LGBTQ Ph.D. graduates will soon be counted in key U.S. survey

The U.S. National Science Foundation (NSF) plans to add questions about sexual orientation and gender identity to its Survey of Earned Doctorates (SED) in the coming months, the agency announced last week. The move, which comes after years of pilot testing, pressure from scientists, and a scientific integrity complaint, is being applauded by activists and diversity experts who say the data will be critical for understanding how many LGBTQ people are in STEM. But NSF has yet to release specific details about its plan, including how it will safeguard the information.

The SED serves as an annual census of the nation’s Ph.D. recipients—roughly 55,000 respondents per year, across all academic disciplines—and is a key data source for researchers and policymakers who are interested in the diversity of the U.S. STEM pipeline. It asks questions about their gender, race, disability status, educational background, career plans, and more. But until now, the survey hasn’t included any questions that indicate whether the Ph.D. graduates identify as LGBTQ. In addition, the gender options were limited to just “male” or “female.”

Last week’s announcement is “exciting” and “very long overdue,” says Jon Freeman, an associate professor of psychology at Columbia University. Freeman has spent years drumming up support for NSF to collect data on LGBTQ scientists through its suite of workforce surveys—particularly on the SED, which he considers the “crown jewel.”

Last year, NSF added a question about gender identity to its National Survey of College Graduates (NSCG), a biennial survey of more than 160,000 U.S. bachelor’s degree holders. But data at all educational levels are needed, says Bryce Hughes, an associate professor of adult and higher education at Montana State University, “to understand where inequities need to be addressed and to make visible the participation of these communities.” Hughes, whose research has documented lower retention rates for gay men pursuing STEM bachelor’s degrees than for their heterosexual peers, hopes that as these kinds of questions get added to more surveys researchers will be able to study the retention of LGBTQ researchers through different stages of the STEM pipeline.

NSF conducted a pilot test last year to determine how best to phrase the questions on the SED. Of primary concern was whether the questions were clear, the respondents felt comfortable answering them, and their presence might decrease the overall survey response rate if respondents chose to leave the survey after seeing the questions. After collecting responses from more than 30,000 Ph.D. graduates who saw different iterations of the questions from May 2023 to March 2024, NSF found that less than 7% didn’t answer or selected “I prefer not to answer” for all test questions. Few respondents left the survey because they were asked about their sexual orientation and gender identity—approximately the same amount as left on other survey questions. Most respondents also reported feeling comfortable answering the test questions, although between 40% and 61% of LGBTQ respondents—depending on what question was asked—said they were concerned about how the data would be used.

In a statement last week that accompanied the release of the pilot data, NSF announced it would add new questions to the 2025 SED that will allow respondents to specify their sex at birth, their current gender identity, and their sexual orientation. Data collection for the 2025 SED will run from July 2024 to June 2025, and a summary of the data is expected to be released in October 2026.

It’s not clear how the new questions will be phrased or what the response options will be, and NSF declined to provide specific details in an email to Science Careers. The plans are subject to approval by the U.S. Office of Management and Budget, which oversees changes to federal surveys. NSF says it plans to submit those plans later this month.

Many are hopeful that, if approved, the new data collection effort will provide a window into LGBTQ representation in the academic research community. The pilot data released last week offered some starting estimates: Roughly 2% to 3% of respondents identified as a gender minority and roughly 13% to 15% identified as a sexual minority. Given that these numbers point to a sizable presence of LGBTQ people in the Ph.D. population, “I feel like that indicates really strongly the need for these kinds of questions,” says Zara Weinberg, a postdoc at the University of California San Francisco.

Weinberg, who co-authored a commentary in Cell this year listing actions scientists and institutions can take to better support transgender researchers, points out that studies have identified numerous workplace challenges faced by LGBTQ professionals in STEM, such as higher incidents of harassment. “But we don’t actually have any data about how that affects actual representation at different stages of career progression,” she says.

NSF has been exploring the idea of adding sexual orientation and gender identity questions to its workforce surveys since at least 2018, telling Science Careers at the time that it would require a “lengthy, deliberate process involving extensive experimentation.” The first pilot test came in 2021, when the agency asked 5000 respondents on the NSCG to answer questions about their sexual orientation and gender identity.

In 2023, NSF decided to officially add a question about gender identity to that year’s NSCG, allowing respondents to select whether they currently describe themselves as male, female, transgender, or a different term that they can write in. (The survey also includes a question about the respondents’ assigned sex at birth.) But it declined to add a question about sexual orientation, citing pilot results indicating that respondents took longer to complete the question, were more likely to change their answers, and that some exited the survey after seeing the question. In response, Freeman drafted a letter last year, cosigned by hundreds of researchers, protesting the decision and requesting that the pilot data be made public. Ramón Barthelemy, an assistant professor at the University of Utah and a signatory on the letter, said at the time, “We have fought so hard for so long to try to get representation in the scientific community, and what NSF is communicating to us is, they don’t want us to have that representation.”

A report describing the NSCG pilot data was posted online in January, but that didn’t clarify for Freeman and others why the sexual orientation question was left off the 2023 survey. “At least 16 other questions on the survey led respondents to exit at an equal or higher rate,” Freeman wrote in a letter published last month in Science.

Freeman filed a formal complaint with NSF’s Office of Scientific Integrity last month, as well as with the U.S. House of Representatives. Both confirmed to Science Careers that they’d received the complaint. The Democratic staff of the House Committee on Science, Space, and Technology added that it has held meetings with experts who corroborated the concerns raised by Freeman.

“I am concerned about the non-scientific biases—described in the complaint—that may have contributed to a delay in the implementation of a sexual orientation question. This complaint must be assessed expediently,” Representative Zoe Lofgren (D–CA), who serves as the ranking member of the House committee, communicated in an email to Science Careers. “We can only ensure our scientific enterprise is safe and welcoming for all if we collect data on the experiences of people who historically may not have been welcomed in our scientific enterprises.”

An NSF spokesperson wrote to Science Careers that the complaint is being handled according to the agency’s scientific integrity policy. “Given that it is an open case, there is no information that can be shared,” they added.

It’s unclear whether those events factored into NSF’s decision to add sexual orientation and gender identity questions to the 2025 SED. But Freeman says he was glad to see a detailed report of the pilot data accompanied NSF’s announcement this time around. It’s what he and others have been requesting for years, he says.

“They did a really comprehensive job,” agrees Clair Kronk, a lecturer at the Yale University School of Medicine who is an expert on sex and gender data collection standards in health care and biomedicine. She’s glad NSF is moving to add new questions to the full survey, saying she knows some nonbinary Ph.D. graduates who didn’t complete the SED in prior years because it lacked their gender as an option and didn’t allow respondents to skip the question or select “I prefer not to answer.” But she remains cautiously optimistic until NSF releases the final details of its plans. “I’ll believe it when I see it fully,” Kronk says.

One detail that isn’t clear is how NSF will handle individually identifiable sexual orientation and gender identity data. Currently, SED data are shared with respondents’ home institutions under data sharing agreements. Although the answers to the pilot test questions were not shared, it remains an open question whether that will change once the questions are added to the full survey, or whether the data will be withheld or shared “in aggregated form,” an option that was mentioned in the report released last week.

Given the wave of antitrans laws, it’s important to carefully consider who has access to the data, Kronk says. “There are a number of institutions that exist wherein I could see negative outcomes for people if this information were to get back to those institutions,” she says. “That’s something that NSF should definitely very much consider.”