Women in Engineering: A Review of the 2022 Literature

Social science research on the state of women in engineering and STEM continued unabated in 2022. This year, we identified more than 200 publications and conference papers on a wide range of topics pertaining to diversity in STEM. We review here the most important, substantial research findings from this year’s literature and their implications for future research and/or policy interventions.

Last year’s literature review provided an overview of 20 years of social science research on women in engineering as analyzed in SWE’s previous two decades of literature reviews. There, we identified a number of questions on which researchers appeared to have arrived at a consensus, and a number of others where debate was ongoing. The research published this year did not break much new ground. For the most part, it examined well-established research questions, either refining existing knowledge or confirming the results of earlier studies. Discussion of the areas of disagreement we identified in last year’s review continued in this year’s literature, as scholars debated why young women are not more attracted to engineering, whether recruitment or retention should be the primary focus of efforts to improve diversity, and what sorts of interventions are most likely to produce significant change.

There were, of course, some differences between this year’s published research and the material we reviewed in the past. For example, there was an unusual amount of research on women’s performance and confidence in math and computer science this year, although it largely confirmed the emerging consensus that women’s confidence about math, not their skills, is the problem to be addressed. There was also an increased amount of interest in the impact of the COVID pandemic on the status of women in engineering and science, most of which emphasized the negative consequences of the pandemic for diversity in STEM. We explore this portion of the literature in the concluding section of this review.

Another overview of the literature published this year examined the question of whether research is focusing on sex or gender. Mary Frank Fox (2022) and her colleagues reviewed social science research on women, science, and engineering over a 46-year period (1965–2010), finding that there has been a shift toward a focus on gender, both in the number of publications that appear and in the frequency with which articles are cited. This shift has significant implications, as it signals movement toward a view that diversity in engineering and science can be achieved through policy interventions. As Fox et al. argue:

“Toward these policy decisions, the more the differences are thought to be grounded in gender rather than in sex, the more they are perceived as malleable and subject to initiatives for change at the individual level and/or at the level of institutions.” (806)

They acknowledge that contemporary researchers also often focus on issues of “difference,” which can lead to new forms of biological “essentialism.” But much of the literature on difference also calls for equity, insisting that differences need not be an obstacle to change. Rather than pressing for equality, i.e., treating everyone in the same way and thereby ignoring real differences among different groups of people, advocates for equity acknowledge difference and insist that efforts be made to ensure that everyone has the resources needed to be successful. The shift in emphasis identified by Fox and her colleagues suggests that researchers are increasingly optimistic that progress toward diversity in engineering and science can be achieved.

That said, there is also reason to be concerned. Women’s share of STEM employment and degrees has increased over the long term, but that progress has slowed considerably and even plateaued in recent years, as was documented in last year’s long-term review. As mentioned above, the short-term impacts of the COVID pandemic on women in STEM have largely been negative; if they are sustained, there is reason to believe that the progress women have made will continue to stall or even be reversed. It is, thus, all the more important that high-quality research on the state of women in engineering and science be supported and that the results of that research inform interventions and policies that seek to promote greater diversity in STEM.

Math skills, computer skills, and confidence

The idea that girls are less capable than boys in math and differ in spatial skills was a central element in traditional sex-centered explanations of the low numbers of women in engineering and science. This type of argument has become less and less common as evidence has accumulated that girls’ and boys’ math abilities are not significantly different. Tian et al.’s (2022) longitudinal study of 690 fourth, fifth, and sixth graders did find that girls performed less well on tests of spatial skills, and that these skills predicted STEM major choice later in life. They don’t regard this difference as immutable, however, and actually discuss ways in which girls’ spatial skills can be improved, making them more likely to select STEM majors in college.

Most of the rest of the research we reviewed this year extended the argument that it is not girls’ math skills but girls’ confidence in math that should be the focus of attention, adding that a similar argument can be made about computer skills. Zhao and Perez-Felkner (2022) analyzed data on a very large sample of students who participated in the National High School Longitudinal Study in 2009, finding that students who perceived their math ability as strong were more likely to major in STEM subjects. Girls were less likely to perceive their math ability as strong, although the researchers added that girls who see their ability as strong were more likely than comparable boys to gravitate toward the social or natural sciences, not computer science or engineering.

The effects of differential math confidence continue among older students, as John, Vierra, and Robnett’s (2022) research on 754 college undergraduates showed. They classified students into those with a “high math self-concept” and those with a “low math self-concept” and found that women were overrepresented in the latter category. Their research showed that students with high and low math self-concepts reacted differently to challenges and low points in their experiences with math. High math self-concept students tended to blame problems on their teachers and/or to talk about how a bad experience spurred them to work harder. Low math self-concept students were much more likely to talk about math anxiety, aversion, and how their bad experiences had turned them away from the subject. Battey et al.’s (2022) description of the coping strategies minority and female students develop to deal with bad experiences in calculus demonstrates that some students are able to overcome bad experiences in math. But, John, Vierra, and Robnett’s study indicates that students with a low math self-concept, who are disproportionately female, are more likely to turn away from math and math-dependent fields such as engineering.

As computer skills have become increasingly central to success in STEM majors and careers, a similar discussion of the role of computer skills and computer self-confidence has developed. As with math achievement, the gender gap in computer science achievement is narrowing. Bahar et al. (2022) examined more than 20 years of available data from three advanced placement computer science programs, finding that the number of girls participating is rapidly catching up with the number of boys (they anticipate parity in less than two decades), and that the ratio of boys to girls who scored a 5 had been reduced from 8:1 to 2:1.

Computer science self-confidence and self-concept are another matter, however. Ashlock et al.’s (2022) mixed methods study of middle school students in three school districts in the southern United States found that the gender gap in computer efficacy was twice as large as the gender gap in math efficacy, despite the fact that boys and girls earned similar grades. This gender gap in computer self-confidence continues into adulthood. Liberatore and Wagner (2022) asked 239 professionals to complete a task on one of three platforms and to answer questions of varying complexity about those tasks. They found few differences between men’s and women’s performances, but the female participants’ self-efficacy was significantly lower, despite their performance. Neither of these studies links women’s self-evaluation on computer skills to their choice of major or career, but it is possible that, as with math, lower self-confidence in computer science may be discouraging even some high-achieving women from pursuing careers in engineering.

Early influences on major and career choice

Over the years, many researchers have found that gender stereotypes about science and engineering develop very early in children’s lives. These early stereotypes are of great importance, as researchers have also found that interest in engineering, or an aversion to it, develops early. Moloney and Ahern (2022) studied more than 600 high school students in Dublin, Ireland, finding that young women were much less likely than young men to express an interest in studying engineering in university. The researchers attribute this not to different levels of achievement in engineering-related subjects, but to their often inaccurate feeling that they weren’t good at math and science, a fear that they wouldn’t be successful, and a perception that engineering was dominated by men. These attitudes had developed long before the point at which students make a final decision about what their college major would be, underlining the importance of influences earlier in childhood.

Several studies published this year confirmed that children continue to hold gender stereotypes about engineering and science. Canessa-Pollard et al. (2022) asked nearly 200 children ages 6 to 11 in the United Kingdom to describe the gender ratio of 24 occupations and to indicate whether men or women were more capable of performing those jobs. The children assigned men and women to stereotypically male and female occupations and tended to gender-type competence as well. However, girls’ competence ratings were less stereotypical than boys’ and both boys’ and girls’ stereotyping of competence in male-dominated occupations decreased with age. Interestingly, stereotyping of competence in predominantly female occupations did not decrease with age.

Sims et al. (2022) asked a sample of 144 young people, ranging in age from 5 to young adulthood, to comment on two vignettes involving science supplies, one of which showed boys as advantaged (more supplies), the other showing girls as advantaged. Overall, respondents saw the vignette in which boys had the advantage as less wrong. The researchers also found that many respondents justified disadvantaging girls using moral reasoning, implying that they felt boys deserve or would use resources more than girls. Still, there were complexities that offer hope that stereotyping can be overcome. Older respondents tended to see the vignette in which girls were disadvantaged as more wrong than did younger ones, suggesting that gender stereotypes about science may weaken with age.

Other studies published this year examined small-scale interventions designed to counter stereotyping in young people and found that, at least in the short term, they were effective. Shachnai, Kushnir, and Bian (2022) conducted an experiment involving boys and girls participating in a science game. Girls were substantially less likely to persist in the game than boys, but being told about a female scientist (Marie Curie) virtually eliminated the gender difference. Similarly, Yabas et al. (2022) found that participation in a “girls meet science” program significantly increased STEM interest among a group of 6- to 10-year-old girls in Turkey, while Buckley, Farrell, and Tyndall (2022) found that being told stories about female scientists reduced stereotyping in a group of British schoolgirls.

If stereotypical attitudes among children can be modified, and if girls’ self-confidence in engineering-related subjects is a concern, teachers could be a powerful force encouraging more girls and young women to consider engineering careers. Unfortunately, much of the research on the role of teachers published this year indicated that some may be inadvertently perpetuating the lack of diversity in STEM. Fox-Turnbull et al.’s (2022) study of student teachers in New Zealand found that they held stereotypical views of engineers (good analytical thinkers with poor social skills) and had gender-stereotypical views about which children would make good engineers. Wammes et al. (2022) studied a group of teachers in the Netherlands, finding that they had gender stereotypical expectations of their students’ technical skills. Burgess et al. (2022) examined data from Statistics Denmark’s administrative high school data base to analyze the gender gap among students who subsequently completed a math or math-demanding degree program. They found that the gender gap among students who participated in an external assessment in math was much smaller than among students who were evaluated only by their teachers. They conclude that teachers’ assessments may be discouraging some capable girls from pursuing math and math-demanding majors and fear that the fact that the external assessment was discontinued during the COVID pandemic may lead to widening gender gaps in STEM careers.

Copur-Gencturk, Thacker, and Cimpian (2022) examined grading patterns and expectations among teachers working in virtual classrooms in the United States during the COVID pandemic. They found no evidence of gender or racial bias in grading but did find that teachers were more likely to recommend male than female students for gifted programs based on the same work. They also found that teachers were more likely to recommend Black students than white students for individualized education programs, again based on the same work. The possibility that teacher beliefs and attitudes may be contributing to racial as well as gender gaps is also raised in McNeill, Levya, and Marshall’s (2022) study of instructors in undergraduate calculus courses. They interviewed seven instructors at a large, public, historically white university in the Northeastern U.S. and found that the instructors defined math and math instruction as an objective, color-blind space where race and gender were meaningless. The researchers argue that this attitude absolves instructors from having to consider the social determinants of success in mathematics, enables them to define it as purely meritocratic, and thus to persist in practices that help to perpetuate lower levels of success among minority students.

Of course, many teachers are strongly committed to encouraging diversity, nor are teachers the only influences on young people’s decisions about whether to major in engineering or science. Nguyen et al.’s (2022) study of more than 100 young women in high school who participated in the SWENext program offers insights into the complex set of influences at work. Respondents, all of whom had expressed engineering aspirations, were asked to answer a set of survey questions about individuals who supported or discouraged them. Thirty-three of the respondents were also interviewed about their experiences. The survey results indicated that the young women had received a great deal of support, although more from adults than peers, and more from female than from male peers. The interviews revealed that teachers, both male and female, had been particularly supportive (countering the negative view of teachers in the studies discussed above).

In contrast, young male peers, particularly in STEM spaces, had been discouraging. Respondents reported having been excluded physically by male peers who took control of material, excluded socially (ignoring, avoidance), or denied recognition. These young women persisted in the face of this discouragement. But their accounts are consistent with those reported in research and memoirs discussed in earlier literature reviews that found that women interested in engineering often encountered a chilly, hostile climate that caused some of them to look elsewhere.

The university experience

As in previous years, many researchers focused their attention on the experiences of students who entered university-level engineering programs. Many studies reported that female and minority students struggled psychologically as they pursued engineering and science degrees. Wilkins-Yel et al. (2022) interviewed a diverse group of a dozen female graduate students in STEM, finding that they all had encountered a pervasive lack of support from significant others in their programs and repeated encounters with subtle and overt forms of sexism and racism. All experienced psychological distress, and several described more severe mental health issues. Several decided to leave, saying they were trying to protect their mental health.

Danowitz and Beddoes (2022) surveyed more than 900 first- and second-year engineering students from eight U.S. universities in the period just before the outbreak of COVID-19. The survey revealed that half had experienced significant mental health problems, including depression, anxiety disorders, eating disorders, and post-traumatic stress disorder, but most reported never having received a diagnosis. Female students were more likely than male students to screen positive for anxiety disorders or a major depressive disorder. Students with disabilities were more likely to suffer from mental issues than students who reported no physical disabilities, and Hispanic students were more likely than other groups of students to experience a major depressive disorder.

Other studies described experiences of self-doubt among women and minority students. Dumitru et al. (2022) conducted a mixed methods study of 120 students and tutors in STEM programs at a New York area university. They found that female students and tutors experienced more math anxiety than males and that male participants expressed more confidence than their female counterparts. Male tutors also rated their own performances more highly than female tutors.

Atwood, Gilmartin, and Sheppard’s (2022) study of more than 6,000 engineering juniors and seniors also found gender and racial differences in self-efficacy. For female students, this correlated with reduced intentions to pursue a career in engineering, but no similar relationship was found for minority students. The authors speculate that minority students may learn to activate interpersonal relationships to counter negative experiences, enabling them to persist.

Lapan et al. (2022) interviewed a diverse group of 13 college women majoring in computer science and participating in an internship program through a public university in the Southeastern United States. They reported that all participants experienced impostor syndrome and questioned their competence and readiness to participate in the internship. The students felt they had to prove their worth for being in the internship. Many also reported noticing the lack of other women on the teams that they joined. Those students who experienced more diverse teams felt a stronger sense of belonging. The authors emphasize the importance of these internship experiences, arguing that they affect students’ decisions about whether to pursue a career in the field.

Several studies published this year emphasized the importance of viewing the experiences of engineering and science students through an intersectional lens. For example, Campbell-Montalvo et al. (2022) interviewed 29 undergraduates, finding that students who were sexual minorities continue to experience problems with “fit” as a result of things such as refusal to use preferred pronouns and negative reactions to self-expression that did not fit gender norms. Problems of “fit” were greatest for students who were also racial minorities. Sexual minority students valued having similar peers and faculty members with whom to share their experiences. Casper, Atadero, and Fusilier (2022) note there is a tendency to lump all sexual minorities together under the omnibus LGBTQ+ heading, as if all these identities were the same. Their survey finds that respondents would like to see more research that allows them to self-identify more clearly and that this would result in a more nuanced understanding of the experiences of sexual minorities in science and engineering.

Park and Rottinghaus’ (2022) study of 585 female STEM students of color points to the importance of attending to differences among these groups and avoiding lumping them together under the heading “minority.” They found that these groups differed in a variety of ways. For example, Black female students reported higher levels of academic self-efficacy than the other groups, while Latina students were more likely to have “proactive personalities,” and Asian students reported lower levels of discrimination. Riegle-Crumb et al.’s (2022) study of 33 SWENext participants showed how different groups of pre-college young women respond differently to gender inequality in engineering. Black respondents were most likely to engage in critical reflection on inequality while Asian respondents were least likely. Black and Latinx respondents were more likely than others to indicate a willingness to challenge gender inequality in their future careers, while white respondents were more likely to endorse an individualistic “lean-in” response. Spencer et al.’s (2022) study of 21 Black graduate students in engineering and computer science found that they felt compelled to “code-switch,” i.e., to modify their appearance and/or speech patterns in order to avoid negative stereotypes, and that this could be a source of stress they needed to find ways to manage.

The fact that female and minority engineering and science students experience various forms of psychological distress and self-doubt has led to much discussion of whether they leave engineering programs at higher rates than men. In last year’s overview of research on this topic, we described the ongoing debate as to whether retention rates for women and minority students are lower than for white males and noted that many studies don’t support this conclusion. Koch et al. (2022) published research this year on 2006 to 2009 College Board data on a sample of more than 40,000 seniors who expressed an interest in pursuing a STEM major. They followed up on those students to see who had persisted in STEM, finding that men persisted at significantly higher rates than women (65% vs. 48%). This is a large-scale, carefully constructed study, so it may provide strong evidence that STEM retention rates for women are indeed lower than for men. However, the authors used intended major from the College Board data as the starting point, not declared major in the first year of college. So, it is possible that many students who said they intended to pursue a STEM degree changed their minds before even entering college. It also should be added that the retention rate for women was highest in engineering, so it is possible that the overall gender gap in retention is the result of patterns in other disciplines.

Park et al. (2022) noted that the percentage of women and minority students earning STEM degrees continues to be higher than their share of workforce jobs in these fields, so it is important to examine factors that cause graduates to stay after they earn their degrees. They used data from the National Longitudinal Study of Freshmen on more than 500 students who declared a STEM major on entering college, stayed in a STEM major after their fourth year of college, and who identified a career or graduate study plan after graduation at the end of the fourth year of college. They found some differences between female and male students and minority and white students. Specifically, women were more likely to report that experiences such as volunteering and extracurricular activities were important influences on postgraduate plans. Underrepresented minority students were also more likely to list volunteering and extracurricular activities as influential in pursuing STEM graduate study as majority-status peers. Unfortunately, Park et al. focused their analysis only on students who intend to continue in STEM careers or graduate study and didn’t compare them with those who left. Thus, it is difficult to know whether these differences account for different levels of persistence in the field.

What is clear is that, if women leave engineering and science programs at higher rates than men, it is not because of poor academic performance. Maries, Whitcomb, and Singh’s (2022) analysis of more than 18,000 student records from a large public university found that female students in STEM earned higher grades than male students in STEM, whether or not they stayed in a STEM major. They also found that women who dropped computer science or engineering majors had GPAs comparable to male students who stayed.

Barth and Yang’s (2022) study of 251 STEM students’ goals implies that another explanation frequently given for women’s leaving engineering (or not being attracted to it in the first place) may not be as persuasive as some think. Many have argued that female students’ interests and goals are different from men’s and that women struggle to find ways to realize their goals within engineering and science. However, Barth and Yang found that both male and female STEM students’ goals change as they progress through their programs. They found decreases over time for both men’s and women’s career-status goals, while men’s marriage-family and social-impact goals both decreased, but women’s did not. The result was that more advanced female STEM students valued marriage-family and social-impact goals more highly than men, but this did not prevent them from staying in STEM. Barth and Yang speculate that these women may be able to visualize ways to realize their goals within STEM. Perhaps more importantly, they show that students’ goals are not fixed, so assuming that their goals at a single point in their careers inevitably determine what they will do later is probably ill-advised.

There are many programs designed to improve the retention of female and underrepresented minority students in engineering and STEM, several of which were described in research published this year. Carrigan et al. (2022) reported on positive results from the University at Buffalo’s Navigate program, which involves using case studies to help STEM graduate students recognize gender bias and learn how to overcome it in the workplace. Zurn-Birkhimer and Serrano (2022) examined data on students who participated in the Women in Engineering Program at Purdue University between 2009 and fall 2014, finding that the various components of the program helped improve completion rates for female and minority students. Knezz, Pietri, and Gillian-Daniel (2022) reported that participation in an online equity training course helped graduate students to recognize gender bias and increased their motivation to counteract it. Rincon et al. (2022) provided preliminary evidence on a community college networking program, finding that exposure to diverse women in STEM is beneficial to undergraduate students’ intentions to continue their path toward a career in engineering or computer science.

While all these programs show promise, most involve small numbers of students and demonstrate only short-term results. Lagesen, Pettersen, and Berg’s (2022) study of participants in an Ada program in Norway pointed explicitly to the need to find ways to sustain the gains achieved in this way. Ada is a program involving outreach, recruitment, and efforts to create and sustain a supportive environment for female students who participate. The researchers found that the program improved the retention of women (and men) in engineering, but they added that they feel the gains are not sustainable unless continuous effort is exerted.

Perez, Motshubi, and Rodriguez (2022) added that efforts to improve departmental climates with a view to retaining more female and minority students can sometimes be problematic. They reported on two National Science Foundation (NSF)-funded programs focused on improving racial climate, finding that, although the departments involved were participating voluntarily, faculty often resisted the work involved, meaning the burden of creating change fell on graduate students, many of whom were minority women. These students described the conflict they encountered: On the one hand, they needed to do change work to make it possible for them to do their scientific work, but their change work was regarded by faculty as a distraction from their “real” work as graduate students. Graduate students’ activity also effectively validated faculty members’ inaction.

