Researchers are examining a notable fall in the share of younger US teenagers identifying as lesbian, gay or bisexual, particularly among ninth-grade girls. The CDC’s latest Youth Risk Behaviour Survey offers an important snapshot, but it doesn’t explain the reasons behind the change, leaving social media, the pandemic, politics and wider cultural shifts in the frame.

Essential takeaways

  • The biggest change: Ninth-grade girls reporting a lesbian or bisexual identity fell from 30% in 2021 to 17% in 2025.
  • Bisexual identification dropped: The proportion of ninth-grade girls identifying as bisexual fell from 26% to 13% over the same period.
  • Age matters: The decline was sharper among younger pupils than among older teenagers.
  • The cause remains unclear: The CDC survey records responses, but it cannot establish why reported identities changed.
  • The survey is broader than sexuality: CDC research also tracks mental health, relationships, digital behaviour and other aspects of adolescent life.

Younger teenagers show the sharpest shift

The most striking figure concerns ninth-grade girls. In the 2025 Youth Risk Behaviour Survey, 17% identified as lesbian or bisexual, compared with 30% four years earlier. The fall was especially pronounced for bisexual identity, which moved from 26% to 13%, according to the CDC data and subsequent analysis. [2]

That doesn't automatically mean fewer young people experience same-sex attraction. Surveys capture how respondents describe themselves at a particular moment, and teenagers' answers can be shaped by privacy, language, confidence and the atmosphere around them. A form can record a change, but it can't explain the feeling behind it.

The contrast with older pupils is also revealing. Among twelfth-graders, LGB identification slipped from 18% in 2023 to 16% in 2025, while ninth-grade reporting fell from 17% to 11% across that period. The age pattern is one reason researchers are treating the results as a question to investigate, rather than a neat verdict. [5]

Why the numbers need careful reading

The CDC's Youth Risk Behaviour Survey is designed to track health-related behaviour and experiences among secondary-school pupils in the United States. Its subjects include sexual behaviour, emotional wellbeing, substance use, safety and online habits, making it a useful national barometer but not a laboratory experiment. [4]

That distinction matters. The survey can identify associations and shifts in responses, yet it cannot prove that a particular political event, online trend or cultural influence caused them. Researchers have pointed to several possibilities, including the disruption of Covid, changing social attitudes and the way teenagers encounter identity discussions online.

Digital life is an especially tempting explanation, since young people spend so much time navigating social platforms, group chats and algorithm-driven content. Still, a plausible theory isn't the same as demonstrated evidence. The CDC's wider work on digital behaviour is valuable context, but it doesn't settle the identity question. [3]

Politics is part of the debate, not the answer

Some conservative commentators have linked the decline to changes in government policy and education funding. Others have argued that social influences helped drive the earlier rise in LGBT identification and that the latest figures may represent a reversal.

The timing complicates that argument. The decline began before the most recent change in the White House, and the larger movement among ninth-graders doesn't neatly fit a simple political explanation. As analyst Jean Twenge has noted, if politics were the main driver, it's reasonable to ask why younger pupils appear more affected than their older classmates.

There is also a more basic caution here: survey answers are not a scoreboard. A fall in the number of pupils choosing a label shouldn't be treated as proof that a community has disappeared, just as a previous rise couldn't by itself prove a single cultural cause. Teenagers are still working out what words fit, and sometimes the honest answer is that none does.

Social media and adolescence add another layer

The wider CDC survey programme looks at online behaviour alongside health and relationships, reflecting how closely those parts of teenage life now overlap. A young person's social world can be immediate, public and surprisingly intense, with identity language changing quickly from one year group to the next. [2]

That doesn't make online culture a convenient villain. Social media can offer isolated teenagers support and vocabulary, while it can also amplify peer pressure, trends and polarised arguments. The same platform may be reassuring for one pupil and exhausting for another.

For parents and teachers, the practical response is less about guessing a cause and more about creating room for honest conversation. Ask open questions, avoid treating a survey category as a diagnosis and remember that a teenager's answer may change without being insincere. Quiet, patient listening is usually more useful than a dramatic theory.

Researchers are watching what happens next

The latest results sit alongside other research suggesting that some forms of trans and queer identification among young adults may also be levelling off. But different surveys use different questions, age groups and methods, so the findings shouldn't be stitched together as though they were one continuous dataset.

Future waves of the Youth Risk Behaviour Survey should help show whether the 2025 figures mark a lasting generational change or a temporary wobble. Researchers will also need to examine the effects of the pandemic, social media, politics and changing terminology without assuming any one factor explains everything.

For now, the most responsible conclusion is also the least sensational. The reported decline is real within this survey, the strongest change appears among younger girls, and the reason remains unsettled. Good reporting should hold all three points together.

It's a revealing trend, but the explanation will take more than one survey to uncover.

Source Reference Map

Story idea inspired by: [1]

Sources by paragraph: