Career Choices of Female Engineers: A Summary of a Workshop (2014)

4

Retaining Technical Talent: Data Needs, Critical Transitions, and Career Pathways

Are more data needed to understand why women enter engineering and why they stay or leave? If so, what types of data are needed? Tom Perry, director of education for the American Society of Mechanical Engineers (ASME), led this discussion, starting with a review of ASME’s work in this area. As a point of reference, he presented data from the American Society for Engineering Education (ASEE) that he and his colleagues found particularly disturbing: Figure 4-1 shows that despite seven years of continual enrollment growth in mechanical engineering, the enrollment of women remained stuck at 11–12 percent.

But then they looked at the chart differently and realized that, for the total numbers to increase and the percentage of women to stay the same, there had to be more women. They saw that since 2005 the number of undergraduate women studying mechanical engineering has risen more than 10 percent faster than the number of men—an increase in women of over 55 percent, compared to a 40 percent increase in men (Table 4-1). ASME has begun a study of these women undergraduates to understand the meaning of this increase for mechanical engineering education, precollege education work, and so on.

ASME is also interested in finding the right mentoring networks and how to structure them. For example, is it better to be more prescriptive—offering women’s networks and women’s events—or to pay attention to the kinds of things women are interested in and do that kind of programming? A combination of both may be needed, he said.

Perry and his colleagues also realized that the fact that most women engineers are in civil engineering, followed

FIGURE 4-1 Enrollment in Full-time (FT) and Part-time (PT) undergraduate Mechanical Engineering (ME) Programs, 2005–2012. SOURCE: ASEE Data Mining Site. SOURCE: www.asee.org

TABLE 4-1 The number of undergraduate women studying mechanical engineering (ME) is increasing at a higher rate than the number of men, 2005–2012 and 2007–2012.

SOURCE: ASEE Data Mining Site, www.asee.org.

closely by mechanical engineering, is important information for ASME. What draws women to mechanical engineering? They survived the front-end filter by getting a degree in the field; next is the other filter—entry into the profession—and then the issue becomes retention again.

When research results and surveys showed that the answer to questions about retention problems was “It’s climate, stupid,” Perry said he started wondering, Who controls climate? Who has the “thermostat”? Middle managers have control over what happens in the daily lives of people; do they determine the climate? Climate may change because more women are getting into management, but that will take time, Perry noted. It would be helpful for research surveys to track female engineers—and male—into middle management and ask, for example, who taught them to tweak the thermostat?

Catherine Didion opened the general discussion with some questions for the participants to consider: Looking at the climate issues, are there data that would enhance understanding of these questions? Are there issues that should be raised that haven’t been considered at the workshop?

Noting the limitations posed by the small sample sizes in the B&B survey, Jennifer Hunt from the Department of Labor wondered about appealing to the B&B surveyors to either oversample in science and engineering or increase the sample size, although she acknowledged that in these austere times that might not be likely. Perry suggested that ASME (and other engineering societies) could help with sample size, given the society’s 100,000 members. ASME could do a certain amount of database selection and then the survey instrument itself could filter further.

Joanne Cohoon commented that in her work with the Computing Research Association Committee on the Status of Women (CRAW), they have consistently run into the problem of not being able to learn what they need to learn, such as how many students—men and women, of different ethnic groups—switch from computing to another major. To address the lack of data, CRAW recently launched the creation of a massive dataset based on surveys of students at critical points in their education and on department-level data on outcomes; the surveys cover retention in and attrition from the major, entry into graduate school, and faculty status, among other outcomes. About 70 institutions are participating, and CRAW’s goal is to make the data publicly accessible while protecting confidentiality.

Engineering as a whole has the same sort of problem, Cohoon continued, and urged collaboration among those concerned about these issues to collect data and make it accessible. A single survey, for example, might present certain questions one year and different questions in other years, consistently collecting the same basic information for shared use among researchers.

CRAW sponsors undergraduate research experiences for students, and the data show that CRAW programs have a statistically significant effect on participants’ likelihood of going into a graduate program in computing. These data are very valuable for determining outcomes for different intervention programs, said Cohoon.

Tamiko Youngblood from Robert Morris University suggested that sample size concerns can also be addressed by engaging the National Society of Black Engineers and other societies whose member demographics are of particular interest to ensure the inclusion of populations that aren’t as well represented in the national datasets.

Victoria Rockwell, past president of ASME, wondered whether, in talking about datasets and how to retain women and minorities in engineering, other cultures have been benchmarked. She had recently learned from Kuwaiti women that women there make up over 50 percent of the engineering population, even though as undergraduates they cannot account for the same percentage in the classroom. Yet in a culture that is challenging for women, women constitute the majority in this profession and are very successful. This anecdote illustrates the need to benchmark datasets, conduct comparisons, and examine why women are better retained in some cultures, disciplines, and professions and why they drop out in others.

In response, Cohoon cited studies of engineering in 32 countries¹ and of computing,² and a 34-country study showing that interest in math-based fields or careers at the 8th grade level predicts women’s participation in fields such as

_________________

¹Charles, Maria, and Karen Bradley. 2002. “Equal but Separate: A Cross-national Study of Sex Segregation in Higher Education.” American Sociological review 67:573–599.

² Charles, Maria, and Karen Bradley. 2006. “A Matter of Degrees: Female Underrepresentation in Computer Science Programs Cross-nationally.” Pp. 183–203 in Women and Information Technology: research on the reasons for Underrepresentation, ed. Joanne McGrath Cohoon and Bill Aspray. Cambridge, MA: MIT Press.

engineering or computing in those countries.³ Rockwell too had heard that the odds of anyone going into science, technology, or engineering can be predicted based on how they perform in Algebra 1: if they do well, they’ll typically be able to perform in one of the technical professions. Cohoon cautioned that although studies show the need to be able to perform well in math, most women who do extremely well in math choose to major in the humanities.

Constance Thompson, who is responsible for diversity programs at the American Society of Civil Engineers (ASCE), said that she struggles with partnering the Engineering Workforce Commission (EWC) and the Bureau of Labor Statistics (BLS) data. Civil engineering comprises construction, environmental and water resources, and transportation. Does it look like women are well represented in civil engineering careers because a lot of women are going into environmental and water resources areas? There is a need to cross-reference the data and learn how engineering students progress from their undergraduate and postgraduate degrees into the workforce. She recommends that universities, professional societies, the EWC, and the BLS collaborate to provide a clearer picture of the career outcomes of those who major in engineering disciplines. She also works to promote data sources that are less well known, such as the National Association for Minority Engineering Program Advocates (NAMEPA) and the Women in Engineering Programs and Advocates Network (WEPAN). Each has a data collection resource; why aren’t they talking to one another, she asked, and why aren’t we talking about those data?

Thompson also noted the need to seriously consider the number of foreign students in the United States who are graduating with BS, MS, and PhD degrees in engineering. According to the EWC,⁴ in 2012 foreign students were

• the fourth largest population awarded bachelor’s degrees in engineering,
• the largest population awarded MS degrees in engineering, and
• the largest population awarded PhD degrees in engineering.

Peggy Layne from Virginia Tech pointed out that researchers tend to approach the study of women in engineering from their own disciplinary perspective—whether psychology, sociology, economics, or another field—and often do not talk to one another or read each other’s work, so they sometimes seem to be talking past one another as they attempt to understand the factors affecting women’s persistence and success in engineering careers. She encouraged researchers to take a broader perspective on how to engage various disciplines that study women in engineering. The National Academies or another organization might take a role in bringing together scholars from different disciplines to look at these issues.

Nimmi Kannankutty from NSF said that the Foundation has been collecting extensive data on scientists and engineers in the workforce for almost 40 years. In the early 1990s the agency substantially redesigned its surveys to collect information on people with science or engineering degrees and/or in science or engineering occupations, an important change that made it possible to follow people once they got their degree as they entered the labor market and progressed in their careers. One striking finding was an unexpectedly loose relationship between degree and occupation: 70 percent of people with a science or engineering degree don’t work in a science or engineering occupation. There is more of a connection in engineering, but it is still not as high as expected: even among those with a PhD in engineering, only about 60 percent work in occupations classified as engineering. Such data are collected through NSF’s series of three surveys called the Scientists and Engineers Statistical Data System, which surveys about 100,000 people every two years. Though these surveys are not entirely longitudinal, it is possible to look at the cross-sectional patterns that many researchers have discussed.

The NSF data differ in important ways from those of the Bureau of Labor Statistics, Kannankutty continued. BLS asks employers to provide information about their employees, whereas NSF surveys employees: What do you do? What’s your educational background? How do you spend your time? Are you satisfied with that? What kind of entrepreneurial activities do you engage in? How do you keep up your training? How do you stay relevant? Why did you change your job? Why did you leave the labor force? Moreover, once a decade NSF collects information on all college graduates—engineers, scientists, and those in nonscience fields—thus providing data that can provide broad national context.

Didion raised the question of critical transitions in the education and careers of women engineers. Are there certain transitions that seem to be of particular importance?

Hunt wondered why women seem to get onto a career path that leads to far fewer patents than men. Engineering is a major source of innovation, which is a driver of growth in the economy and in standard of living. So what engineers do is important, and one of the few tangible, measurable indicators of innovation is whether people patent. In her research Hunt has observed that men’s patenting life cycle starts in their 20s, peaks in their 40s, and then drops off somewhat; in contrast, women’s patenting life cycle remains at or near

_________________

³Nosek, B.A., Smyth, F.L., Sriram, N., Lindner, N.M., Devos, T., Ayala, A.,…and Greenwald, A.G. 2009. “National differences in gender–science stereotypes predict national sex differences in science and math achievement.” Proceedings of the National academy of Sciences 106(26):10593–10597.

