Every year, after clicking “submit” on the final copies of their Ph.D. dissertations, thousands of scientists answer all sorts of questions—for example, about their age, sex, race, ethnicity, and career plans—as part of the U.S. National Science Foundation’s (NSF’s) Survey of Earned Doctorates (SED). The questionnaire has served as an annual census of U.S. doctoral degree-grantees since 1957 and provides useful demographic information, which can be used to track the success of diversity efforts. In the years ahead, the survey may start covering even more ground: During a meeting last week at NSF’s headquarters in Alexandria, Virginia, the agency said it plans to test the feasibility of adding questions about sexual orientation and gender identity.
NSF’s move was catalyzed by a letter arguing that comprehensive, nationwide data on LGBT (lesbian, gay, bisexual, and transgender) scientists and engineers are needed because the group experiences disadvantages and disparities that are akin to other underrepresented groups, such as racial and ethnic minorities and women. Only a handful of studies have examined LGBT representation in STEM (science, technology, engineering, and math) graduate programs and the scientific workforce, so there’s a clear need for more information, says Jonathan Freeman, an associate professor of psychology at New York University in New York City and the lead author of the letter. The letter writers used LGBT—rather than, say, LGBTQ—because it’s the most generally recognized term, and they didn’t want to confuse audiences who may not be as familiar with others. “In wanting to have a conversation with folks about these issues, oftentimes it’s a way to meet them where they’re at in terms of language,” says letter co-author Laura Durso, the vice president of the LGBT Research and Communications Project at the Center for American Progress in Washington, D.C.
The letter writers asked NSF to include questions about sexual orientation and gender identity on the SED, as well as two biennial surveys administered by the agency: the Survey of Doctoral Recipients and the National Survey of College Graduates (NSCG), both of which are designed to examine the career trajectories of STEM degree holders living in the United States. The letter was cosigned by 251 scientists, engineers, and legal and public policy scholars, as well as 17 scientific organizations (including AAAS, which publishes Science Careers).
If LGBT data were available from these surveys, “you’d have tons of people chewing on these data” to figure out if and where underrepresentation exists and to suggest interventions, Freeman says. Data collected by NSF could help us understand a whole host of questions about STEM’s LGBT community—“whether they’re here, whether they’re being retained, what their work trajectory is, whether they get paid as much”—notes Lauren Esposito, an assistant curator at the California Academy of Sciences in San Francisco and cofounder of the 500 Queer Scientists visibility initiative.
NSF is still in the early planning stages, so details are sparse regarding how it will move forward. But a spokesperson told Science Careers via email that changes to demographic data collection require working with an interagency group and a “lengthy, deliberate process involving extensive experimentation” in order to ensure that the agency generates “accurate, reliable data sets.” NSF plans to start with the biennial NSCG. The earliest that LGBT questions would be added is 2021 because the window for testing questions to add to the 2019 survey has passed.
Durso, who has worked to add sexual orientation and gender identity questions to federal surveys across the U.S. government, understands why implementing changes to the survey will take time. “There’s actually quite a bit of testing that has to happen,” says Durso, who holds a Ph.D. in clinical psychology and has studied the LGBT community. “These are federal government surveys; you want to do a deliberate and well thought out process.” For instance, it’s important to get the wording right to ensure that people fully understand the question that they’re being asked. Before making any changes to an ongoing survey, statisticians also want to confirm that adding certain questions won’t cause some people to refuse to answer the survey entirely—for instance, because they are offended by the questions.
Esposito is also concerned that collecting this type of information could be risky for the survey respondents themselves. Esposito—who hadn’t read the letter that was sent to NSF until Science Careers emailed her a copy—agrees that there’s a need for the data. “We should be informed and have tools at hand by which we can make policy and bring about change,” she says. But she worries about these kinds of data being in the hands of the federal government. “Sexual orientation and gender are not protected classes federally and in many states in this country,” she notes. When “you can be fired for that information, it seems risky and it seems like a risk that many people would have to think twice before taking,” she says.
Told about Esposito’s concerns, letter co-author Adam Romero—the director of legal scholarship and federal policy at the University of California, Los Angeles, School of Law—acknowledged her concerns but expressed confidence that they were not a reason to scrap the survey questions. “In my experience, the federal government does a very good job to keep the personal demographic and other responses of survey takers highly confidential and protected,” he says. In addition, existing federal surveys that ask these kinds of questions usually give an option to decline to answer or to say that you don’t know. “For any particular person who may be uncomfortable, there’s no mandate to disclose their sexual orientation or gender identity.”
Policy decisions in higher education often hinge on information gleaned from federal surveys, notes Bryce Hughes, an assistant professor of education at Montana State University in Bozeman. So if NSF doesn’t collect data on sexual orientation and gender identity, then “we’ll miss opportunities” to make policy decisions that benefit LGBT communities, he says. “I’m just excited to see this moving forward.” Hughes wasn’t involved with the push to nudge NSF to add LGBT questions, but he understands the value of these kinds of data: Earlier this year, he published a study showing that sexual minorities are more likely to leave STEM undergraduate programs than their heterosexual peers.
Freeman—the author who spearheaded the letter—wants to get data into the hands of Hughes and other social scientists because he’s concerned that LGBT issues have been sidelined in STEM diversity discussions. “There is a tendency to see LGBT information … [as an] overly personal demographic detail … that should have no place in science and engineering,” Freeman says. But he says that people shouldn’t view it that way. “This is about a social identity that is like any other, like gender or race or ethnicity.” That’s why it’s important to have LGBT role models and adequate representation across STEM fields, he says.
“These are scientists and engineers, and so numbers speak and data speak, and I think having actual data on this would really change things,” Freeman says. “I think it would trigger a snowballing event of getting more people to study this issue and getting universities and federal funding agencies” to think about LGBT diversity initiatives, he says.