New US health survey figures show a sharp fall in lesbian, gay and bisexual identification among younger high-school girls, especially ninth-graders. The change has prompted plenty of political commentary, but researchers caution that the numbers don’t reveal a single cause. Here’s what the data show, and why the pattern deserves careful reading.

Essential Takeaways

  • The steepest change: LGB identification among ninth-grade girls fell from roughly 17% in earlier surveys to about 11% in 2025.
  • Girls drive the shift: Across high-school girls, identification dropped from around 27% to 22%.
  • Boys were comparatively steady: Boys’ figures moved up and down but returned to roughly 7% in 2025.
  • The reason remains unclear: Researchers say politics may play a part, but the age-specific pattern points to other influences too.
  • The survey has limits: Students who selected another identity, were unsure, or didn’t understand the question were excluded from the calculations discussed.

Ninth-grade girls show the sharpest fall

The most striking figure sits at the youngest end of the high-school sample. According to analysis of the Centers for Disease Control and Prevention’s Youth Risk Behavior Surveillance System, about 11% of ninth-grade girls identified as lesbian, gay or bisexual in 2025, down from close to 17% in both 2021 and 2023.

That’s a sizeable change over a relatively short period, and it stands out from the gentler movements in older year groups. The CDC describes the Youth Risk Behavior Surveillance System as a recurring national survey designed to track health-related behaviours and experiences among secondary-school students. Its results offer a useful snapshot, though they’re not a complete explanation of why young people answer as they do. ([2], [3])

The drop isn’t quite as dramatic among 10th-graders, while figures for 11th- and 12th-graders have broadly returned to earlier levels after a rise in 2023. In other words, the headline isn’t a universal retreat across every teenage age group. ([4], [5])

The overall movement is being driven mostly by girls

When the figures are separated by sex, the contrast becomes clearer. Around 27% of high-school girls identified as LGB in 2021 and 2023, compared with about 22% in 2025. Among boys, the comparable figures were much lower and ended close to where they began. ([6], [7])

Much of the female decline appears to come from fewer students choosing bisexual as their label. That matters because broad categories can conceal very different experiences: a fall in one identity group can move the overall number even when other groups remain fairly stable.

It’s also worth resisting the temptation to read the figures as a simple measure of changing attraction. Survey answers can be shaped by wording, privacy, confidence, social context and whether a young person feels comfortable selecting a particular label on that day. The CDC’s published materials provide the framework and results, but they don’t claim to identify the personal reasons behind each response. ([2], [4])

Why the pandemic-era rise matters

The latest numbers follow a noticeable rise in LGB identification among teenage girls after the pandemic period. Increased time online, disrupted routines and changing conversations around identity all formed part of the wider backdrop, but the survey itself can’t neatly assign responsibility to any one of them.

Older teenagers still appear more likely to identify as lesbian or bisexual than comparable teenagers did before the pandemic, even as younger groups have moved down. That mixed picture is important. It suggests the story is less about a single national switch being flipped and more about different age groups responding to a changing social environment in different ways. ([5], [6])

The survey’s design also deserves attention. Students who said they identified in another way, were unsure, or didn’t understand the question weren’t included in the particular calculations discussed by the analysis. That makes the percentages easier to compare, but it also means they shouldn’t be treated as a full map of teenage identity. ([2], [5])

Politics is part of the conversation, not the whole answer

Conservative commentators have presented the decline as evidence of a wider cultural shift, while LGBTQ advocates and researchers are likely to be wary of claims that turn young people’s identities into political scorecards. The numbers may be used in that argument, but they don’t settle it.

Psychology professor Jean Twenge, who analysed the CDC data, told The Washington Post that the pattern was difficult to explain and questioned why politics would affect ninth-graders so differently from older pupils. That’s a sensible caution. If the change were mainly caused by a national political mood, you’d expect a more even effect across the age range. ([6], [7])

The CDC’s surveillance programme is valuable precisely because it can show broad movements over time. But trend data raises questions; it doesn’t automatically answer them. Further surveys, research into question wording and closer attention to students who select other or uncertain responses may help clarify what’s happening. ([2], [3])

What readers should take from the figures

The safest conclusion is also the least dramatic: younger teenagers, particularly girls, are reporting LGB identities at lower rates than they did in the previous two survey cycles, while older teenagers and boys show a more mixed or stable pattern.

That doesn’t prove that fewer young people experience same-sex attraction, nor does it demonstrate that one administration, platform or cultural trend caused the change. It shows how students responded to a standardised survey at a particular moment. The distinction may sound small, but it’s the difference between reading evidence carefully and turning a percentage into a slogan. ([4], [5])

For families, teachers and youth organisations, the practical message is to keep conversations open and non-judgemental. Labels can be important, but they’re not compulsory, and uncertainty is a normal part of growing up. A calm question is likely to be more useful than a confident assumption.

The next rounds of CDC data should reveal whether this is a temporary dip, a lasting change or simply a more complicated pattern emerging across different ages. For now, the mystery remains the most honest part of the story. ([2], [6])

A striking trend deserves careful questions, not easy political answers.

Source Reference Map

Story idea inspired by: [1]

Sources by paragraph: