Nearly 5,000 Schools Locked the Phones Away. Year One Looked Like a Mistake.
The largest study yet of lockable phone pouches found test scores flat, suspensions up and pupils feeling worse in the first year — and pupils feeling clearly better by the second.
In short
What happened. Researchers tracked the 4,607 US public schools that have bought Yondr lockable phone pouches since 2014 and compared them with matched schools that did not. Phone use in school fell steeply. Average test scores did not move.
What it means. The first year is bumpy. Suspension rates rose by roughly 16 percent in the year a school switched, and pupils reported feeling worse. By the second year afterwards, reported wellbeing was clearly better than in comparison schools.
Risks and impact. This is US data, on one product, followed for at most three years, in a working paper not yet peer reviewed. It describes school averages, not what a phone does to one child.
What can be done. Ask a school what its phone rule was before the new one, since the size of any change depends on that starting point, and whether it publishes discipline or wellbeing figures comparable year to year.
What to watch. Whether a fourth and fifth year of data confirm the second-year improvement, or reveal it as an artefact of which schools adopted when.
Shown as a summary because of your reading settings.
What happened
A Yondr pouch is a fabric sleeve with a magnetic lock. The phone goes in at the door, the pouch seals, it stays in the bag all day, and a magnet by the exit opens it at home time. Since 2014 the company has sold them to 4,607 US public schools.
In an April 2026 working paper for the National Bureau of Economic Research, six researchers — Hunt Allcott, Jason Baron, Thomas Dee, Angela Duckworth, Matthew Gentzkow and Brian Jacob — used Yondr’s own sales records to date each school’s switch, then compared those schools with matched schools that never adopted. Their evidence spans state test scores, attendance and discipline records, pupil surveys from Panorama Education, GPS pings and two new national surveys.
The pouches worked as hardware. Teachers reported the share of pupils using phones in class for personal reasons falling from 61 percent to 13 percent. GPS pings during school hours dropped about 30 percent by the third year.
What that bought was less obvious. The pooled effect on test scores was −0.004 school-level standard deviations, standard error 0.010: close to zero, and precisely so. Attendance moved −0.138 percentage points against a mean of 93 percent, and perceived online bullying did not shift.
Two things did move. In the adoption year, suspension rates rose about 16 percent and pupil-reported wellbeing fell by roughly 0.2 student-level standard deviations. By the second year after adoption, the discipline effect had faded and wellbeing stood 0.16 standard deviations above comparison schools.
What the evidence supports
Three things in this study are solid. The analysis was pre-registered, so the outcomes were chosen before the answers were known. Pouch adoption really does cut phone use in school, confirmed by two independent measures — anonymous device pings and teacher reports — that have nothing to do with each other. The academic result is not a shrug of uncertainty but a tight null: the authors can rule out average improvements larger than about 0.008 student-level standard deviations. And the averages hide a split. High schools gained modestly in maths, about 0.024 student-level standard deviations, or 0.9 of a percentile point. Middle schools went slightly the other way.
Two things rest on thinner ground. The wellbeing measure comes from one survey provider, and the adoption dates come from the pouch company itself, which partnered with the researchers. Nothing suggests the numbers were shaped by that, and the funders are named openly, but it is a single supplier of a key input.
The most important caveat is the authors’ own. The story of a first-year dip followed by a rebound is partly a story about different schools. The sharp fall in wellbeing is driven by schools that adopted in 2025 and are only observed in their first year. The recovery comes from schools that adopted in 2023 and have been followed longer. With at most three years of data, the paper cannot fully separate a genuine adjustment curve from the possibility that late adopters were simply having a worse time.
How the story is being framed
Pouches do not work. The headline reading, and the one that travelled. It has support beyond this study: a survey of 1,227 pupils across 30 English secondary schools, published in The Lancet Regional Health Europe in February 2025, found no wellbeing difference between restrictive and permissive schools — an adjusted mean difference of −0.48, confidence interval −2.05 to 1.06. What it leaves out is that the same study found restrictive schools did cut phone use during the school day by about 40 minutes, without changing total daily use at all.
You are measuring the wrong thing. Wellbeing did move, and by the second year it moved the right way: 0.16 student-level standard deviations. For scale, an earlier experiment by the same lead author found four weeks off Facebook raised participants’ wellbeing by 0.09. A test score is what schools happen to count, not what most parents lie awake about.
Bans work, but the first year has a bill, and it is not shared evenly. In one of the ten largest US districts, a bell-to-bell ban studied by David Figlio and Umut Ozek pushed suspension rates up 25 percent in year one. In schools with heavy pre-ban phone use, in-school suspensions for Black pupils rose about 30 percent, with no significant change for white or Hispanic pupils. Both effects vanished in year two, when spring scores in those high-use schools came out about 1.1 percentiles above schools where phone use had been lighter.
It depends who the child is. Sara Abrahamsson’s study of Norwegian middle schools, published in the Journal of Human Resources in March 2026, reports no average effect on education or mental health — but lower bullying for both sexes, fewer psychological consultations for girls, and better grades and academic track placement for girls, especially those from disadvantaged backgrounds.
The background
This is now a majority policy. UNESCO’s Global Education Monitoring Report counted 114 education systems with a national ban on phones in schools in March 2026 — 58 percent of countries, up from 24 percent when it first measured in June 2023 and 40 percent by early 2025. Florida legislated first among US states in 2023; writing in May 2026, Figlio and Ozek counted forty more states with restriction laws, roughly two dozen of them bell-to-bell.
What the label hides is how much variation sits underneath it. In a Gallup survey of about 2,000 US adolescents and their parents run for this study in February 2026, 39 percent of pupils described a bell-to-bell rule, 48 percent said phones were allowed at lunch or between lessons, and 13 percent reported no formal rule at all. Only 8 percent used pouches. Half were on a no-show policy: keep it in your bag, out of sight.
That matters more than it sounds, because every estimate in this field measures a change from something. Most schools that bought pouches were previously running loose rules. Among the adopting schools whose teachers answered the researchers’ survey, about 62 percent had been on a no-show policy and another 23 percent had no formal rule at all. So the finding is not “phones versus no phones”. It is “sealed pouches versus a rule that was already there and half-working”. A school with a strictly enforced policy should expect a smaller change than the study shows. A school with no rule at all should expect a larger one.
The same survey found parents and pupils looking at the same policy and seeing opposite things. On a scale from −1 to 1, parents supported pouches at 0.369 and expected mental health to improve most of all, at 0.597. Pupils opposed them at −0.353.
The deeper story
Look at what the study could count. Test scores, attendance, suspensions — three things schools already record because they are easy to record. Then one thing that is hard: whether a fourteen-year-old feels all right. That last one was the thing parents ranked highest when asked what they hoped a pouch would fix. It is also the only one that clearly moved.
It moved down first. That detail is worth sitting with, because it is the shape of almost every rule worth having, and it is the shape most institutions are least equipped to survive. A school introduces a policy in September. Suspensions climb, because staff are now catching something they previously ignored. Pupils are cross, and say so on the survey. Parents read the write-up in the spring newsletter and conclude the thing failed. The improvement, if it comes, arrives two years later, quietly, in a measure nobody was watching.
The pupils in that survey were not wrong, either. They were describing their first year accurately. Something was taken away, and the loss was real and immediate, and the compensating benefit was invisible and slow and not theirs to choose. Adults tend to file that as adolescent unreasonableness. It is closer to an honest report from inside the transition — the same report an adult would file about any rule imposed on them for their own eventual good.
What the evidence cannot tell any parent is whether a phone is bad for their particular child. It was never that kind of study. It can say what happens when several thousand schools take the phones for the day: a hard autumn, a suspension spike that fades, exam results that stubbornly refuse to notice, and a slow return of something the school can barely measure and cares about most.
Something to sit with
If a rule in your own household got worse before it got better, how long would you give it before deciding it had failed?
When a policy improves something that cannot be scored, who decides whether it was worth it — and on what evidence?
Sources
- NBER working paper 35132 — The Effects of School Phone Bans: National Evidence from Lockable Pouches — https://www.nber.org/papers/w35132
- Allcott, Baron, Dee, Duckworth, Gentzkow and Jacob — full text of the phone-pouch study (Stanford) — https://web.stanford.edu/~gentzkow/research/PhoneBans.pdf
- Education Next — Figlio and Ozek on Florida's bell-to-bell cellphone ban — https://www.educationnext.org/wp-content/uploads/2026/05/ednext_26_...
- NBER working paper 34388 — The Impact of Cellphone Bans in Schools on Student Outcomes: Evidence from Florida — https://www.nber.org/papers/w34388
- University of Birmingham research record — SMART Schools, The Lancet Regional Health Europe — https://research.birmingham.ac.uk/en/publications/school-phone-poli...
- Journal of Human Resources — Abrahamsson, Smartphone Bans, Student Outcomes and Mental Health — https://jhr.uwpress.org/content/early/2026/03/04/jhr.0224-13403R2
- UNESCO Global Education Monitoring Report — Phone bans in schools are spreading worldwide — https://www.unesco.org/gem-report/en/articles/phone-bans-schools-ar...
We report facts from the sources above in our own words and link to the originals. Interpretation is ours, not theirs.
In the nationwide US study of lockable phone pouches, what happened to average standardised test scores after schools adopted them?
The pooled estimate was close to zero and tightly measured. The authors say they can rule out average improvements larger than about 0.008 student-level standard deviations.
♻︎ Free to republish
Copy this HTML into your CMS. Credit line and licence are included. Republish our work — free
Every headline has a deeper story. This is ours.
What we are doing here