respect that a lot and I think even adds credibility to them because everybody’s like yeah, they have to be legit because they did all this and they have kids. But then I think people who are less established . . . people are questioning whether or not they want to be a scientist or whatever. Or does this mean that she’s ready to just like be a housewife or that sort of thing, and not pursue her career in the same fervor as she might have previously . . . either late PhD, early career. . . . I definitely could see people thinking that . . . women who become mothers at that point as . . . less dedicated scientists than their peers who chose to forgo having children until they’re . . . further along in their career.
These examples illuminate the decoupling of cultural motherhood and actual motherhood: Young women have to navigate an environment in which the timing and management of their fertility decisions are closely intertwined with their ability to obtain professional legitimacy.
In sum, the specter of motherhood informed young women’s concerns about the future and it shaped their present experiences. Young women’s status as potential mothers led colleagues to question their competence and commitment, especially if they did not, or could not, repudiate motherhood.
Future Career Plans
Overall, women respondents were more uncertain than men about their future in academia. At the time of the interview, 30 women (76.9 percent of women) and 9 men (50 percent of men) were either undecided or intending to leave academia (Table 2). We also asked respondents to retrospectively report whether they had been open to an academic career when they started graduate school. Notably, women and men were fairly equally likely to report that that they had initially been open to an academic career when they started graduate school (Table 3, column 2). However, women were more than twice as likely as men to report they had subsequently changed those plans, such that they no longer planned on an academic career (Table 3, column 4). We also counted, stratified by gender of respondent, whether the respondent was, at the time of the interview, planning on leaving academia, planning to stay, or undecided, cross-classified with race, academic discipline, marital status, dual academic-career status, age, and stage in graduate program (analyses not shown). We do not find evidence that gender differences in career plans are patterned on any of these characteristics.
Career Plans at Time of Interview.
Reports of Changes to Career Plans at Time of Interview
NOTE: “Open to academic career” includes individuals who had plans to stay in academic science or engineering or who were undecided about whether they would pursue academia or not.
We interpret these findings with caution, because our sample is small and nonrandom (i.e., it is not representative of the entire population of graduate students or postdocs). Moreover, it is possible that there is some recall bias regarding plans at the beginning of graduate school. However, we expect that such bias in this case would be fairly minimal, given that assessing retrospective states using recall is standard in social science research (Patler and Pirtle 2018), and memories of major life events and transition points tend to be reliable (Belli 1998; Brewin, Andrews, and Gotlib 1993).
Our qualitative interviews suggest that there is a relationship between gender-differentiated patterns of intended persistence and the specter of motherhood. Sixteen women and three men reported, retrospectively, that they had been open to an academic career at the start of their graduate program, but expressed definite plans to leave academia at the time of the interview (Table 3). An additional 11 women and four men remained undecided about the academic track. Of those who had either shifted plans away from academia or remained undecided, all 37 of the women (100 percent) and 5 of 7 men (71 percent) made comments related to the specter of motherhood during their interviews.
To more directly assess the connection between the specter of motherhood and attrition from academia, we analyzed the content of our conversations with each of the respondents who had shifted their plans since starting graduate school. In these interviews, we asked participants to discuss why they were no longer open to an academic career. Our analysis reveals that 13 of the 16 women who had changed their plans (81 percent) mentioned issues related to the specter of motherhood in their rationale. By contrast, specter of motherhood themes did not emerge in any of the three men’s reasoning for changing their plans.
Christin discussed her decision to leave in terms of her dissatisfaction with the culture surrounding motherhood. After referring to her advisor as an “alpha-scientist,” she explained that “No one wants to be beta. If you’re going to be a scientist, you’re supposed to be one of the . . . best. [Being a mother] is a huge struggle if you are an alpha scientist.” Christin’s “alpha” and “beta” language articulates how motherhood makes it harder to enact the dominance that is culturally expected of successful scientists. Saanvi similarly expressed distaste for the culture: “The culture of the workplace is that it can be very hard to . . . have a family and be in science, because you are expected to put in so many hours and it seems like that’s less acceptable for women who have families.” She continues that, in lab meetings, “I felt like my suggestions, my ideas—contributions weren’t taken as seriously . . . contributions weren’t perceived in the same way [as male peers].” Saanvi’s present experiences of gender bias contextualized the specter of motherhood for her: The prospect of taking on motherhood as an academic scientist was especially unappealing given how much she already struggles to be taken seriously.
Elizabeth’s concerns about future motherhood were also informed by her present experiences of bias. She described a sexist lab environment in which she and another young woman in her lab were made to feel “really, really dumb.” This made the prospect of motherhood especially daunting. “I work with [a toxic chemical] . . . if god forbid I got pregnant, I would have to ask for nine months leave. I’m not planning on having kids for quite a while, but it’s a really big consideration.” For Elizabeth, the idea of having children seemed especially scary (“god forbid I got pregnant”) because it would significantly handicap her ability to prove her competence, a challenge she already faces.
By contrast, concerns about fatherhood did not factor into men’s decisions to leave. Alex wanted a career with more “hands on research.” Jason did not believe he was “creative enough to come up with [research] ideas.” Like some of the women we interviewed, Terrence was deterred by the culture. “[My advisor] likes to, I think, publicly humiliate me. I think he almost gets a pleasure out of that. . . . I just don’t like this. I actually don’t like the people involved in this.” But despite his dislike of academic culture, Terrence viewed academia as advantageous for family: “[Academia] is more or less a stable job. There’s security, there’s health benefits . . . it would probably go well with having a family.” Terrence’s statement reflects how academia is well aligned with the breadwinner role and reveals his implicitly gendered assumptions about family responsibilities. By contrast, Alex acknowledged that academia is not “particularly favorable” for having a family, and Jason noted that it would be “tricky.” But neither raised it as a significant problem, and they did not mention it until prompted by the interviewer. In sum, among respondents who reported changing their intentions from a career in academia to a career outside of academia, the specter of motherhood emerged as a strong theme in women’s accounts of their decision making, whereas it was not present in men’s accounts.
Do Findings Vary by Academic Institution or Discipline?
Our participants span four universities and a range of science and engineering disciplines. It is possible that the presence and intensity of the specter of motherhood could vary according to institutional or disciplinary features. However, we find consistent evidence of the specter of motherhood across respondents at all universities and aggregate disciplinary groups: The majority of interviews (83 percent) included comments that were coded as pertaining to the specter of motherhood, and these codes were present at a similarly high rate across institutions and disciplines. Whereas some of the most egregious examples came, not surprisingly, from the most male-dominated fields, the overall sex composition of discipline was only one of many factors that came into play when determining the severity or frequency of the specter of motherhood. Of equal, and in some cases greater, importance were factors such as the gender composition and social environment of the lab or an advisor’s temperament, work ethic, or views about women in science. For example, several of the most distressed respondents were in biological sciences, whereas two astronomers, who had a supportive female advisor and a female-dominated lab, had positive experiences.
Discussion
Previous research has demonstrated the importance of motherhood as a mechanism undergirding gender inequality. We build on this work by showing how cultural beliefs about motherhood circulate in academic workplaces and shape the experiences and aspirations of potential mothers. Our study makes important theoretical contributions and offers insights into organizational strategies for change.
First, we extend previous work on gender inequality in academia by developing an account of how narratives and practices around motherhood disproportionately disadvantage young women and shape aspirations before they have children. Our findings offer one explanation for mixed evidence on the relationship between women’s family plans and their attrition from academic careers. Young women’s concerns are not based solely on long work hours or a lack of family-friendly policies. Rather, during the course of their training, they are exposed to a pervasive cultural belief that motherhood is something to fear because it holds the power to undermine their legitimacy as a scientist. This fear is rooted in their present experiences. Childless women already experience bias from faculty and graduate students related to both gender and potential motherhood and are bombarded with the message that such difficulties will worsen with actual motherhood. They are faced with the knowledge that, in order to gain professional status and respect, they will likely need to engage in practices that repudiate, denigrate, or hide motherhood—something that not all women are willing to do.
Second, our findings provide new insight into motherhood penalty research. Whereas initial research focused primarily on how ideal worker norms disadvantage pregnant women and mothers (Blair-Loy 2003; Correll, Benard, and Paik 2007), other work suggests that ideal worker norms additionally problematize anticipatory motherhood (Cech and Blair-Loy 2014; Ecklund and Lincoln 2016; Mason, Wolfinger, and Goulden 2013; Stone 2007; Williams, Alon, and Bornstein 2006). Our research extends this literature by (1) elucidating the mechanisms by which cultural beliefs about motherhood can circulate in professional workplaces and (2) specifying how these cultural practices can disproportionately shape the workplace experiences and future aspirations of young women without children.
Third, our findings offer insights for institutions hoping to increase the percentage of women faculty in STEM. Because a larger presence of mothers could help dispel the specter of motherhood, policies that aim to recruit and retain them, such as tenure clock extensions that require parents to opt out rather than in, are necessary. But our work also reveals that interventions that target the content and tone of cultural narratives about motherhood are critical. Programs that raise awareness about the many success stories of mothers in academic STEM or that mitigate motherhood bias via modifications to graduate advising guidelines and hiring and promotion procedures would be especially beneficial. For example, the findings from this project are being used to develop one such program on our campus. In collaboration with a team of administrators and media professionals, we are producing a video series that exposes graduate students to a more balanced dialogue about motherhood. The series features a diverse group of mothers who are STEM tenure-track faculty on campus. Faculty members relate both challenges and rewards associated with combining motherhood with an academic career (e.g., they describe how the career provides flexibility and job stability and how motherhood has brought personal joy and a broader perspective to their academic work). When filming and editing is complete, the videos will be made available online and will be widely publicized to graduate students.
The scope of our study is broader than that of other interview studies of gender in STEM because it includes both women and men, across many disciplines, at four universities. This approach enabled us to identify themes that cut across multiple settings. However, we cannot offer a fine-grained analysis of how experiences systematically differ by setting, given that the number of respondents within any single lab, discipline, or institutional type is low. By design, our qualitative study focused on an in-depth understanding of experiences and perceptions. A non-random sample of 57 participants is too small to conduct detailed comparisons among disciplines and institutions by gender, race, ethnicity, or national background. Future studies should engage in purposive sampling on the basis of these characteristics so that such an analysis is possible. Furthermore, although our qualitative approach was well suited for describing cultural dynamics, our results are not generalizable. A survey could better identify the prevalence of, and variability in, the specter of motherhood, its effect on career aspirations, and the relevance of factors such as race, ethnicity, and nationality. Finally, our study is focused on academic science and engineering. Future studies should assess whether the specter of motherhood is present in other disciplines and professions, and whether it may contribute to gender inequality in those contexts. If the content and circulation of narratives about motherhood in those workplaces is similarly powerful in shaping women’s career aspirations, then measures that target such narratives will be critical for addressing the stalled progress toward gender equality in elite professions.
Footnotes
Author’s Note:
Both authors contributed equally to the work. We thank Weihua An, Stephen Benard, Jessica Calarco, Youngjoo Cha, Jennifer Lee, Andrew Halpern-Manners, Elaine Hernandez, Lauren Rivera, Rosalind Thébaud, Ellis Gonzales, and Charlotte and Felix Philipp for their helpful comments and suggestions, and Katelynn Bishop, Heather McLinn, and Annie Russian for skilled research assistance. Support for this project was provided by the Center for the Study of Social Organization at Princeton University, the Institute for Social, Behavioral, and Economic Research at the University of California, Santa Barbara, the Robert Wood Johnson Foundation Health & Society Scholars program, and a grant from the National Science Foundation (NSF Advance Institutional Transformation Award [HRD-0547373]).
Sarah Thébaud is Associate Professor of Sociology at the University of California, Santa Barbara. Her research identifies cultural, social psychological, and institutional processes that contribute to gender inequalities in the workplace, families, entrepreneurship, and higher education. She earned her PhD in Sociology at Cornell University and was a postdoctoral fellow at Princeton University.
Catherine J. Taylor is an Associate Professor of Sociology at the University of California at Santa Barbara and a mother. Her main research and teaching areas are gender, work and occupations, social psychology, health, and methods. Before joining the faculty at UCSB, Professor Taylor earned her PhD in Sociology at Cornell University, was a Robert Wood Johnson Health & Society Scholar at Columbia University, and was a faculty member at Indiana University.