Did Phones and Social Media Really Make This Generation Anxious? (Plain Version)

This is the condensed edition of the deep dive of the same name. Every key number has been verified against primary sources; for the full argument and citations, read the deep dive. Figures as of July 2026. This is not medical or parenting advice.

First, split one question into three

The Anxious Generation says: childhood moved onto phones, and that is the single largest reason for the wave of adolescent mental illness. The scientists criticizing the book say: that claim is not supported by science. Heavyweights on both sides, seven years of fighting, no resolution.

One reason nobody wins is that this is really three separate questions: Are the kids actually doing worse? Did phones cause it? Do phone bans work? The answers are not a package deal — this article takes them one at a time.

The other reason is yardsticks. One side's "screen time," the other side's "social media," and the chart's "smartphone adoption" are not the same thing; "telling a survey you feel sad" and "being hospitalized for self-harm" are even less so. We fact-checked 24 groups of load-bearing claims for this piece: 23 needed corrections, and the corrections split roughly evenly across the two camps. The biggest problem in this debate isn't that anyone is lying — it's that both sides' signature numbers say more than the data does.

Question one: are the kids really doing worse? — Yes, with boundaries

The deterioration is real. Three independent anonymous US surveys (no doctors, no names) agree: the share of high-school girls "persistently sad or hopeless" went from 36% in 2011 to 57% in 2021. One survey has asked the exact same four questions for 28 years — girls' depressive symptoms were actually falling from 1991 to 2011, then turned upward in 2012. The hardest indicator — suicide — is real too: the rate for 10-14 year-olds tripled from 2007 to 2018. Norwegian researchers formally tested whether kids just changed how they answer questionnaires. Answer: no. The worsening is real.

But the "explosion in kids seeking care" contains a lot of water. In October 2016 the US changed one medical-coding rule (doctors were told to also record "suicidal thoughts" as a diagnosis) — and that category of visit jumped 43% nationwide in a single year. In the scary numbers that merge "thoughts, self-harm, and attempts," most of the rise comes from how thoughts get recorded, not from behavior. The water and the real deterioration coexist — real self-harm ER visits rose (+77%) even slightly faster than suicide deaths (+66%).

The boundaries: the crisis is solidly documented in English-speaking countries, mostly in girls' self-reports and in US hard outcomes. German kids didn't deteriorate in sync (at least their parents didn't see it); "globally synchronized" is oversold. And US indicators have been improving or flat since 2021 — while kids' screen use has not dropped. That's a genuine problem for "phones are the main cause."

Question two: did phones cause it? — A real effect, yes; "the largest reason," no evidence

Start with the numbers both sides fight with. The critics' most famous line — "social media's link to well-being is in the same league as eating potatoes" — used the widest possible yardstick ("all digital technology"); narrow it to girls and social media and the link genuinely grows. But the proponents' most famous number ("for girls, a correlation of 0.15-0.22") traces to nowhere — we chased it to the end and no paper has ever actually produced that range. Both camps' signature numbers took corrections in our audit.

What's actually worth money are the experiments and natural experiments — and they agree with each other to a startling degree. Pay people to quit Facebook for four weeks: well-being improves by 0.09 standard deviations. Have twenty thousand people quit for six weeks: 0.06. Watch Facebook roll out college by college in 2004-06: student mental health worsens by 0.085 where it arrives. Watch broadband reach Italian towns: psychiatric hospitalizations of the young rise by 0.08. Two things are true of all these numbers: the direction is harm, and the size is small — roughly a tenth to a third of the total crisis, and mostly measured on adults.

So the honest conclusion takes three sentences:

• "There is no causal evidence at all" is wrong — quitting experiments, natural experiments, and tracking studies all point the same way: more use first, worse mood after, not the reverse.
• "Phones are the single largest reason" also has no evidence — everything proven so far, added together, falls far short of the size of this crisis, and no study has closed that gap.
• The likeliest truth is in between: a small, real harm, multiplied by nearly universal exposure, stacked on a crisis with many causes. "Small" doesn't mean "ignore it" — but it does mean you shouldn't expect the crisis to retreat just because phones get banned.

Question three: do phone bans work? — For mental health, so far: zero

This may be the most counterintuitive part. The policy camp's favorite study — Norway, built on national health registries, now published in a top journal — opens its abstract with: banning smartphones in middle schools "has no average effect on education or mental health." The circulating "girls' psychologist visits dropped 60%" is one subgroup, one specific year, and counts number of visits — the master switch, whether a child sees a doctor at all, didn't move. Better still: of Norway's 477 "phone-ban" schools, half of the "bans" meant "silent mode in class." The number of schools where students genuinely couldn't reach their phones: four.

England ran the comparison directly: strict schools versus lenient schools — students' well-being, anxiety, and depression all show no difference. The 40 minutes less used at school got made up after school. America's data from 4,600 schools using lockable pouches: average test-score effect "consistently close to zero." Australia's under-16 ban, three months in: 12-13 year-olds using restricted platforms fell from 95% to 86% — down 9 points, and 86% still use them.

Academics do get a small sweetener: Brazilian and Norwegian data both show stricter bans nudge grades up a little (weeks-of-instruction sized). But "banning phones improves kids' mental health" is currently supported by no rigorous study.

This does not mean phones are harmless — in the English study, individual kids who used more were still doing worse. What it shows is that banning a few hours at school doesn't change the day's total dose, let alone the whole environment outside the gate.

What officials and experts actually say

The US Surgeon General's formal advisory stops at "we do not yet have enough evidence to determine if social media is sufficiently safe" — not "proven harmful." But in the same batch of documents, his own quote in the press release calls social media "an important driver of that crisis." The official voice runs on two registers: cautious on paper, alarmed at the microphone.

A large opinion survey of 120+ researchers (note: a preprint, its questions drawn from Haidt's book, with the main critics declining to take part) has two genuinely interesting readings. Asked about social media's overall effect on teen mental health, two-thirds of the experts chose "it depends," and only 26% said clearly negative. Asked about Haidt's three policies (no smartphones before high school, a 16+ age gate, phone-free schools), more than nine in ten agreed the evidence is "too preliminary to support or challenge" them. That survey circulates as "expert consensus backs the bans" — the opposite of what its final version concludes.

One-sentence verdict

The crisis is real (English-speaking countries, especially girls). The harm from phones is real but small (and mostly measured on adults). "The largest single reason" has no evidence. School phone bans measurably do nothing for mental health so far, and a little for grades. Both camps' signature numbers needed corrections — the most useful skill here isn't picking a side, it's asking: which yardstick is this number on?

How to check this article's judgments

The deep dive lists twelve testable claims. Here they are in plain words, strongest first:

1. The deterioration among English-speaking teens (especially girls) is real, not a statistical illusion — watch the US 2025 survey (due fall 2026): does the girls' number stay near 53%?
2. The "care-seeking explosion" is mostly bookkeeping change, but real worsening exists too — watch for other states replicating the New Jersey decomposition.
3. The causal harm from phones/social media exists and is small (0.04-0.11 standard deviations) — watch where the next big deactivation experiment lands.
4. Nobody has closed the "largest reason" gap — if a rigorous study ever attributes double the known magnitude, this claim dies.
5. The time order runs use-first, mood-later (and not the reverse) — watch for an independent team re-running the big US adolescent cohort.
6. School phone bans: zero average mental-health effect — watch any rigorous evaluation of a policy where phones are genuinely out of reach.
7. Small academic gains, growing with strictness — watch the US national pouch study's follow-ups.
8. How much bigger the girls' effect is remains unsettled — the camps' numbers differ by 2-3× — watch for a genuine joint analysis; until then, discount every one-sided number.
9. Australia's ban showed no usage cliff in its first three months (in an underpowered study) — watch the one-year official evaluation.
10. US youth suicide fell once in 2022, then flattened; girls 15-19 ticked up in 2024 — watch whether the official final data confirms it.
11. Kids' usage is no longer climbing across the board (TikTok up, overall intensity down for the first time) — watch Pew's 2026 wave.
12. "Indicators improving while usage stays high" is a live stress test for dose thinking — watch whether the two curves keep diverging.

The things that matter most

• The crisis is real — don't let "it's all a statistical artifact" reassure you. But the cause list is long, and phones are one small proven entry on it.
• For any scary number, ask three things: which yardstick (saying you're sad? seeing a doctor? suicide?), which years, and what is the speaker selling (a book, a platform, a policy)?
• For parents, the evidence points at small average effects with big individual differences: your kid's sleep, offline friendships, and what they actually do on the phone matter more than the daily-hours number.
• Supporting school phone bans is reasonable (small academic gains, fewer classroom distractions) — just don't expect them to cure anxiety: every rigorous evaluation so far says zero average mental-health effect.
• Three signals worth watching: the girls' curve in the US 2025 survey; Australia's one-year evaluation; and whether the two camps ever run a joint analysis.