⁴2012 Engineering & Technology Degrees. Engineering Workforce Commission of the American Association of Engineering Societies. Available at http://ewc-online.org/publications/degrees.asp.

zero.⁵ (She clarified, though, that a relatively small percentage of men patent, so it’s important not to exaggerate the difference.) Aside from the fact that women are largely missing from mechanical and electrical engineering, which are the big engineering fields that do a lot of patenting, even in these fields women tend not to be in jobs that involve design and development, the tasks most associated with patenting. And that’s true right from graduation. Why, right from the beginning, do women get onto a different path from men? Cohoon mentioned two reports from the National Center for Women in Information Technology on women’s representation among IT patents,⁶ which might shed some light on the issues raised by Hunt.

Echoing Perry’s point about whether women will help change workplace climate once they reach greater numbers in middle management, Nadya Fouad raised the issue of women in leadership positions. When she presented her research to a large defense industry company that has a technical path, the attendees expressed concerned that women were not seeking those leadership positions. They wondered why: What’s wrong with the women? Fouad reframed the question: What’s wrong with the environment that isn’t supporting women? In talking to the women, their perspective was, why would we want that opportunity? Patenting, she added, may be an opportunity for leadership.

Didion noted that the career pathways of many colleagues at the workshop may no longer exist for women who are getting their degrees now or did so in the recent past. Much has changed—for example, the assumption that a person may stay with one company for 30 years may no longer be true.

Rockwell questioned whether there has ever really been a career pathway. Each of us is responsible for our own career path, she said. When talking to women who are students or are in entry-level positions, she tells them to keep their eyes on their next position and to set their own career path. She urges them to think about where they want to be in one or two years and in five or ten years; that can change along the way, but they need to have that goal in mind.

Thompson described civil engineering as a highly entrepreneurial field—many women leave and start their own firms after 10 or 15 years. She also pointed out that engineering careers and the engineering curriculum are increasingly interdisciplinary, and that the discussion was not taking this into account. A person might start as a civil engineer and end up working in a field related to mechanical engineering. Current approaches to counting engineers and engineering careers may not capture these transitions.

Referring back to Hunt’s observation about patenting, a student said she is about to graduate with a degree in mechanical engineering and that it has never entered her mind to pursue a patent. Instead, she plans to go into a leadership position in engineering soon after she graduates. Didion related this observation to the role of faculty: “To what extent do we help students think about careers, and is there a gulf between what we see as seminal markers of career pathways and whether, for example, students would even think about patenting?”

Hunt said she would ask why not, but she also thinks people should do what they’re good at. She cited a friend at Pratt and Whitney who has five patents and believes that women go into management because that’s what they’re good at—they have better people skills compared to male engineers. If that’s what your strength is, you should take advantage of that. On the other hand, if the problem is lack of information or ideas or examples, and that’s why you haven’t considered a more technical path, then that’s a problem that warrants attention.

Cohoon observed that in computing, patenting varies tremendously by company. There is a company element that either encourages patenting as a valuable thing to do or doesn’t mention it, and either assists with it or doesn’t. That is one big variable in women’s patenting. It might never have occurred to you as a student, she said, but if you’re hired by a company that values patenting and values you, then you will be encouraged to patent and assisted with that process.

Why are we so concerned about long-term retention of women or men in strictly disciplinary occupations? asked Kannankutty. In engineering education, you learn quantitative skills and project management skills as well as how to conceptualize a problem and how to work in a team. On a macro level, if many people have this training and are out in the workforce, isn’t that a good thing?

Gretchen Fougere from Boston University agreed that the point is for people to be productive and help society. Engineering is a great training ground for getting many necessary skills, and a lot of both men and women who get engineering degrees go on to do different things because they are driven by different factors. Students, young or old, need to be educated in how the world works and how to take care of themselves, and that includes advocating for themselves and asking the right questions, such as, What do I need to do next? She wondered about the flip side of the research discussed: What did women do to try to navigate the barriers and hostility? What approaches did and did not work?

Romila Singh commented that an important reason to care about people staying in engineering is innovation. Engineering skills can be applied in all walks of life and they help a person to be productive and innovative. And innovation should include diverse perspectives.

_________________

⁵Hunt, J., Garant J-P., Herman, H., Munroe, D.J. 2012. “Why Don’t Women Patent?” NBER Working Paper 17888. Cambridge, MA: National Bureau of Economic Research. Available at www.nber.org/papers/w17888.

⁶Ashcraft, Catherine, and Anthony Breitzman. 2006. “Who Invents IT? An Analysis of Women’s Participation in Information Technology Patenting.” Available at www.ncwit.org/sites/default/files/legacy/pdf/PatentReport_wAppendix.pdf. Ashcraft, Catherine, and Anthony Breitzman. 2012. “Who Invents IT? An Analysis of Women’s Participation in Information Technology Patenting: 2012 Update.” Available at www.ncwit.org/sites/default/files/resources/2012whoinventsit_web_1.pdf.

Cohoon agreed that engineering skills are very valuable and can contribute to a productive and fulfilling career even if one does not remain in technical work. But when women are lost disproportionately as technical creators, they do not participate in creating the world we will all live in, she said. Engineering defines the built world, and the world should reflect the interests, perspectives, and needs of the world’s diverse society.