LGBTQ+ people are everywhere. Trump is cutting the data that shows it

A new Human Rights Campaign report estimates that 12 percent of U.S. adults are LGBTQ+, a higher share than several widely cited estimates. But the federal government is removing the questions the HRC report and similar research relied on.

The consequences extend beyond knowing how many LGBTQ+ people there are.

Andrew Flores, a distinguished visiting scholar at UCLA School of Law’s Williams Institute, has used federal data to document that transgender people experience higher rates of crime victimization than cisgender people. Without questions that identify transgender respondents, researchers lose a way to measure that disparity.

Even if a future administration restores those questions, Flores told The Advocate that collecting enough new information for comparable research would take substantial time.

“If you really count the numbers, it’s nearly a decade of information that’s being lost, based upon the federal government’s action,” he said.

Related: LGBTQ+ Americans are going back in the closet under Donald Trump: report

Why estimates of the LGBTQ+ population differ

HRC’s study uses U.S. Census Bureau Household Pulse Survey data collected from 2021 through 2024. It exceeds Gallup’s 2025 finding that 9 percent of U.S. adults identify as LGBTQ+ and the Williams Institute’s 2023 estimate that 5.5 percent of U.S. adults identify as LGBTQ+, based on 2020-2021 data.

The HRC report estimates that about 2.3 percent of U.S. adults are transgender or gender-expansive. Gallup estimates 1.1 percent of adults identify as trans, while the Williams Institute’s 2025 estimate is 0.8 percent, based on 2021-2023 data.

Charleigh Flohr, a senior statistician at the HRC Foundation, called Census Bureau data the “platinum standard,” citing the agency’s ability to select households through random sampling. Flohr said answering questions online may even allow more respondents to feel comfortable disclosing their sexual orientation or gender identity.

HRC’s methodology nevertheless acknowledges limitations. Some groups may be more likely than others to participate online, and statistical adjustments cannot fully correct for those differences. Researchers also have to decide how to handle skipped questions or responses indicating uncertainty, which can also affect who is represented in the findings.

Both Flohr and Flores stressed that, despite the differing results, they don’t see their respective surveys as correct and the other’s as wrong. Flores said differences are expected because definitions and understandings of terms change and generations shift over time.

The Census Bureau classified the Household Pulse Survey as an experimental data series, allowing it to change questions and release findings more quickly than it can for the census or its traditional surveys. Flohr said changes to the questionnaire during the final period, when the survey sampled a new group of households each time, led HRC to exclude that period from its analysis.

The Williams Institute also uses government data, including the Behavioral Risk Factor Surveillance System for adult estimates and the Youth Risk Behavior Survey for youth estimates. States participating in the adult survey can choose whether to include questions about sexual orientation and gender identity. Flores said participation by states across the U.S. has allowed researchers to estimate the national population.

While recent estimates differ, historical estimates of the transgender population’s size in particular have varied even more.

Over decades, leading estimates have gone from suggesting a statistically representative stadium might only have one or two trans people in it to indicating you can reasonably expect to find a trans person on a full city bus.

Related: Trump administration scrubs LGBTQ+ questions from federal surveys

Trump’s data removals leave gaps in LGBTQ+ research

The HRC report comes as the Trump administration removes measures of LGBTQ+ identity from federal data collection, including one used in the report. On January 20, 2025, President Donald Trump signed Executive Order 14168, directing the federal government to recognize only two sexes and reject gender identity as a basis for federal policy.

Since then, agencies have frequently cited the order to stop collecting gender identity data. According to a Williams Institute analysis, approximately 360 federal data collections lost at least one sexual orientation or gender identity measure during the administration’s first year.

Myriad federal departments and agencies, as well as states, local governments, nonprofits, journalists, and researchers, widely use Census and other federal data. In HRC’s case, Flohr said population estimates and demographic breakdowns are vital to adjusting its survey results to better reflect the LGBTQ+ population.

Unlike the lengthy processes used to test questions before adding them, the removals happened rapidly. The Williams Institute’s February 2026 report found that 83 percent of removals occurred through “non-substantive change requests,” a process that did not require formal public notice or an opportunity for public comment.

The changes also affect the availability of previously collected information, according to Erin Meyer, senior data project manager at the Inter-university Consortium for Political and Social Research.

Meyer said the National Center for Education Statistics asked colleges and universities to report the number of students whose gender fell outside the male and female categories, but the released data didn’t include those figures. She said she and her colleagues hoped the information had been withheld from publication rather than deleted.

Flores said an employee told him they had been instructed to disregard information about respondents’ transgender identities, including when respondents volunteered it.

“When you ask questions that may sometimes be quite sensitive, and then you just mark it as residue, it’s an incredibly unethical thing to do,” he said.

Related: Anti-LGBTQ+ policies harm the health of not only LGBTQ+ people, but all Americans

Federal statistics face questions of political interference

Federal law sets standards for the integrity of government statistics. The Foundations for Evidence-Based Policymaking Act, signed by Trump in 2019 and commonly called the Evidence Act, established additional requirements for statistical agencies like the Census Bureau. Under the act, these agencies must protect respondents’ confidentiality and produce accurate, credible, and objective data.

Jonathan Freeman said those obligations should protect federal statistics from political interference. Freeman submitted Freedom of Information Act requests for records concerning a 2024 test of sexual orientation and gender identity questions for the American Community Survey, an ongoing survey that supplements the decennial census. He filed a case against the Department of Commerce and the Census Bureau in September 2025, seeking release of the records.

“These surveys are supposed to exist, and data are supposed to exist free from political interference, and so when you’re removing questions because they’re politically inconvenient, that is, you know, something to consider,” he said.

Freeman joined the lawsuit as a private citizen, but said his interest also ties to his work as a psychology professor.

NPR reported in August that the Department of Commerce, which oversees the Census Bureau, removed language banning political interference from its scientific integrity policy.

The Census Bureau did not respond to The Advocate’s requests for comment.

Related: LGBTQ+ non-U.S. citizens at risk under Trump’s immigration crackdown: study

Counting LGBTQ+ people depends on trust

Research links trust in government and other institutions to participation in some federal surveys. That raises questions about participation amid harsh anti-transgender policy changes and rhetoric. Flohr said federal surveys generally have strong confidentiality protections, but acknowledged the current administration’s pattern of upending long-standing practices.

“It really underscores the importance of having policies in place that protect surveytaker confidentiality, build that trust, and then that people we elect to office who administer those policies and shepherd that trust honor it,” she wrote in an email.

She hoped the current HRC report would make clear that LGBTQ+ people exist across society.

“They’re your coworkers, doctors, and teachers,” she said. “People should take note of that.”