Some observers have argued that having more female professors would increase the numbers of students who choose to major in engineering and other STEM subjects and who eventually succeed in completing those programs. Mansour et al. (2022) conducted research at the U.S. Air Force Academy on students who graduated between 2004 and 2008 to see if they could find evidence of this. The Air Force Academy is an unusually interesting site for this kind of research, as first-year students are randomly assigned to math and science classes. Thus, self-selection of instructor was not a factor. The researchers found that having a female professor during the first year had a positive effect on some female students, but only those in the upper quartile of the ability distribution. Those students were more likely to complete a STEM degree, to earn better grades, to complete an advanced STEM program, and to go on to a career in a STEM occupation. There was, however, no effect on other female students, and the upper-quartile male students who had a female professor in the first year were less likely to pursue a STEM career than those who did not.

Finally, mentoring has frequently been mentioned as an important element in any program to improve diversity in engineering and science. Beck et al. (2022) provide a critical review of almost 50 mentoring programs that were documented in the literature between 2010 and 2018. They praise some of these programs but also note that many are limited in scope and fail to target intersectionality or power relationships at work. They are critical, as well, of the fact that many programs seem designed “…to help mentees navigate hostile work environments rather than by addressing aggressive individuals who negatively impact a collegial climate within the institution.” (180)

Beck et al. are also critical of some mentoring programs’ failure to pay attention to matching mentors to mentees. Genovesi et al.’s (2022) study of EngWINS, a mentoring program for low-income high school students, attempts to address this criticism by training mentors in culturally responsive pedagogies. They found that the mentors responded positively to the training, but were not confident that they were well-prepared to mentor students with whom they did not have much in common. Student participants indicated that the program expanded their horizons and knowledge of engineering, but they were unclear as to what mentorship meant and who their mentors actually were.

Johnson and Pietri’s (2022) experimental research points to another way in which mentoring might be made more effective. They analyzed the circumstances under which a white male faculty member might become a viable ally for a white female student. Endorsement of the professor by either a white female or a Black female was shown to promote perceptions that the faculty member was indeed an ally and could enhance feelings of belonging. Of course, this is experimental data; in the real world, one would need to examine whether the male faculty members identified as potential allies actually follow through and provide effective guidance to the female students to whom they are recommended.

Careers

In past literature reviews, we noted that the bulk of research on working engineers focused on the experiences of researchers and faculty members in universities. Rather less research was devoted to the much larger group of practicing engineers outside the academy. Surprisingly, and happily, the pattern this year was reversed, with more attention being paid to the experiences of practicing engineers.

Academic engineers

The research on academic careers published this year focused on negative aspects of the academy, pointing to the continued disadvantages experienced by women and minority STEM faculty. Rennane et al. (2022) analyzed data from several large, national data sets (including the Survey of Earned Doctorates), showing that female faculty in STEM are more likely than men to hold nontenure track positions and that these positions are less well-paid and offer lower job satisfaction, job security, and opportunities for advancement. Goodman (2022) examined data from a large number of universities (ranging from Ivy League schools to HBCUs) and demonstrated huge gender and racial gaps in the likelihood that faculty will be named on university patents: Men are 5.43 times more likely to be named than women, white people are 24.33 times more likely to be named than Black people, and white people are 57.68 times more likely to be named than Hispanic faculty. Goodman’s data did not allow her to explain the reasons for these discrepancies, although she does point to other literature that emphasizes differential teaching burdens, expectations, resources, and other factors. Nevertheless, the size of the gender gap she documented is clear evidence that equity in the academy has not yet been achieved.

Trapido (2022) conducted an interesting analysis of engineering faculty specializing in information theory. His goal was to determine whether male and female faculty receive the same level of recognition for “novel” ideas. He compiled and analyzed a comprehensive data set on the creative output of academic engineers specializing in information theory, finding that female authors received fewer citations for work that was identified by others as “novel.” He explains this result by arguing that people are generally unsure how to assess work that is novel, so, in making a determination, they are likely to rely on status characteristics such as gender, prestige of the author’s mentor, or prestige of the institution from which the author graduated. All of this works to the advantage of male authors, who are more likely to possess high status characteristics. Ross et al. (2022) also found that women receive less credit for their scientific work than men. Their analysis showed that at least part of the explanation for this is that women are less likely than men to be listed as an author on work produced by teams of which they are a part.

Researchers also reported that gender was a factor affecting the tenure process in university STEM departments. Gregor et al. (2022a) interviewed more than 100 pre-tenure female STEM faculty at a Midwestern university, finding that female faculty in departments where there was less gender diversity were more likely to report experiencing discrimination, difficult colleagues, and a lower level of support from administration. However, Gregor et al. (2022b) found that encountering barriers such as these actually tended to increase respondents’ career aspirations, a result that might suggest that they will find a way to persevere in their careers. The same study found that some female faculty suffer from “impostor syndrome” and that this correlated with negative tenure expectations on their part. Finally, Griffith et al. (2022) studied more than 400 STEM faculty members at a Northeastern university, finding that having an “underrepresented identity” (i.e., being female, minority, or both) was related to a more negative perception of faculty relations within the department, more negative perceptions of gender and race equity, and more negative views of the transparency of tenure and promotion criteria.

