STEM is a short form for the academic disciplines of Science, Technology, Engineering and Mathematics. National Science Foundation of the United States initiated an education policy and curriculum to improve competitiveness in the development of afore mentioned academic fields. Additionally, STEM has implications for workforce development, national security concerns and immigration policy.
In recent years, STEM has got great attention online and STEM guidelines have become a hot topic in social media, as policy makers are re-looking at the ideology and design of the STEM structure. However, the big question remains, where do doctoral students go and what do they do after their PhD. Until now, it is unclear, unknown and unasked. Below are the details from five interesting Science Careers blogs, suggesting how STEM strategy should be revisited.
In a recent 72-page report from the Council of Graduate Schools (CGS), a clear picture on how underrepresented minority (URM) students get affected by STEM doctoral programs was revealed. This report titled Doctoral Initiative on Minority Attrition and Completion (DIMAC) documented with interesting and rather shocking facts about how black/African-American students and Hispanic/Latino students fare in doctoral programs STEM (including social and behavioral sciences).
The project examined enrollment data of 7575 black/African-American and Hispanic/Latino students who entered the STEM graduate programs at 21 universities between the academic years 1992-93 and 2011-12. A total of 1640 of those students answered surveys about their experiences, advisers, and various features of their programs, and 322 participated in focus groups discussing those topics. The researchers also collected information on the universities’ programs and policies relating to URM students and conducted focus groups discussing these topics with about 320 university officials. For the 3829 URM students identified as entering their doctoral programs before 2005, 44% had completed their degree by the 7-year mark, and 20% were continuing their studies. The remaining 36% of the original enrollees had withdrawn from graduate school.
Life science had the highest rate of completion at the 7-year mark (52%), followed by engineering (48%), physical and mathematical science (39%), and social and behavioral science (38%). Attrition was highest in the physical and mathematical sciences (47%), followed by engineering (36%), social and behavioral sciences (33%), and life sciences (31%).
By the 10-year mark, 54% of the students who entered between 1992 and 2002 had received their doctorates-including 63% of those in life sciences, 56% of those in engineering, 52% of students in the behavioral and social sciences, and 45% of those in the physical sciences. Although no rigorous comparative data exist, the top-line figure-54% appears “pretty close” to the 55% 10-year completion found in an earlier CGS study of the overall graduate student population, DIMAC co-author Hironao Okahana, a research associate at CGS, observed during a webinar for reporters prior to the report’s release.
Women attained higher completion rates and lower attrition rates than men did, with 45% of women earning their degrees, compared to 42% of men. The data also showed that 33% of women dropped out by the 7-year mark, compared to 40% of men. At 7 years, Hispanic/Latino students had higher completion and lower attrition rates than black/African-American students: The authors found that 48% of Hispanic/Latino students had earned their degrees and 35% had left, compared to a 40% completion rate and a 38% attrition rate for black/African-American students. “Racial/ethnic differences persist after controlling for the main and interactive effects of gender and field in seven-year completion and attrition,” the authors write.
It would be useful to know why so many minority students left-but because they only surveyed graduates, that question isn’t answered. Other literature, though, addresses this topic. Much of this writing carries the implication that graduate student attrition represents failure, either of the student to master the work or of the institution to enable success. As Okahana observed in the webinar, though, some fields, especially computer technology, offer attractive employment opportunities that may draw students away from graduate school.
A wide range of other motives may cause students to withdraw once they have experienced academic life. For example, some may find scientific research a poor fit with their personal values. Economic realities may also come into play for some: The prospect of long, poorly paid grad school and postdoc years followed by uncertain career prospects in a glutted job market may persuade some to seek more lucrative and reliable alternatives. Research shows that the desire to earn a good income and improve their financial status motivates first-generation college students more strongly than it does students with college-educated parents. A substantial portion of those who answered the DIMAC survey appear to fit the profile of first-generation students with limited financial means; 38% have parents who lack college degrees and 44% received Pell grants for undergraduate study, indicating modest family incomes.
Some URM students may find the culture of graduate school uncongenial, especially if they attend institutions with small URM populations either on campus or in the surrounding community. Although very few of those surveyed mentioned overt racism or racial discrimination-certainly good news-a number cited the importance of “fit” with the program, their advisers, their professors, and their peers. Comfort and cultural familiarity may help explain the outstanding ability of historically black colleges and universities (HBCUs) to produce science students who pursue graduate degrees. DIMAC allows no sweeping conclusions except that for URM students as for all students, the experience of graduate school is influenced by a host of personal and institutional factors. In addition to its detailed characterizations of URM populations, it offers several recommendations for institutions to improve the experience of URM graduate students.
