Schools that made students lock their phones in pouches for the whole school day saw slightly faster growth in math and reading, according to a study released November 17, 2026, by Harvard's Center for Education Policy Research (CEPR). The researchers estimated the boost at about 0.01 standard deviations a year, which they equate to one to two extra weeks of learning.
A standard deviation is a yardstick for comparing results across tests, and 0.01 of one is small. The authors argue it matters because the drop in U.S. scores since 2013 also built up a little each year.
The study tracked about 3.3 million students in grades 3 to 10 at 4,600 schools from the 2021-22 through 2025-26 school years, using fall and spring scores on NWEA's MAP Growth test. Researchers at Dartmouth College and NWEA worked with CEPR, and the Overdeck Family Foundation and the Walton Family Foundation funded the work, according to the report.
"The earlier studies on cellphone bans have been dismissed as implying small impacts, and in-school cellphone use has been ruled out as playing a big role in the decline," said Thomas Kane, a Harvard Graduate School of Education professor and one of the study's authors. "But we think that's mistaken."
How the study measured bans
A strict ban in this study means one product: lockable pouches sold by Yondr, a for-profit company. Students seal their phones in the pouch in the morning, and staff open it with a magnetic device when classes end. The researchers identified pouch schools from Yondr's data and assumed a school kept using pouches after it bought them.
Each pouch school was paired with five similar public schools based on enrollment, poverty, race, past test scores, charter status, and federal pandemic aid. The team then compared how much students grew from fall to spring before and after pouches arrived, a method called difference-in-differences. The final samples included 677 pouch schools in math and 814 in reading, and 72% of the math group were middle schools.
The pouches did cut phone activity. Location data from Advan, a company that collects signals from phone apps, showed about 20% fewer devices on school grounds by the second year. Staff surveyed in five states reported a 28-percentage-point drop in students misusing phones in class.
What the numbers show
| Policy | Math | Reading | Both subjects |
|---|---|---|---|
| Yondr pouches (school-level bans) | 0.008 (no) | 0.009 (borderline) | 0.009 (yes) |
| State bell-to-bell laws, 2025-26 | -0.006 (no) | 0.015 (borderline) | 0.006 (no) |
| State class-time laws, 2025-26 | 0.019 (yes) | 0.006 (no) | 0.012 (yes) |
Estimated yearly change in fall-to-spring growth, in standard deviations. Yes means statistically significant at the 5% level; borderline means significant only at the 10% level; no means not statistically significant. Source: Harvard CEPR, Phone-Free Schools and Achievement Growth, Tables 2 and 6, November 2026.
Statistical significance is a test of whether a result is likely to be real or could be chance. The combined math and reading estimate for pouches passed that test. The math estimate on its own did not, and the reading estimate was borderline.
To turn the result into weeks, the report starts with typical growth. American students gain about 0.3 standard deviations per grade between 4th and 8th grade, so 0.01 is about one-thirtieth of a year's learning. Spread over a 36-week school year, 36 divided by 30 is 1.2, or a little over one extra week, the report said. The authors round the benefit to one to two weeks.
The authors also compare the gain with the national slide. Students who entered 4th grade in 2013 and 2015 grew about 0.02 standard deviations less per grade on the National Assessment of Educational Progress (NAEP) than the 2007 and 2009 groups did. A 0.01 boost is half of 0.02. The gains appeared in the first year and continued after it, though at that pace it would take about two decades to reverse the past decade of losses, the authors' executive summary said.
Results were similar for elementary, middle, and high schools, and by race and gender, except that in math, students who started above the median gained more. In reading, pouches did the most for students having the worst years. The estimated gain was 0.017 at the 10th percentile of growth, compared with 0.009 at the average and 0.002 at the 90th percentile, meaning fewer students slid far backward.
The study also looked at 15 states plus Washington, D.C., that began bell-to-bell laws in 2025-26 and seven states that banned phones only during class time. The authors called those results exploratory because the ban states were already on different score trends before the laws. Phone signal data showed no measurable drop in those states, possibly because students in those states may leave phones on in backpacks, the report said.
Two earlier studies came out differently. David Figlio and Umut Özek, studying one large Florida district, found suspensions rose in the first year of a ban, especially among Black students, and found "significant improvements in student test scores in the second year of the ban." A national pouch study released in April 2026 found that "average effects on test scores are consistently close to zero," with modest gains in high schools, especially in math, and small losses in middle schools.
Brian Jacob, a University of Michigan professor and an author of the April study, said the new findings are "encouraging. And if [the benefit] really persists and accumulates over time, it really could be a meaningful benefit." Another author of the April study, Thomas Dee of Stanford, said he was not comfortable projecting that gains this small, sensitive, and based on a few years of bans will add up.
The report puts pouches at roughly $22 to $25 per student, or $425 for a class of 17. One week of an average public school teacher's $74,910 salary is about $2,080, close to five times as much.
State lawmakers have moved faster than the research. As of October 29, 2026, 46 states had enacted laws or policies on student cellphone use, according to Ballotpedia. Of those, 37 ban or limit phones in classrooms, five require districts to adopt some policy, and four encourage limits. allk12's guide to which states ban cellphones in school covers the bell-to-bell rules state by state, and a separate post looks at why parents and teachers don't agree on phone bans.
The authors said the bigger unknown is phone use outside school. They urged states to use the recent Meta settlement, under which the company agreed to share platform data with a research foundation that has yet to be set up, to track how much time students spend on social media during and after school.
Sources
Harvard CEPR: Phone-Free Schools and Achievement Growth: A Student-Level Analysis (November 2026)
Harvard CEPR: Can In-School Cell Phone Bans Boost Student Achievement? Executive Summary
Harvard CEPR: Phone-Free Schools
NBER: The Impact of Cellphone Bans in Schools on Student Outcomes: Evidence from Florida (Working Paper 34388)
NBER: The Effects of School Phone Bans: National Evidence from Lockable Pouches (Working Paper 35132)
Ballotpedia: State policies on cellphone use in K-12 public schools
The New York Times: New Evidence Shows School Phone Bans Are Gradually Improving Test Scores
NPR: Does locking up a student's cell phone improve test scores? A new study suggests so