Even seemingly “objective” rating systems were found to work to the disadvantage of women in academic STEM. Blair-Loy et al. (2022) examined the use of a hiring rubric in an engineering department at a Research 1 university. The rubric was designed to reduce the influence of subjective judgments and biases in the hiring process. The researchers found that gender bias persisted in spite of the rubric — reviewers underrated the contributions of female candidates and made more positive and fewer negative comments about male candidates. The rubric also did not address other identities that could be the cause of bias, such as race. Blair-Loy et al. recommend that rubric usage be accompanied by “…strategic application in departmental meetings to counteract individual bias and check interactional bias during the discussion of candidates.” (37) Nygaard, Piro, and Aksnes (2022) showed how an “objective” publication rating system in use within the Norwegian university system assigns more points to men than women. They found that small, but measurable gender differences in each of the criteria being used to rate publications’ significance and impact produce a more significant cumulative advantage for male authors. They added that similar publication rating systems are in use in many countries in Europe, so there is a real danger that female faculty are not receiving as much credit for their work as their male colleagues, resulting in lower salaries and slower career advance.

Engineering practice

Female engineering graduates who leave the profession after earning their degrees have been the focus of a growing body of research. Several articles published this year contributed to this discussion. Delaney and Devereux (2022) analyzed data from the U.K. Quarterly Labour Force Survey for the years 1960–2020 and concluded that female STEM graduates are much less likely than males to work in STEM occupations. The gender gap varies by field of study, with engineering and technology having one of the largest, as much as 20% five years after a degree is awarded. Women who leave STEM are more likely than men to enter fields such as education and health, which is partly responsible for the earnings gap between male and female STEM degree recipients.

Skrentny and Lewis (2022) offered a somewhat different view of the retention issue. They used U.S. data from NSF’s biennial National Survey of College Graduates to track employment outcomes for STEM degree recipients. In their view, the retention problem has been misinterpreted because analysts use too narrow a definition of STEM employment, i.e., they generally define only clearly identified STEM occupations (engineer, chemist, mathematician) as STEM employment. This misses the increasing number of jobs in which STEM skills are an important part of the work being done, which they refer to as the “stemification” of employment. Using a broader definition of STEM employment, they found that retention gaps for minority STEM graduates are much smaller than previously thought. The gap for women is also not as large but is still a sizable 16%. There are sure to be questions asked about Skrentny and Lewis’ definition of STEM employment. For example, they emphasized the need to acknowledge the many STEM graduates who move into managerial roles, a phenomenon that has been identified as one of the ways in which women can be marginalized in fields such as engineering since it takes them away from core technical work. Nevertheless, Skrentny and Lewis make a strong argument that measures of STEM employment and retention may need to be refined.

Several articles published this year explore the reasons female engineering and science graduates leave the field. Beddoes (2022) interviewed 18 early-career civil engineers, the majority of whom were female, about their experiences at work. She found that women reported a range of problems, including having their competence doubted, being sexually harassed, and gender discrimination in job assignments and performance evaluations. As a result, the women felt a lower sense of commitment to their employing organization, and several were already considering leaving the field altogether.

There may also be differences in the reasons women and men give for considering leaving engineering and STEM jobs. Smith, Costello, and Chowdhury (2022) surveyed more than 400 engineers employed in the private sector in New Zealand. They found that women’s leaving intentions tended to be motivated by different factors than men’s. Where men were more likely to stress ambition and the ability to effect change quickly, women emphasized feeling authentic and being able to build connections to higher-ranking professionals. Osten (2023) analyzed Canadian data from the National Graduates Survey of 2013 and found that women are more likely than men to look for a job with another employer when they feel overqualified for the work they are doing, are supervising someone at a job, are a visible minority, or when they have children.

Retention rates were not the only employment-related issue on which research focused this year. Murciano-Goroff (2022) offered an explanation of how gender can become a factor in recruitment, focusing on the case of software engineers. He examined data from a recruiting platform where individuals can self-report their computer programming skills, finding that women are less likely than men to self-report knowledge of programming languages and that these self-reports are a factor in expressions of employer interest. Further, he found that recruiters were somewhat less likely to express interest in a female candidate than a male candidate with comparable observable qualifications. In sum, a combination of women’s own behavior and recruiters’ apparent bias combines to reduce the possibility that a woman will be recruited for positions in this field.

Several studies also noted gender inequalities in entrepreneurial activities. Schillo and Ebrahami (2022) used data from a commercial data base, Capital IQ, to examine investors’ attitudes toward female-led startup companies. They found that the involvement of women in leadership is associated with lower levels of funding in all fields considered but was more pronounced for digital startups than biotech startups. Nigam, Benetti, and Mavoori (2022) found a similar pattern of lower venture capital funding for female than male entrepreneurs in 100 Indian digital startups, but argue that much of the difference is attributable to educational differences between the male and female entrepreneurs in their sample. Macchione et al.’s (2022) experimental study found that groups of undergraduates and respondents recruited from Amazon Mechanical Turk perceived male-led STEM companies more positively, which may offer an explanation of the gender bias in investors’ choices.

Gender is also a factor in promotion to management in technical fields. Yates and Skinner’s (2021) interviews of 32 female engineers in the U.K. revealed that they felt that promotions were most often given to people who were “well-known,” who had “high status,” and who conformed to “ideal worker” expectations, i.e., who were men. Respondents were, however, reluctant to frame this pattern as “sexism.” Adams and Flores (2022) show a similar pattern with respect to minority identity. Using data from the Ontario Society of Professional Engineers, they found that minority engineers in Canada were half as likely as their white counterparts to hold managerial positions and exercised less managerial authority on the job. This inequality affected minority women in particular, pointing to the impact of intersectional identities.