Complete PDF file of the report can be accessed here.
One of the great mysteries of the scientific world is what happens to PhD recipients after they finish their degrees. Only a small percentage get the tenure-track faculty positions they ostensibly spent years training for. The rest move on to other careers, obviously, but little is currently known about their exact destinations. Now Melanie Sinche of the Labor and Worklife Program at Harvard Law School is trying to learn more. “If you earned a PhD in any of the physical, life, computational, engineering, or social sciences between 2004 and 2014 and have ever worked, trained, or studied in the U.S.,” you can help by participating in a confidential online survey estimated to take about 15 minutes.
Sinche wants to know “where recent science PhDs are currently employed” so that she can “create a visual map of career clusters,” according to the survey website. The study also aims to “identify the skills and experiences required to enter different fields, and determine whether these skills were developed in the educational/training period of the PhD or on the job, thereby informing the design of graduate and postdoctoral training programs.” One hopes.
If helping to answer questions central to the welfare of early-career scientists is not reward enough for participating, you can enter a drawing for one of five $100 Amazon gift cards. Beyond that, you may ultimately enjoy the benefit of “enhanced understanding of your career options through the publication of a career map for PhDs in science.”
In keeping with human-subjects protection requirements, the survey consent form warns of the risk of “mild discomfort with some of the questions, depending on how you feel about your employment situation.” A clear picture of the available career paths, though, could help reduce the discomfort, mild or otherwise, that numerous PhDs feel when contemplating their employment. You can find the survey here.
Why are African Americans, Latinos, Native Americans, and women underrepresented in academic science? A large literature answers this question by citing obstacles such as bias, lack of role models, and shortcomings in the academic preparation many members of those groups receive. Much of this work rests on an unstated assumption that a research-oriented faculty career is so desirable that qualified scientists will naturally prefer it to other possibilities. If that were true, attracting more people from underrepresented backgrounds would merely require repairs to the “pipeline” that is supposed to deliver young people into those careers.
Kenneth Gibbs Jr., a Cancer Prevention Fellow at the National Cancer Institute in Bethesda, Maryland, and one of the study’s authors, points to work showing that scientists from underrepresented minority (URM) groups and women seek research-focused academic careers less often than males from well-represented groups, even when they have “the same level of research productivity and … the same mentoring. If you control for everything and you still see differences, the only thing at fault is the system itself,” Gibbs says in an interview with Science Careers.
Something “fundamental is probably influencing those choices,” Gibbs believes—most likely the “tension” many URM and female scientists say they feel between the culture and expectations that govern academic science and the values they bring to their work. Gibbs and his co-authors have examined the career motivations of minority scientists using focus groups, a survey, and interviews. Two published articles present part of this work; a third is forthcoming.
This and other work shows that women and URM scientists on average “choose differently.” Their choices are made “outside of ability, outside of competence”—but in keeping with expressed desires to pursue social justice, community involvement, and altruism, he says. In contrast, men from well-represented groups more often seek academic research careers that incorporate the value of “scientific freedom, the ability to research what you want on your own terms.”
For scientists with strong social concerns, scientific and social motivations are “intertwined,” Gibbs says. He “chose science because I perceived that it would allow me to express the values that I already had: helping the world, using my work to help humanity, particularly the parts of humanity that have fewer resources than I have.” For those who share such motivations, both the scientific and the social are “necessary,” but neither is sufficient. It is therefore regrettable that “the system currently says that science—the research—is the essence, the end to which you work as opposed to the means to an end.”
A feeling of obligation
“Scientific research tends to focus on individual enhancement,” Gibbs explains. But many people—“disproportionately, but not exclusively, scientists from populations that are underrepresented within that workforce”—bring to their scientific training an “ethos that I am obligated beyond my own personal advancement. Every person comes from some sort of community, and that community has some sort of narrative. You develop your personal and community-rooted identity prior to your vocational identity.”
You take all that with you when you enter a career, Gibbs says. “You don’t leave your social background because you have a vocational background in science. All these things are always present. I am always a scientist. I am always a black man descended from slaves. I am always married to my wife. These things are always true all at the same time, and I think that sometimes we’re not thoughtful enough about how these things are in play.”
One scientist Gibbs knows—“from an underrepresented group who produced at a very, very high level”—turned down a faculty post at a top research university and “ended up teaching at an undergraduate-serving institution. He said that at the elite institution, his job would be to build as big an empire as he could for himself. That was not what he was motivated to do. He wanted to think about how he could influence more students. He said, ‘How can I do the most good for my people? Can I do the best good by going to [the renowned university] and becoming very famous, or could I do that by doing the hard work of training students to move forward?’” A high percentage of black scientists on university faculties, Gibbs notes, work at master’s-degree-granting institutions, especially historically black colleges and universities (HBCUs). (Two thirds of the nation’s HBCUs fall into that category, while a third offer doctorates.) Such schools have an enviable record as producers of black students who go on to obtain science doctorates.
Other recent publications reveal similar preferences. Dustin B. Thoman of California State University, Long Beach, and co-authors found “altruistic motives uniquely influential to URM students,” often playing “an important role in influencing their interest in scientific research careers.” URM trainees from undergraduates to postdocs express a desire for “STEM training to include or make room for a social justice component,” report Andrew Campbell of Brown University and co-authors. The trainees studied want “opportunities to do science with a purposeful social justice component, a desire that does not preclude performing traditional bona fide research at the highest level. This desire appears to reflect the senses of disconnect and marginalization that trainees feel within the academy and the scientific community. It also appears to align with their concerns for issues such as health disparities, which are evident for underrepresented/disadvantaged groups.” Other studies we’ve reported on also found an important role of personal values in scientists’ career choices.
Gibbs emphasizes, however, that the various studies only reveal general tendencies within groups, while individual opinions cover a wide range. “Any number of underrepresented minorities or women only want to be basic scientists, and that’s fine,” he says. And numerous men from groups well-represented in academic science harbor strong social motives. Overall, though, this values clash causes academe to “lose way more of the underrepresented folks, and there are smaller numbers of them to begin with,” he notes.
Making room for social impact
Academic science could attract and keep more URM scientists, Gibbs argues, by making room for their need to have broader impacts on society than basic research positions currently allow. “Those are values that I brought into science and that you hear many bringing into science,” he says. The tension between science values and social justice values grows as people advance in their training; they find that “[social] values are no longer able to be expressed. Not only are they not rewarded, but expressing those values makes you look less serious about the scientific work.”
For socially motivated scientists, “there are lots of ideas of what it means to serve in the community.” One could, for example, “spend 3 hours a week in an under-resourced school exposing children to what science is.” At present, such an activity “would not be valued” by academe as much as spending those same hours in the lab. Changing this “would take some thought about what our work structures are and how we evaluate people.” He believes change is possible though: “We do lots of really hard things in science. We’ve put men on the moon. We’ve done in vitro fertilization. We’ve pushed lots of boundaries.”
“What if someone were hired to do research and also spend 20% of their time extending that research to the community through K-12 education or working with low-income people [as part of] a tenure-track job at a major research university? What if that were well-supported throughout the system as a viable path?” Such a plan, the evidence suggests, could make academic careers attractive to scientists from a wider range of backgrounds, with a wider range of beliefs.
No single factor can explain the skewed demography of academic science, and no simple solution can fix it. But Gibbs’s critique has a great deal of explanatory value and could prove to have important implications for setting science policy.
“Among the top twenty-one college producers of future blacks with science doctorates, seventeen were HBCUs and none were Ivies.” —Richard H. Sander and Stuart Taylor Jr.
For more than a century, resources had been distributed unequally among the races, so the college’s facilities, its academic standards, and the academic preparation of its students could not match those of selective white colleges. Prestigious white institutions were just beginning to recruit talented blacks.
The data that is marshalled persuasively demonstrate that the larger the mismatch between the academic credentials of the mismatched students and the rest of their class, the graver is the danger that they will receive poor grades, lose confidence and self-esteem, drop hard courses, leave college without a degree, and learn far less than they would have if placed among more closely matched peers. Devoutly wishing to enhance minority students’ access to academic and career success, and disdaining universities’ self-serving desire to assemble racially diverse student bodies at the expense of young people already shortchanged by inferior K–12 schools; large minority admissions preferences are hypocritical and a severe disservice to many able students.
Giving preferences to African Americans, Hispanics, American Indians, and some other minorities (but not Asians) has become widespread and entrenched. “By 1980, more than three-quarters of the black students, and a majority of the Hispanic students at selective colleges and professional schools were there … because they had received a preference,” authors Sander and Taylor write.
“The vast majority of students who are admitted with large racial preferences are talented people who are well equipped to succeed in higher education,” they continue (italics in original). At institutions from community colleges to world-class universities, able and motivated students of all races can and do thrive academically when they enter with credentials that generally match those of the admitted class.
A policy of racial preference, however, sets up many unsuspecting students for failure and disappointment, depriving them of degrees and careers they would have attained had they attended colleges that suited them better, and depriving the nation of more STEM-trained minority professionals.
The effect of mismatch turns out to especially damage minority students’ chances of earning STEM degrees. “As seniors in high school, blacks were somewhat more likely than whites to report an interest in majoring in” STEM—45% to 41% respectively, the book states, citing research on students at Ivy League colleges by psychologists Rogers Elliott and A. C. Strenta of Dartmouth College. These once-aspiring minority scientists, however, “were only slightly more than half as likely as whites to finish college with a STEM degree.”
Why such high attrition? Wherever a college stands in the academic pecking order, STEM courses are always among its most demanding. “Relative academic weakness, not absolute weakness” explains why STEM students who enter college “with comparatively low credentials and an interest in the sciences tend to stream for the exits—and into less challenging courses—after their freshman year,” Sander and Taylor continue. Furthermore, STEM curricula are sequential, so disadvantage accumulates. “A student who performs only passably in the first course of a sequence will be at a still-bigger disadvantage in the second and third courses,” they write. “A student who starts as a chemistry major and whose preparation puts her roughly in the middle of her class will probably do fine and will gain a greater sense of mastery and confidence with each passing semester. A student whose preparation puts her near the bottom of the class can easily feel progressively more lost, and the poor grades that take her to the bottom of a STEM curve add insult to injury,” they write.
On the other hand, “historically black colleges and universities (HBCUs) such as Howard, Fisk and Clark Atlanta” enroll students who are “on average significantly weaker academically than the black students at Dartmouth and the other Ivies,” the book states. Yet, because their students do not suffer mismatch, “these schools were producing large numbers of STEM graduates.” Many of these alumni go on to success in graduate school, though probably not mainly at the most prestigious ones. “[A]mong the top twenty-one college producers of future blacks with science doctorates, seventeen were HBCUs and none were Ivies.”
Across the academic spectrum, “over half of the STEM degrees went to students whose [SAT math] scores put them in the top third of their class; those in the bottom third earned about one-sixth of the degrees,” Sander and Taylor continue. They write that research by psychologists Frederick Smyth and John McArdle, who were both then at the University of Virginia in Charlottesville, shows that, stunningly, “had all the black and Hispanic students in their sample enrolled at schools where their credentials were close to the class-wide averages, 45 percent more of the women minorities and 35 percent more of the men minorities would have completed STEM degrees.”
Colleges make the claim that giving racial admissions preferences enhances minority students’ opportunities for academic and career success. These institutions perversely condemn to the bottom of their classes students who could succeed—many of them in STEM fields—in more favorable environments.
The argument Sanders and Taylor make is unpopular among academic administrators, and, they illustrate, it has been systematically suppressed. But the evidence that they present makes obvious that the solution to educational inequity is not to be found in continuing to mask it with racial admissions preferences that harm students.
PhD graduates can take part in a survey to help create a visual map of career clusters.
Melanie Sinche is a nationally certified career counselor focused on STEM careers, currently serving as a Senior Research Associate at the Labor and Worklife Program in Harvard Law School, studying employment patterns of science PhDs. She formerly served as Director of the FAS Office of Postdoctoral Affairs at Harvard University. She is an accomplished career counselor, trainer, and speaker. In addition to building three career centres for graduate students and postdoctoral scholars, she has delivered career development presentations and training sessions for universities, government agencies, professional associations and non-profit organizations across the country on career-related topics for graduate students and postdocs. Her current focus is to improve data collection on PhDs and postdoctoral scholars across the U.S. She is also working on a book-length project on careers for PhDs in science with Harvard University Press, scheduled to be on the market in the fall of 2016. In this interview, Julie Gould asks Sinche about how she got involved and interested in this field, her new book and how PhD graduates can help with her research.
How did you get involved in the STEM careers space?
I don’t actually have a STEM background – other than my dad having a PhD in physics and being involved in scientific organizations over the years, such as the Society for Native Americans, Chicanos in Science and the Biomedical Sciences Careers Program in Boston. I was actually in graduate school decades ago for Russian and Eastern European Studies at the University of Michigan (UoM). But while I was there, I volunteered at the career centre to be a peer counsellor and work with other grad students, reviewing CVs and advising. I absolutely loved it.
When I was close to finishing my studies and was thinking about my career – whether or not to do a PhD – a job opened up in the career centre at Michigan and I took it. I’ve been in this field ever since.
What was it about working in a careers centre that you enjoyed so much?
I have always found it rewarding to help others find satisfying work. It has been such a gift to work closely with so many PhDs over the years, and to have an opportunity to be exposed to their work. Every day is different, and everyday brings new challenges, which keeps career counselling and advising interesting to me.
What are you doing now?
Following my time at the NIH, a job opened up at Harvard University to serve as Director for the Office of Postdoctoral Affairs. I spent 3 ½ years there, building programmes and serving postdocs, and then moved into a research position with the Labor and Worklife Program at Harvard Law School, headed by economist Richard Freeman. I have been working with Richard and this programme for a year now, conducting research on employment outcomes of PhDs in science.
What are you working on at the moment?
I know there are so many scientists out there who are still completely isolated, with no access to career services or advisors. I wanted to create a handbook for those who really need help, and so I’m writing a book. The idea is that the book will serve as a guide for any scientist anywhere who wants to change jobs or find a job. My goal is to take PhD scientists through the career development process: self-assessment, career exploration, goal setting, and the job search. This book will serve as a manual for finding a satisfying job in any field of interest.
As part of your book research, you’ve recently launched a survey to find out where PhDs in science are going after they graduate. How did this get started?
The National Science Foundation (NSF) collects comprehensive data annually on this population, but existing NSF data sets still contain gaps in the data. Their most recent, large-scale survey effort, the Early Career Doctorates Survey, does not address some of the questions I wanted to explore, so I decided to draft my own survey and collect data on my own. I am grateful for the previous work of Geoff Davis, Michael Roach, Henry Sauermann, Kenny Gibbs, Michael Teitelbaum, and Paula Stephan, who have all assisted me in drafting this new instrument.
What information are you trying to capture with your survey?
The survey completes the piece that is missing from our understanding of the PhD employment landscape: where are PhDs currently employed? What are the career choices that future PhDs have? We know that current students and postdocs are aware of faculty jobs, and most PhDs in science recognize that there are also opportunities in private industry, but if you dig a bit deeper, you come to realize that many have a limited understanding of what people actually do. I wanted to illuminate this kind of information on a granular level: where are PhD graduates working? What are their job titles? What sector are they in? What activities do they regularly engage in? I will use this data to create a visual representation of where people are working—a careers map for PhDs in science.
The second research question that I’m asking (and that the NSF doesn’t cover): What skills do scientists develop organically during their PhD programs, and are these skills required for the jobs they eventually move into?
My goal is to demonstrate that there are significant skills that PhDs build organically during their studies that are attractive to employers across different sectors. To prove this, I knew I needed hard data. If we have this kind of information, it might inform how we train graduate students, PhDs and postdocs in the US.
How are you exploring what skills employers are looking for in PhD graduates?
I developed the list of skills used in my survey from data collected every year by the National Association of Colleges and Employers (NACE). NACE results are collected from HR recruiters/employers from the private sector. Every year, they publish a set of outcomes called Job Outlook. They survey all of the employers and ask: What skills are you seeking from your students? However, their focus is on the undergraduate population hired by employers, not on the graduate student and postdoc populations. Still, as you read the list of skills desired by employers on the most recent NACE survey, it is clearly aligned with what goes on in graduate school: the ability to analyse large amounts of data, critical thinking, problem solving, independent thinking, team work. I want to show PhDs everywhere that they do in fact have skills that employer’s value!
Are these skills enough though?
The PhDs that I work with come in to see me and say: “I have no skills.” “I’ve been studying this one molecule or protein and I have nothing to offer.” “I’ve got nothing of interest to an employer.” And as much as I say “yes you do,” it’s meaningless. I need data to back it up. But the conversation then follows on to: you cannot expect to find a job with these skills alone. You need to engage in some sort of transitional experience. Trainees need to spend time in some kind of transitional experience. Either as a volunteer or intern, shadowing an employer or taking extra classes in finance or big data, for example. Students need to demonstrate experience to be more attractive for the jobs they are trying to move into. So in addition to my hypothesis that PhDs do emerge from their programs with important skills, I also want to present data on the kinds of transitional experiences that people who were hired used to get into the job in the first place.
How can PhD graduates help you?
They can take part in my survey! The survey project is called “Identifying PhD Career Pathways in Science” and is open to all PhDs earned between 2004 and 2014 in any discipline within the physical, life, engineering, computational, or social sciences. The survey was closed on 28 April 2015, and results will be submitted to a peer-reviewed journal.