What the Largest School Phone Ban Study Actually Found, and What It Means for Your Own Phone Rules

Taking a phone out of a teenager's hands is the easy part of a phone ban. It is a pouch, a lock, a rule. The harder part comes later, when someone asks the plain question: did it work? The honest answer from the largest dataset yet is "yes, and also not in the way you were promised."

A 2026 multi-university effort involving researchers at Stanford, Duke, Michigan and Penn looked at schools using locked pouches, roughly 4,600 of them, inside a wider dataset tracking more than 40,000 US schools between 2019 and 2026. I am relaying the figures as they were reported; the study itself is the place to check the detail. What interests me is the shape of the result, because that shape applies well beyond schools.

What improved

The headline number is the one you would hope for. Non-academic phone use during the school day fell from 61% of students to 13%. A rule that sounds soft turned out to be, at least on this measure, quite enforceable.

The softer outcomes moved too. Bullying went down. The gap between students narrowed for girls and for lower-income students, who in several earlier accounts seemed to carry more of the social cost of constant connection. If you have ever watched a lunch table where everyone is looking down, you can picture why a room without screens might feel safer for the person who is easiest to leave out.

What did not move

Test scores did not budge, and neither did attendance. This is the finding that gets quoted in the "bans don't work" headlines, and it is true as far as it goes.

But a ban is not a tutoring program. Nothing about locking a phone away adds a teacher, a lesson, or an hour of sleep. Expecting math scores to rise because a pocket is empty assumes a very short causal chain, and most things that matter in education have long ones.

The suspension spike nobody mentions

There was a cost during the adjustment period: suspensions rose at first. That makes sense if you think about it for a minute. A new rule creates a new way to break the rule. Some students tested it, some forgot, and some schools enforced it with the tools they already had.

The useful reading is that transitions have a price and the price is front-loaded. A school that expects the spike and plans for it (clear consequences, a grace period, a way to hand the phone back) is in a different position from one that is surprised by it in week two.

Why "no effect on grades" does not mean "no effect"

This is the sentence I would put on a card if I could. A study measures what it measures. Grades and attendance are lagging, noisy, heavily influenced by things a phone policy never touches: home life, teacher quality, sleep, a bad year.

Meanwhile, a student who is no longer being photographed in the hallway, or who eats lunch while talking to someone, may be a very different student without being a different test-taker. Both of those can be true at once. The outcomes that moved are real outcomes. They are just not the one that fits on a report card.

A framework: attention outcomes versus achievement outcomes

When you evaluate any intervention meant to help people focus, it helps to keep two columns, because they run on different clocks.

Attention outcomes are about the day itself: how often you reach for the device, how long you stay with one thing, how present the room feels, how you feel at 4 p.m. These change fast, often within days.
Achievement outcomes are about what accumulates: grades, shipped work, skills, income. These change slowly, and many other forces push on them.

An intervention can win clearly in the first column and show nothing in the second for months, or ever, and still be worth keeping. It can also fail in the first column, in which case you should stop. The mistake is judging a first-column tool by second-column numbers, then abandoning it.

Applying it to a personal phone policy

Say you decide the phone stays out of the bedroom, or off the dinner table, or in a drawer during the first two hours of work. Before you start, write down what you are actually testing.

Pick one attention measure. Pickups per day from your phone's own screen-time report, or simply "did I read for twenty minutes without reaching for it."
Pick one achievement measure, and expect it to lag. A finished chapter, a shipped feature, a savings transfer made on time.
Give the adjustment period room. The first week is mostly friction. Judging it then is like judging a new route on the day of the road closure.
Check again at four weeks, not four days.

In my own experience, the first thing that changes is not output. It is the quality of the gaps: the minute in the queue, the pause before a reply. Those do not show up on any dashboard, but they are where a lot of thinking used to happen.

At home and at work

The study also hints at something about how bans work. They were effective in part because everyone was inside the same rule. A single student putting a phone away at a table where nobody else has is doing something much harder than a whole room doing it together.

For a household, that suggests the adults go first, or at least go along. For a team, it suggests that a shared norm (no devices in the first ten minutes of a meeting, say) is worth more than ten individual resolutions. Where it is just you, it still works. It simply costs more willpower, so make the rule smaller and the cue clearer.

FAQ

Do school phone bans improve grades?

In this dataset, test scores and attendance did not change measurably. Improvements showed up in phone use, bullying and the gap for girls and lower-income students. Grades may simply be the wrong place to look for early effects.

Why did suspensions rise at first?

A new rule creates new violations, and enforcement tends to be strictest while everyone is still learning it. The spike was described as part of the adjustment period rather than a lasting feature.

Does this apply to adults?

The study was about schools, so it does not prove anything about offices or homes. The logic carries over reasonably well, though: judge a focus tool on attention first, and give it a few weeks before looking for results in your work.

Is a full ban necessary, or does a partial one help?

This study does not answer that, and I would be guessing. A partial rule, such as phones out of one room or one time block, is a cheaper experiment to run on yourself.

Maybe the useful question to carry out of the study is the one it quietly asks of every habit we try to change: what exactly did we hope would be different, and how long did we agree to wait before looking?