Varma (2022) also emphasized intersectionality in analyzing the situation of women in science and engineering. She focused on the experiences of Asian American women, arguing that they face a “double bind.” They may have superior academic credentials but face a “prove-it-again” bias once they are hired. They are viewed as passive, but face negative reactions if, as women, they act assertively. As women, they are viewed as less competent, although Asian males are typically seen as competent. And, both Asian men and white women have a better chance of being promoted.

Kim and Meister (2022) described the existence of microaggressions against women in technical fields, based on interviews with 39 women leaders in STEM from the U.S. and Canada. Their respondents reported experiencing and observing a variety of microaggressions: having their technical competence devalued; being ignored or interrupted; denials of the existence of gender bias; abnormalizing female values and communication styles; and highlighting female characteristics in an unwanted, negative way. Dealing with these experiences can be difficult. Kim and Meister’s respondents described experiencing a range of negative emotions, but found that having allies can help. Cruz and Nagy (2022) surveyed approximately 500 female scientists and found that the most common response is what they call a “preservationist” response, which involves the lowest level of stereotype threat management and involves lower levels of thriving. Wolfe and Powell (2022) conducted a series of experiments with groups of female engineers to determine whether being prepared with a “script” for managing conflict would help women avoid backlash when they respond to problems. The goal was to help women avoid emotional reactions and to focus on solutions that would benefit everyone, but the results were rather mixed, leaving open the question of how much such scripted responses to situations will actually be of use.

Two studies did find that women can and do make use of networks in dealing with the various situations they encounter in engineering and scientific workplaces. Papafilippou, Durbin, and Conley (2022) interviewed 48 female engineers, mostly in the U.K., about their use of networks. They found that respondents used internal networks to create a “safe space” from which to defend themselves in a male-dominated environment. External networks were also important, both as sources of validation and psychological support, but also as resources for building careers and engaging in broader change activities. The study’s authors questioned the widely held view that women’s networks are not very powerful or effective and urged employers to encourage their development and recognize their worth. Jung and Welch’s (2022) study of almost 10,000 female academics in STEM also found that women benefit from having external networks that are demographically similar. However, they found that female academics felt more included if they had more diverse internal networks that included men. Perhaps the fact that this study focused on feelings of inclusion, rather than on coping with negative experiences, explains the differences in outcome.

Considering the impacts of COVID

The literature on women in engineering and science has generally reflected an expectation that women’s share of employment in STEM will gradually increase, eventually reaching equality with men. And, for a time, the progress made over the last 40 years made that expectation seem achievable. In the last two decades, however, as progress has slowed, a note of concern has crept into discussions regarding efforts to diversify engineering and science. And, as noted in the introduction to this review, the growing body of research on the effects of the COVID pandemic on STEM employment suggests that expressions of concern are likely to continue.

It has been widely reported that the workplace effects of the COVID pandemic had a more significant impact on women. Women made up the bulk of the workforce in the most-affected sections of the economy (service and administrative jobs). And, while the dramatic expansion of remote work options during the pandemic made it possible for many workers, including many women, to remain employed, parents found that closed schools and child care facilities made balancing work and caregiving responsibilities more difficult. Parents working from home had to find ways to work while keeping an eye on their kids and overseeing their online educational experiences. Parents who were not able to work from home had to decide whether to give up their jobs in order to care for their children when schools and child care facilities were not available. By all accounts, the people most affected by these problems were mothers and people with lower incomes, many of whom were members of minority groups.

Research on academic STEM employees confirms that the effects of the pandemic fell disproportionately on women and minority engineers and scientists. Caldarulo et al. (2022) conducted surveys of academic scientists at Research 1 universities in the United States, finding that, in both 2020 and 2021, female respondents had more domestic responsibilities and that this interfered with their research productivity. Dunn et al.’s (2022) study of 84 female untenured STEM professors echoed these findings, reporting that mothers had less time for academic tasks, experienced higher levels of stress, and feared that their tenure journeys had been disrupted. Douglas et al.’s (2022) survey of more than 4,000 early-career academic scientists added that underrepresented minority respondents reported more significant effects of the pandemic than other racial groups. These included a greater increase in workload; growing concern about career advancement; and more work disruptions from COVID-related physical health problems, mental health problems, and increased caretaking responsibilities. All these studies emphasized that universities missed opportunities to provide support for graduate students and junior faculty in the early stages of their careers.

Considerably less research on the effects of COVID on nonacademic engineers and scientists was published this year. The one study that was published, however, reported on an interesting “natural experiment” showing how effective interventions could reduce the impact of COVID on engineers and scientists. Early in 2020, before the outbreak of the pandemic, Melin and Correll (2022) collected baseline data on a group of early-career women employed by a biotech company. They were interested in assessing the effects of an online intervention on employees’ soft skills. After the pandemic-induced lockdown, they continued with the intervention and studied its effects. They found that those women who received the intervention reported improvements in their soft skills, while those who did not reported decreases. Further, managers’ assessments of the treatment group improved more than for those in the control group, and this was associated with improved chances of retention. This is at least suggestive evidence that, for women who did not receive this kind of support, the COVID pandemic may have had negative effects on their performances (although it is also possible that the same can be said of men).

COVID’s effects on women were not confined to the United States. Heo et al. (2022) conducted an online survey in six languages of 4,494 scientists in STEM fields across 132 countries during October–December 2021. They found that the percentage of participants with increased work hours was the highest among female participants, especially those without children. Disproportionately higher increases in work hours were found for teaching and administration in women than men and for research/fundraising in nonparent participants than parent participants. Female participants were more concerned about the negative impacts of the pandemic on publications and long-term career progress, and less satisfied with their career progress than their male counterparts. Female respondents also reported higher levels of stress, anxiety, and depression associated with work/life conflict.

Two other studies indicate that the effects of COVID on women in engineering and science may have varied across countries. Singh’s (2022) study of a group of female engineers in India did not find that they were differentially affected, perhaps because of the high demand for their skills at the time the pandemic broke out. Haghani et al. (2022) studied data on publications from the global citation database Web of Science, finding that the gender gap in publications during the COVID pandemic increased in some countries, but decreased or stayed the same in others.

Research on the pandemic’s effects on students provides another reason for concern about future prospects for increased diversity in engineering and science. Kahn et al. (2022) studied nearly 1,000 students in three STEM departments (civil and environmental engineering, computer science, and physics) that required lab work, computer access, or both. The study identified numerous demographic differences that affected students’ self-efficacy and motivation resulting from the shift to online learning and the loss of in-person lab experiences, including the fact that Black, low socioeconomic status, and first-generation students reported poorer internet access than students from other demographics; students who were in more precarious positions socially were more likely to opt into pass/fail for a class; first-generation women reported having the most caregiving responsibilities at home; and female students were more likely to report a decline in the quality of interactions with their peers. Overall, female students and students from low socioeconomic backgrounds were the least likely to feel prepared for the next class in the sequence. It is, thus, possible that the COVID pandemic may at least temporarily reduce the number of well-prepared female and minority students in fields such as engineering.

Conclusion — what’s next?

While it is too soon for academic research on this topic to have been published, journalistic accounts point to one other reason for concern about diversifying STEM employment: the wave of layoffs affecting technical workers, particularly in the heart of the tech industry itself. Big technology companies such as Google, Apple, Amazon, and Meta announced significant layoffs in the past 12 months, after having maintained or even expanded their workforces during the early phases of the COVID pandemic. In some cases, these companies had been attempting to recruit more diverse workforces by offering remote work options. Preliminary analyses of the layoffs indicate that they are likely to undo ongoing efforts to diversify the tech sector. For example, an analysis by Revelio Labs Inc. (2022) found that women represented 46.64% of the layoffs between September and December 2022, while Latino workers represented 11.49%. Since they represent, respectively, 39.09% and 9.96% of the workforce in the industry, the layoffs will likely reduce diversity in the tech sector. A possible reason for the effects on women and Latino workers is that many companies are focusing layoffs on more recent hires, among whom minority workers were overrepresented.

It is not clear whether this will have a significant long-term effect on the diversity of the engineering and technical workforce, as many of the layoffs are on the business side of the affected firms. Women and minority workers are more likely to be employed in these nontechnical positions, so it is possible that female and minority engineers and scientists are not being disproportionately affected by the cuts. It also appears that Black and Asian American workers were not affected in the same way, so any progress made among those groups might be sustained. Still, reports indicate that some companies have cut back their expenditures on diversity efforts, and it has become more difficult for new graduates to find jobs, reducing the chances that the percentage of female and minority workers in the engineering labor force will be increased (Nix 2023; Oladipo 2023).

All of this underlines the reality that progress toward diversity in engineering will not happen automatically, and one cannot assume that even long-term trends will be sustained. We were struck, in this regard, by Rudenko et al.’s (2022) report on trends in Russia, a country to which people once pointed as a place where women’s status in science and engineering was relatively high. This study found that there has been significant growth in gender inequality in engineering since the collapse of the Soviet Union. It is also worth noting Stoet and Geary’s (2022) ongoing research, showing that gender gaps in STEM aspirations may actually be larger in more affluent societies where women have more opportunities. So, there is little evidence that, left to their own devices, societies will “evolve” toward gender equality in fields such as engineering. This means that continued efforts by researchers are needed to understand why the engineering and scientific workforce is so overwhelmingly white and male and to help identify policies and interventions that can sustain and accelerate the progress already made.

About the authors

Peter Meiksins, Ph.D., is professor emeritus of sociology at Cleveland State University. He received his B.A. from Columbia University and Ph.D. from York University, Toronto. Major publications include Putting Work in Its Place: A Quiet Revolution, with Peter Whalley, Ph.D. (2002), and Changing Contours of Work: Jobs and Opportunities in the New Economy, 4th edition, with Stephen Sweet, Ph.D. (2020).

Claudia Vanessa Garcia is a Ph.D. candidate in the higher education administration program at Kent State University and a research intern with the Society of Women Engineers. Her research interests focus on higher education policy, access, and representation of marginalized groups in STEM disciplines.

Natali Huggins is a doctoral candidate in the higher education program at Virginia Tech. Her research expertise lies in diversity and inclusion in graduate education, particularly international, Latinx, and Hispanic graduate students’ persistence and development. She specifically focuses on supporting students in their transition, adaptability, and socialization to higher education in the United States.

Maya Menon is a Ph.D. candidate in the engineering education program at Virginia Tech. Her current research interests include engineering education for sustainable development, global engineering, faculty development, and interdisciplinarity.

Olivia Ryan is a Ph.D. student in the engineering education program at Virginia Tech. Her current research interests include curriculum barriers in engineering related to math readiness.