Schools across the United States have increasingly moved to restrict or ban students’ cellphones during the school day, but a new nationally focused study found that the policies are not delivering the rapid academic and discipline improvements many advocates have expected. The research, published Monday by the National Bureau of Economic Research, examined how bell-to-bell bans worked in practice by using data tied to lockable-phone pouches rather than relying on whether phones were theoretically kept out of sight.
The study draws on data from about 4,600 schools and is described as the first nationally representative look at cellphone bans. It also aims to be the first large analysis to track actual phones that were locked away during the school day, not just school “no-show” policies that ask students to keep phones hidden in backpacks or pockets, according to Thomas Dee, a Stanford economist who co-led the project.
Dee said the team wanted to “leverage the data from Yondr because it gives us much more confidence that in-school use of phones is actually being restricted.” Yondr, a California startup, provides lockable pouches for schools, businesses and entertainment venues. Dee contrasted that approach with “no-show” rules, which he said are inconsistently enforced and therefore are a weaker basis for evaluating impacts.
The study’s results begin with an effect on phone use itself. Teacher surveys in schools that banned phones bell-to-bell found that the share of students reporting using phones in class for personal reasons fell from 61% to 13%, and GPS data suggested phone usage declined “with a large and persistent” drop on campuses with bans. Researchers cautioned that device pings are not a perfect direct measure of student use because adult phone use can be included and pings can occur when phones are on but not in active use, though the data still pointed to substantial reductions among students.
Beyond phone use, the study found academic achievement changes were limited early on. Average effects on standardized test scores were described as “consistently close to zero” across the first three years after adoption, and the findings were reported as similar across subjects. Attendance, attention and bullying also showed little measurable movement, with effects on attendance reported as “close to zero” and researchers finding no measurable improvements in perceived online bullying or self-reported classroom attention.
The study also reported changes in discipline and student well-being that appeared to be front-loaded, then stabilized. Schools saw about a 16% increase in suspension rates—both in- and out-of-school—during the first year after adopting bans, but researchers said that effect faded in subsequent years. Subjective well-being dropped in the first year and then rebounded, turning positive by the second year, according to the study.
Dee said the results should be viewed as “sobering,” adding that the lack of strong early gains was “somewhat disappointing.” He also argued that as schools continue the bans, some indicators improve, pointing to the timing pattern in suspensions and well-being: student self-reported well-being fell substantially in the first year, but “within three years” it was above baseline levels, and exclusionary discipline such as suspension “really only occurs in the first year of the phone bans,” returning to baseline by the third year.
The study tracked cohorts of schools that adopted phone bans in 2022, 2023 and 2024. Dee said the newest cohorts had seen test scores rise in a short time, though he said he was not certain why, theorizing that the “entire social context around which we understand phone bans may be changing.” He said people may now be more likely to view the policies as beneficial—something meant to help rather than merely constrain students—compared with several years ago.
Dee cautioned that the findings represent early days of phone bans and he urged giving policies time to show broader outcomes. He said phone bans, at minimum, accomplish what they advertise by reducing student phone use and recapturing attention in classrooms, and he argued that those changes may be “a critical antecedent” to realizing academic potential, suggesting schools may need a couple of years for effects to appear. He also called for schools not to succumb to education “faddishness” and for persistence with a “robust learning agenda” to better understand how devices can be managed while supporting child development.