Governments around the world are moving to restrict or ban social media for teenagers, often citing scientific evidence that less screen time leads to better mental health.
But a new review found a surprising gap in that evidence.
Researchers examined all 40 experiments that form the backbone of the case for youth social media bans. Not one included a participant younger than 16 years old.
In other words, the research most often used to justify restrictions on teenagers never actually tested teenagers.
The finding doesn’t prove the bans won’t work. It does mean policymakers are making decisions about young people without direct experimental evidence showing what those policies will do.
The studies skipped teens
Dr. Monika Neff Lind, a clinical psychologist at the University of California, Irvine (UCI), set out with two colleagues to find the research behind the bans.
She is also a parent and says she would have been thrilled to confirm that pulling kids off social media helps them. The data told a different story.
Her team gathered every experiment that tested whether cutting back on social media improves well-being.
These are the gold-standard studies, where researchers randomly assign people to either quit the apps or keep scrolling as usual and then compare the two groups. They ended up with 40 of them.
Not one of those 40 experiments included a single person under 16. Across every study, the average participant was an adult, and the youngest individual reported was 16 years old.
The teenagers these laws target have never actually been tested. Social media bans are being rolled out on a group nobody studied.
The results aren’t convincing
Sometimes strong results in adults justify extending a conclusion to teenagers. That leap does not hold up here, because the adult results are weak to begin with.
The benefits that politicians describe mostly do not show up in the data.
When the team combined results from all those experiments, the mental health payoff from quitting social media hovered near zero or came out small at best.
Roughly a fifth of the studies found that the restriction either did nothing or actually left people worse off, with effects such as increased loneliness and lower life satisfaction.
Mental health link is unclear
One large review by the National Academies reached a similar verdict. The link between social media and adolescent mental health, the report concluded, is small and far messier than the public conversation suggests.
These experiments had another flaw. Participants knew which group they were in and were surrounded by a culture that constantly tells people social media is bad for them.
Even so, the average benefit barely registered – and older participants tended to benefit slightly more, making any application of these findings to teenagers even harder to justify.
Beyond the thin evidence, Neff Lind’s team flags a second problem. The bans may not even keep kids off social media, and the ways they fail could make things worse.
Teenagers are famously good at finding ways around rules they did not agree to.
Three months into Australia’s ban, authorities reported that close to 70 percent of accounts belonging to under-16s were still active.
Many young people may create fake adult profiles or browse anonymously without logging in.
When that happens, they keep access to the platforms while losing the content filters and parental controls that come with a legitimate youth account.
Problems caused by the bans
Adolescence runs on a growing hunger for autonomy and respect.
Top-down edicts that ignore that tend to provoke defiance, not compliance. Age verification brings its own trouble.
The technology that estimates age from a selfie makes more errors with young faces and people of color.
Enforcement also forces families to hand sensitive identity documents to tech companies.
Banned teens can lose touch with schools and clubs, which rely heavily on these platforms to reach them.
An experiment without answers
The researchers are not arguing for an open internet where kids fend for themselves. Their point is narrower and sharper.
Governments are spending political capital and public trust on an intervention nobody has measured, and they are doing so with children.
“We do not know how social media bans will affect youth because we have never studied that question,” said Dr. Lind.
The honest position, the team argues, is to admit this is an experiment and treat it like one.
That means investing in real evaluation rather than waiting to see what sticks. The first thing to confirm is whether the bans change teen behavior at all, since every other question depends on it.
After that comes the harder work of tracking wellbeing through more than one lens.
Measuring more than feelings
Asking teenagers to rate their own mood on a survey will not cut it. The team wants researchers to draw from several sources at once, combining what kids report, what parents and teachers observe, and what phones can passively record about actual use.
They also warn against quick verdicts in either direction. A few viral stories of a ban working, or a few stories of it flopping, prove nothing on their own.
Comparing whole countries against each other is one of the weakest approaches available because no two nations differ by social media policy alone.
A cleaner option exists. Researchers could randomly assign one region to a later start date and compare outcomes with areas already operating under the ban.
Approaches like that can hint at cause and effect without a perfect laboratory setup, as long as the planning happens before the policy takes effect rather than afterward.
Next steps in testing teens
Until this paper, the conversation treated the science as settled. It is not. The specific finding is striking. The entire experimental case for youth social media bans rests on studies that excluded young people.
A meta-analysis reviewed all available evidence from those studies. The adult data showed little to no benefit, even under conditions designed to favor the intervention.
That changes what policymakers can honestly claim. A government can still decide a ban is worth trying, but it can no longer say strong evidence guarantees the outcome.
The label on this policy should read “untested,” and the researchers want clear benchmarks set in advance so the bans can be continued, paused, or revised based on real results.
For the millions of families now living under these rules, the stakes are concrete. If the bans quietly push kids toward less-moderated corners of the internet or erode trust between teens and parents, those costs fall on actual children.
Taking things away is the easy move, the team argues. The harder, better option is making the platforms safer in the first place.
The study is published in the journal Frontiers in Developmental Psychology.
—–
Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.
Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.
—–
