A blanket ban on mobile phones during school hours appears to bring significant benefits for school performance, according to a study by Harvard University's Center for Education Policy Research (CEPR). The study examined data from 3.3 million students at 4,600 schools in the United States. It found that strict restrictions are linked to better results in maths and reading, with the gain equivalent to one to two extra weeks of learning a year. The researchers themselves stress, however, that the results are modest, and that a ban cannot solve everything.
One of the largest studies to date
The debate over smartphones in schools has taken on international dimensions, as more and more education authorities try to limit their presence in students' daily lives. As The New York Times highlights, the key question is whether removing phones from the classroom can translate into better student performance. One of the largest studies carried out so far now adds new data to the debate.
The study was published on 17 November 2026 by Harvard's CEPR, based at the Harvard Graduate School of Education. It was carried out in partnership with researchers from Dartmouth College and NWEA, a non-profit educational assessment organisation, and was led by CEPR faculty director Tom Kane. Dartmouth economist Douglas Staiger is a co-author. According to Harvard, the study was funded by the Walton Family Foundation and the Overdeck Family Foundation.
The main conclusion is that in schools with strict bans, where phones had to be stored in special lockable pouches throughout the school day, students' results in maths and reading improved.
The gradual decline in student performance
One of the elements that makes the study particularly interesting is its starting point. Since 2013, American students' performance in maths and reading has been in steady decline. This coincided with the widespread adoption of smartphones and the integration of social media into teenagers' daily lives.
According to Harvard, between 2013 and 2024, eighth-grade students in the US lost learning ground equivalent to about one school year in maths and reading. The loss did not happen suddenly. It came through small annual declines of roughly 0.01 to 0.02 standard deviations, or two to three weeks of learning a year, which added up to a significant shortfall. The pandemic and school closures made the situation worse.
The question the researchers set out to answer was whether the presence of mobile phones in schools contributed to this decline, and whether restricting them could reverse part of the trend.
"Previous research on phone bans has often been interpreted as a sign that their effects are small," said Tom Kane, a professor at the Harvard Graduate School of Education. He argues that the new analysis shows the effect of phones on learning should not be underestimated.
How the study was carried out
The methodology is one of the study's most important features. The researchers did not simply compare average results at schools that banned phones with those at schools that allowed them. Instead, they tracked the progress of the same students before and after the bans were introduced.
To do this, they used results from NWEA's MAP Growth tests, an assessment system that records students' progress from the start to the end of the school year. The analysis covered the school years from 2021-22 to 2025-26, in schools in 49 states and the District of Columbia, among students from fourth to tenth grade.
The researchers focused on schools that applied genuinely strict restrictions, mainly through lockable Yondr pouches. A general recommendation not to use phones in class was not enough: phones had to be stored in pouches that were locked at the start of the school day and unlocked at the end. According to Harvard, the study also examined statewide bans introduced in 22 states in 2025-26.
This difference proved crucial. Simply banning use in the classroom does not necessarily mean students lose access to their devices during breaks or other school activities.
Two minutes for every hour of class
According to Harvard, schools using pouches recorded an annual gain of 0.01 standard deviations in achievement growth. That is about one to two extra weeks of learning a year, or roughly two minutes for every hour of class. The gains appeared in the first year and were still present three years later.
The improvement was not limited to maths but also extended to reading, a finding the researchers consider particularly significant. Reading a text, understanding complex concepts and sustaining attention for long periods require a kind of concentration that runs counter to the way modern digital apps work. Constant notifications, switching between content and the habit of consuming information quickly may make it harder to stay focused on an activity that demands time and mental effort. The study also found that fewer students saw sharp drops in their reading performance during the school year.
The benefit was not confined to students with low initial performance. According to Harvard, effects did not differ significantly between primary, middle and secondary schools, by school poverty level or by prior achievement. This suggests that removing phones may benefit the learning process as a whole.
For the statewide bans, the findings are described by Harvard as suggestive. Bell-to-bell bans, covering the whole school day, were linked to a 0.016 standard deviation gain in reading. Bans limited to instructional time were linked to a 0.019 standard deviation gain in maths. The researchers caution that these results may reflect trends that already existed in those states.
Why the results are not spectacular
Despite the positive findings, the researchers avoid presenting phone bans as a solution to education's problems. The gain is a measurable but limited improvement, and it cannot compare with the effect of more intensive interventions, such as personalised teaching.
However, as Kane points out, if the losses built up over years, it is reasonable to expect that the benefits of restricting phones will also take time to become more noticeable. Based on the team's calculations, strict bans appear to offset about half of the annual decline in performance recorded since 2013. "Strict cell phone bans seem to plug about half of that leak," Kane said, according to Harvard.
Douglas Staiger described storing phones as "a cost-effective way to add 1–2 weeks of instruction each year".
In other words, removing smartphones from schools does not automatically bring performance back to previous levels, but it appears to be a measure with a real, if gradual, effect.
Reservations and open questions
The new study is not the first to try to measure the effect of bans. That matters, because earlier studies reached less encouraging conclusions.
A major study published in May 2026 under the auspices of the National Bureau of Economic Research (NBER), involving researchers from Stanford, Michigan, Duke and Pennsylvania universities, examined more than 40,000 schools. The researchers found that the bans significantly reduced phone use during school hours, but did not find a correspondingly significant overall improvement in scores. Positive results were most visible in secondary school students' maths, after three years of restrictions.
Thomas Dee, an education economist at Stanford University and one of the researchers behind the NBER study, has expressed doubts about whether relatively small improvements can really build up and lead to meaningful long-term change.
One problem is that banning phones does not necessarily eliminate digital distractions. Students may still use laptops or other devices to access social media. In addition, bans are not always enforced effectively. In some schools, students find ways around the restrictions, even using second devices or other inventive methods to stay online.
Psychologist Jean Twenge, a professor at San Diego State University, also points out that having a ban policy does not necessarily mean it is applied in practice. Even when phones are removed from the classroom, digital distractions may simply move to other devices.
The bigger problem lies outside school
Perhaps the most important question raised by the new study concerns what happens after school hours. How much does smartphone use during homework affect learning? How much time do teenagers spend on social media instead of studying? And what happens when phone use continues late into the night, at the expense of sleep?
The researchers acknowledge that they do not yet have enough data to answer these questions accurately. "We have lacked the data for understanding the role that cell phones are playing in students' lives," Kane said, according to Harvard.
Kane believes that access to usage data from digital platforms could make a decisive contribution to understanding the link between the time teenagers spend online and their school performance. As he points out, social media companies know very precisely how much time minors spend on their platforms, but this data has not been systematically linked to student performance.
According to Harvard, the researchers point to a recent Meta settlement limiting minors' access outside school hoursas an opportunity to study this. They make three recommendations:
- Extend social media restrictions beyond Meta to platforms such as YouTube, TikTok and Snapchat, so that students cannot simply switch apps.
- Create baseline measures of social media use, using de-identified, aggregated data from the companies, tracked before and after restrictions take effect.
- Track student engagement and well-being, including extracurricular participation, reading time, homework completion, in-person socialising, and anxiety and depression.
The issue, therefore, goes beyond whether a child has a phone in their pocket in the classroom. It concerns a broader shift in how young people concentrate, inform themselves, relax and communicate.
With information from The New York Times and the Harvard Graduate School of Education


