New CDC survey data suggest fewer US high-school pupils identified as lesbian, gay or bisexual in 2025 than in 2023, with the steepest change among younger girls. The figures are prompting fresh debate about how teenagers describe themselves, while researchers caution that the results reveal a trend, not a simple explanation.

Essential Takeaways

  • Overall decline: LGB identification fell across all four high-school year groups measured in 2025.
  • Youngest pupils changed most: Ninth graders recorded the lowest figure, at 11%, compared with 17% in 2023.
  • Girls drove the movement: Female identification fell from 27% in 2023 to 22% in 2025, while the male figure moved from 9% to 7%.
  • Bisexual identification mattered: The largest change was among ninth-grade girls, whose bisexual identification dropped from 26% in 2021 to 13% in 2025.
  • Interpret carefully: Self-described orientation and sexual behaviour aren't identical measures, and the survey can't establish why the pattern changed.

What the latest CDC teen sexuality data show

The headline finding is straightforward but striking: the proportion of US high-school students describing themselves as lesbian, gay or bisexual declined in every grade covered by the 2025 Youth Risk Behaviour Survey. According to the CDC's survey materials, the study tracks a broad range of health and behaviour indicators among pupils in Years 9 to 12 equivalent age groups. [2]

The biggest movement appeared among ninth graders, generally aged 14 or 15. Their LGB identification rate fell from 17% in 2023 to 11% in 2025, while seniors recorded the highest 2025 figure at 16%, still below their 2023 result. The numbers suggest a generational effect may be involved, rather than a uniform shift among pupils who had already answered the question in earlier years.

That distinction matters. A survey snapshot can show how a group describes itself at a particular moment, but it can't tell us whether individual teenagers have changed identity, whether newer cohorts use different labels, or whether some respondents are simply less inclined to choose an LGB category.

Girls account for most of the change

The gender gap is the clearest feature of the results. Female respondents identifying as LGB fell from 27% in 2023 to 22% in 2025, while the corresponding male figure decreased from 9% to 7%. Previous CDC reporting has also highlighted differences in health experiences and wellbeing among LGBTQ+ young people, so the figures shouldn't be read in isolation from the wider survey. [3]

Within the female group, ninth graders stand out. Their LGB identification reportedly fell from 30% in 2021 to 17% in 2025, while the change among senior girls was more modest, from 28% in 2023 to 25% in 2025. The contrast between younger and older pupils is doing much of the analytical heavy lifting here.

It also helps to avoid loaded shorthand. A lower identification rate doesn't mean teenagers are becoming "straight again", nor does it prove that a previous wave of labels was inaccurate. Young people may use different language over time, and some may prefer not to attach a label at all.

Bisexual labels are at the centre of the story

The reported fall was especially pronounced in bisexual identification among younger girls. Among ninth-grade girls, the figure dropped from 26% in 2021 to 13% in 2025, making bisexuality the largest single contributor to the overall change described in the analysis. [5]

That pattern fits a wider point about survey design: headline categories can conceal very different experiences underneath. "LGB" groups lesbian, gay and bisexual respondents together, but the three labels don't necessarily move in tandem, and a change in one group can substantially alter the combined total.

The CDC's Youth Risk Behaviour Survey is useful precisely because it offers repeated, standardised measurements across student populations. Even so, comparisons need care. Question wording, response rates, school participation and the social climate around disclosure can all influence what pupils report. [7]

Identity and behaviour aren't the same measure

The survey analysis also separates self-identification from reported same-sex sexual behaviour. Those measures overlap, but they aren't interchangeable: a teenager may have a same-sex partner without using an LGB label, or may identify with a label without reporting that behaviour in the survey period.

The reported figures suggest same-sex partnering remained more common among girls than boys, although the female measure had declined since 2021 while the male measure edged above 10%. That contrast is a useful reminder that identity, attraction and behaviour are related but distinct parts of adolescent development.

For readers trying to understand the data, this is the practical rule: don't turn one percentage into a claim about every teenager. The survey measures what pupils say about themselves under specific conditions, not a permanent verdict on identity or future relationships.

Why the numbers changed remains unclear

Researchers cited in coverage of the figures have warned that the data can't predict whether the decline will continue or establish a definitive cause. Possible influences might include changing language, peer culture, social pressures, the aftermath of the pandemic years or differences between newer student cohorts, but the survey alone can't rank those explanations. [5]

The CDC has also reported broader improvements in several areas of youth health, including trends that point to changing patterns of risk and wellbeing among high-school students. That wider context is important: sexuality figures sit within a survey designed to monitor health, behaviour and safety, not to explain cultural change on their own. [4]

For parents, teachers and policymakers, the sensible response is curiosity rather than celebration or alarm. The result is notable, particularly among younger girls, but the next rounds of data will show whether it's a lasting shift or a short-term dip.

A few percentage points can tell us a great deal about a population, but very little about any one teenager sitting in a classroom.

Choose language that leaves room for young people to describe themselves honestly and safely.

Source Reference Map

Story idea inspired by: [1]

Sources by paragraph: