1. A public controversy that moved faster than the evidence

The phone ban ceased to be an arrangement internal to the school and became a public controversy before evaluations existed that could say which outcome changes, in whom, and with what degree of compliance. Montag and Elhai (2023) record this in an argumentative essay, not in a finding: France and China had established bans, and the text asks whether a digital school uniform is needed, the same rule for everyone. It weighs better learning environments and face-to-face interaction against the arguments on the other side. It does not estimate an effect.

The turn did not always originate in pedagogy. Paredes-Labra et al. (2025) document, in a three-year project, the passage from initial tolerance to prohibitive policies in Spanish secondary education. The qualitative design draws on actor-network theory: policy and media documents, nine interviews, ten regional cases, twelve focus groups and a session for returning findings. Madrid, among the first to restrict phones, contributed two school cases and three groups with teachers, students and families. The ban was motivated less by pedagogical evidence than by cultural anxieties, a techno-pessimistic shock. They call for nuanced, multi-level governance, not the binary of promotion versus prohibition.

Other jurisdictions reached the same place by different instruments. Santos de Souza Junior (2026) problematises Brazilian Law 15.100/2025 as critical discourse analysis, not as an impact evaluation; the argument returns when Latin America is taken up. Sullivan et al. (2026) describe England: guidance that promotes phone-free environments, without a legal mandate at the time of the study. Moreno et al. (2026) find, among parents in the United States, a school rule that was already nearly ubiquitous in their sample.

Positions do not converge either. Livingstone (2026), in a viewpoint, argues for allowing the phone where policies reduce harm and support learning. Out of that disagreement comes the question of this review: which outcomes the restriction improves, under what conditions of implementation, and at what cost. The answer does not fit a yes or a no: it depends on the type of rule, on compliance and on the outcome.

2. What exactly is banned: a typology of school phone policies

Before asking whether the ban works, one has to ask what object that word names. As a conceptual framework, this review distinguishes seven forms that the corpus studies. Campbell et al. (2024) warned, in a scoping review, that definitions range between a partial ban and a complete ban, and that this makes results difficult to reconcile. Sullivan et al. (2026) confirm the point: the content of the rules varies, and use of the word ban is inconsistent.

Policy type What is restricted and when Example from the corpus Outcomes that have been studied
Total ban on the premises Phone use throughout the school day Vereecke et al. (2026), described as a total ban Social loneliness, friendship support and cyberbullying
Restriction of recreational use versus a permissive policy Recreational use is not allowed, or it is allowed Goodyear et al. (2025); Randhawa et al. (2025) Mental wellbeing, phone time and social-media time, support for the rule
In-class ban with collection The device is collected at the start of each class Grigic Magnusson et al. (2023) Implementation climate, exceptions and pedagogical use
Recess ban Use during recess, four weeks in the study Pawlowski et al. (2021), six Danish schools Physical activity: frequency, moderate and vigorous
Phones off while school is in session Access restricted during the school day; phased rule King et al. (2024), five schools, grades 7 to 12 Problematic use, engagement, belonging, bullying and compliance
Medium restriction, as reported by families The classroom is restricted and the phone is allowed outside class Moreno et al. (2026); the most common formula in their survey Presence of the rule, family contact and violations
Formal policy versus absence of a policy There is a document, with or without the rhetoric of a ban Sullivan et al. (2026); some of the schools had no policy Digital safeguarding incidents and their recording
Table 1. Typology of school phone policies in the corpus: what is restricted, when, in which study it appears and which outcomes have been examined. A conceptual framework, not an estimate of effects.

As a design inference, no row authorises extrapolating its outcome to the others. Vereecke et al. (2026) look at loneliness and cyberbullying, not at grades; Randhawa et al. (2025) describe the rationale, not a health effect; King et al. (2024) evaluate phones switched off during the school day, with the rule phased in during 2023 in South Australia. Moreno et al. (2026) collect from parents an intermediate formula — limiting the classroom and allowing the phone outside it — that the binary debate erases. Sullivan et al. (2026) show that having a document is not the same as sharing a definition. More movement at recess is not better mental health, and less use during school hours is not less use across the day.

3. Method: critical narrative review with a screening flow

This is a critical narrative review, not a meta-analysis. The base is the Crossref REST API. No pooled effect of our own was estimated: each effect size that follows belongs to its authors. Initial screening was lexical and reproducible with screen.py; final inclusion was manual. There was no double coding.

Table 2 fixes the flow: 16 seed queries by 60 rows; 390 unique verified records from 2021 to 2026, with no exclusions by year; 2 corrections, replications or supplements; 4 news items; 61 doubtful journals; 37 out of scope, among them higher education and nursing, myopia, radiation, preschool, educational applications and smartphone addiction with no link to policy or to school outcomes; 286 eligible in the seed. The seed did not retrieve key evaluations. A complementary search of 14 queries added 11 records verified one by one. The flow justifies that complement by gaps such as SMART Schools, Norway, Spain, the economic analysis, Madrid and Brazil; Norway remains in that justification and is not assigned as a country to an abstract that does not declare it. Manual inclusion closed at 24 works, 13 from the seed and 11 from the complement: 21 with an abstract, 3 with title only, 1 Latin American study and 0 from Mexico.

The criterion admitted experiments and quasi-experiments, large observational studies, reviews, qualitative work on implementation and peer-reviewed articles in compulsory education, plus three opinion pieces as contrast and not as effect: the essay by Montag and Elhai (2023), the viewpoint by Livingstone (2026) and the letter by March et al. (2026). Each use is labelled as an empirical finding, a conceptual framework or a design inference, and the design is named: experiment, quasi-experiment (synthetic control, difference-in-differences or event study), cross-sectional observation, prospective follow-up, qualitative work or review.

Stage of the flow Magnitude
Seed queries and rows per query 16 queries; 60 rows each
Unique verified records, 2021–2026 390; none excluded by year
Corrections, replications or supplements 2
News items 4
Doubtful journals 61
Out of scope 37
Eligible after screening of the seed 286
Targeted complementary search 14 queries; 11 records verified one by one
Included by manual selection 24: 13 from the seed and 11 from the complementary search
With an abstract / title only 21 / 3
Studies from Latin America / from Mexico 1 / 0
Table 2. Screening flow of the critical narrative review, with Crossref as the base and final manual selection, without double coding or a meta-analysis of our own.

4. Attention and classroom climate: the most consistent outcome

Compared with mental health, attention leaves more legible signals, not a single gain and not one that is always favourable. That it is the most consistent outcome is a comparison, not an agreement across designs.

Sungu et al. (2026) can be cited only with the deposited sentence. In their massive experiment on phone-free classrooms, those with lower achievement were the ones who benefited most, and even those with higher achievement supported the policy. The title states that phones impede learning. There is no effect size, no country and no sample, and support from higher achievers does not prove an equal gain.

King et al. (2024) measure engagement with a preregistered design: ban or no ban, at the start and at one month. The students were in grades 7 to 12 in five public schools in South Australia, 1282 students and then 1256, with measures from the Longitudinal Study of Australian Children and from PISA. There were no significant differences in academic engagement or in belonging. The authors read the short-term benefits as limited or null. That empirical finding prevents turning Sungu et al. (2026) into a general law.

Bar et al. (2025) analyse a preregistered policy experiment in five secondary schools in South Australia: 1549 students, 7188 responses and 69589 words, in five categories and 16 themes. The benefits that are narrated are face-to-face interaction, health and safety, and engagement in the classroom: experience, not time on task. Goodyear et al. (2026) place learning among the dimensions of wellbeing, together with health, connection, agency, resilience and safe environments. In 40 focus groups, 7 English schools (4 permissive and 3 restrictive) and 177 participants, both rules had positive and negative impacts depending on context. Intervening on the phone, they conclude, has to cross the school and what happens outside it.

Grigic Magnusson et al. (2023), in a longitudinal qualitative study, follow an upper-secondary team that banned the phone in class: notice at the start of the semester and collection before each session. The strategy failed. There were exceptions; collection was slow and generated tensions. With teachers’ consent, the phone was already a pedagogical tool. As a design inference, climate is decided in the procedure.

5. Academic achievement: small, heterogeneous and unequal effects

Achievement is the outcome most often invoked and one of the least univocal. The average is small or null; some subgroups do move.

Beneito and Vicente-Chirivella (2022) use a quasi-experiment. Two Spanish autonomous communities banned the phone from 2015; the contrast with the rest of the regions combines synthetic control and difference-in-differences, on PISA in mathematics and science and on bullying. The abstract attributes to Galicia effects equivalent to 0.6–0.8 years of learning in mathematics and to around 0.72 up to nearly one year in science. It does not detail an equivalent gain in the second treated region, and that silence is not filled in here. The fall in bullying does appear in both. The learning gain is regional.

Abrahamsson (2026), with an event study, finds that banning the smartphone in middle schools has no average effect on education or on mental health. The null conceals gender: girls’ grades improve, and so does their academic trajectory, especially in disadvantaged contexts. The idea of a low-cost policy, above all for them, is the author’s conclusion, not a mean effect. Extending it to the whole student body would betray the design.

Böttger and Zierer (2024), in a rapid review with PRISMA and five studies, estimate d = 0.162 (p < 0.05), modest and more marked in social wellbeing than in achievement. The small academic effect may be due, they warn, to the narrow base. They recommend educational measures and evaluation alongside the ban, and a literacy that might one day make the ban unnecessary: that is their inference, not the finding. The finding is the modest effect. It is not the meta-analysis of this review.

Campbell et al. (2024), with PRISMA-ScR and a preregistered protocol, bring together 22 studies on learning, mental health and cyberbullying, with no randomised trials and with partial or complete definitions that are difficult to reconcile. Sungu et al. (2026) add that those who benefited most were the lower achievers. This is not merged with Abrahamsson (2026): one speaks of prior achievement and the other of gender and disadvantage, and only the second declares a null average.

6. Wellbeing and mental health: what a ban does not solve on its own

In mental health the controversy moved ahead of the data. Campbell et al. (2024) recall the usual argument — distraction, cyberbullying, worse mental health — and do not find randomised trials. The restriction, merely by existing, does not resolve distress.

Goodyear et al. (2025) compare, in a cross-sectional observational study from SMART Schools, 30 English schools: 20 restrictive, with no recreational use, and 10 permissive. In 1227 adolescents aged 12 to 15, the Warwick-Edinburgh Mental Well-Being Scale averaged 47 (standard deviation of 9), with no difference (−0.48; 95% CI: −2.05 to 1.06; p = 0.62). During the school day there was less phone use (−0.67 hours; 95% CI: −0.92 to −0.43; p = 0.00024) and less social media (−0.54 hours; 95% CI: −0.74 to −0.36; p = 0.00018), and no difference on weekdays or at weekends. There is no association with lower overall use or with better wellbeing; school-day bans, in their current form, are not supported.

Perry et al. (2026) estimate cost-utility in the same study. In 815 pupils aged 12 to 15 and 20 schools (13 restrictive and 7 permissive) the differences were 0.009 quality-adjusted years (95% CI: −0.014 to 0.032) and −0.004 mental-wellbeing-adjusted years (95% CI: −0.044 to 0.036). Time equated to 3.1 full-time posts in restrictive schools and 3.3 in permissive ones; the abstract does not say whether this is per school or in aggregate. The incremental cost was −94 pounds per pupil per year (95% CI: −229 to 41), with a 90% probability of cost-effectiveness at 20 000 and at 30 000 pounds per quality-adjusted year. The conclusion is not a health conclusion: health barely differs, there may be a saving of time, and practices must be developed because teachers devote a great deal of time to managing the phone.

Abrahamsson (2026) finds no average effect on mental health and does find less psychological care among girls. Vereecke et al. (2026) follow 960 students (mean of 14.71 years, standard deviation of 1.81; 66.6% female) across eight waves, in a single school. In the short term, loneliness did not change, nor did support from friends, nor cyberbullying victimisation; perpetration fell slightly. Afterwards, loneliness rose modestly, support held and cyberbullying fluctuated at a low level. There was no clear improvement; they ask that social connection and digital literacy be added. Radunz et al. (2026), a prospective study of social media and mental health under the South Australian ban, has no abstract: there is neither direction nor magnitude.

King et al. (2024) measure problematic use in 2.6% of their sample, with no difference between schools with a ban and schools without one: restricting access may not touch the mechanism. Less use at school, in Goodyear et al. (2025), was not less use on weekdays or at weekends. Banning the device, treating problematic use and regulating social media outside school hours are three tasks.

7. Peer relations, cyberbullying and digital safeguarding

Peer relations are where the quasi-experiments agree more than they do on marks or on wellbeing. Even there, the agreement is not total.

Beneito and Vicente-Chirivella (2022) find less bullying in both regions treated from 2015, unlike PISA, which the abstract limits to Galicia. The design is synthetic control and difference-in-differences, at a regional grain. Abrahamsson (2026) finds less bullying in both genders, whereas grades, trajectory and psychological care move above all among girls. Peer relations can change when the rule varies over time and across territory; a school regulation does not inherit that result.

King et al. (2024) see slightly higher bullying in the group with a ban and, at the same time, a significant fall in both groups: the decline cannot be attributed to the rule, and the follow-up lasts one month. Vereecke et al. (2026) find no short-term change in victimisation; perpetration falls only slightly and then fluctuates at low levels. Böttger and Zierer (2024) do say, from five studies, that bans can reduce social problems such as bullying, with more weight in their d = 0.162 than achievement carries. That reading is coherent with the quasi-experiments and too coarse to cancel those nuances.

Sullivan et al. (2026) use information requests in 114 English schools and more than 90000 pupils, in 2023/2024. Sixty of 114 (52.63%) had a policy, with variable content and an inconsistent use of the word ban. Digital incidents were 4.9% (4183 of 85540), above 10% in 36 of 114 and above 25% in 9. Those that had a policy reported more incidents, regardless of the restriction; restrictive schools referred more and reported more events at school, although most occurred outside. The cross-sectional finding associates the rule with recording, not with cause and not with avoidance. They call for definitions of risk.

Pawlowski et al. (2021) contribute a collateral benefit of recess, not of bullying. They banned the smartphone for four weeks, between August and October 2020, in grades 4 to 7 (ages 10 to 14) in six Danish schools. With 814 children beforehand and 828 at the end, the frequency of physical activity rose (odds ratio of 1.370), as did moderate activity (1.387), and vigorous activity fell (0.851), outdoors and indoors, among boys and girls, and almost equivalently across grades. It is not a trial described as randomised. As a design inference, removing the phone from the playground without another activity can change the mix of intensities, not only the amount.

8. Implementation and compliance: the written policy versus the lived one

Moreno et al. (2026) survey, cross-sectionally, parents of young people aged 12 to 17 who attend school in person and have a smartphone. The quota sample, aligned with the United States census, gathers 2502 parents (62.5% women, 15.9% Hispanic, 63.6% White). Of the children, 33.5% had a health condition or a disability. A policy was reported by 98.7%; the most common (46.2%) limited the classroom and allowed the phone outside it. Contact was daily (25%) or several times a day (38.5%). A reported violation was more likely where there was a health condition or a disability (odds ratio of 2.09; 95% CI: 1.71 to 2.57) and where there was more parental contact (1.87; 95% CI: 1.51 to 2.32). This is parental report, not an audit, and the authors ask that families be involved. The secondary analysis by Baggio et al. (2026) concerns implementation of the ban in South Australia: there is no abstract and no direction of effect.

Sheng and Lipscombe (2026) review international research and Australian documents in order to clarify challenges and implications for leadership. The abstract carries no figures; the paper is cited as a review, not as an estimate. Randhawa et al. (2025) describe 30 English schools with 1198 students, 53 teachers and 30 school leaders. The phone appears as a good for safety, learning and communication, yet most schools adopted restrictive rules on grounds of achievement, behaviour and safeguarding. Support differed between students and teachers, and between students in permissive schools and students in restrictive schools.

In King et al. (2024), older age and greater use of social media predict lower compliance. Grigic Magnusson et al. (2023) show exceptions and a collection judged to have failed, with the phone still in class as a tool used with consent. Goodyear et al. (2026) add that context decides the sign for health, connection, learning, agency, resilience and supportive environments, under restrictive rules and under permissive ones alike.

Link What has to happen Where the chain is lost
Written policy Name which use is restricted and at what moment The word ban covers incompatible rules
Communication and legitimacy That students, families and teachers accept the rule Support divided between students and teachers, and across schools
Everyday enforcement A repeatable procedure, with explicit exceptions Costly collection, improvised exceptions and tension
Compliance That the restriction occurs and is not merely announced Lower compliance with older age and more social media; violations linked to family contact
Use of recovered time Give attention and recess a school purpose Lack of a pedagogical alternative or of an activity at recess
Outcomes Read attention, achievement, peer relations, wellbeing and equity separately Expecting the label to move mental health or problematic use
Figure 1. A conceptual scheme, not a fabricated datum: the chain from the written policy to the outcomes, and the points at which implementation is lost. The cells organise mechanisms from the corpus; they are not coefficients.

The chain breaks at divided legitimacy (Randhawa et al., 2025), at failed collection (Grigic Magnusson et al., 2023), at lower adherence among older students or heavier users of social media (King et al., 2024) and at family contact associated with more violations (Moreno et al., 2026). If the time that is recovered has no destination, the final link stays empty. Perry et al. (2026) recall that managing the phone already consumes staff under both types of policy. Implementation is the design, not its residue.

9. Unintended effects and equity

The letter by March et al. (2026) raises unintended consequences for students with diabetes. There is no abstract and no clinical finding. As an inference, not as a result, it is reasonable to think that some of those students monitor glucose with the phone and that a ban without a medical exception obstructs a form of care. Moreno et al. (2026) do not study diabetes, but a reported violation was more likely where there was a health condition or a disability: some of what is called non-compliance may be a need that was not foreseen.

Livingstone (2026) argues, without data, for allowing the phone with rules that reduce harm and support learning. Santos de Souza Junior (2026) maintains that the problem is not the mobile phone and that the school owes students digital literacy. Paredes-Labra et al. (2025) show the other risk: a techno-pessimistic shock, banning from cultural anxiety rather than from evidence. Banning out of panic and allowing without rules are two ways of not designing.

Bar et al. (2025) collect three unintended effects: less independence and less confidence, loss of digital tools, and difficulty regulating distress without the phone. In the same material there are benefits of face-to-face contact, health, safety and engagement, and a call for education in responsible use, with flexibility, set against a ban without nuance. The authors underline dependence on the device for socialising and for regulating emotions.

Heterogeneity is an argument about equity. Abrahamsson (2026) shows, beneath a null average, less use of psychological care among girls, improvements in grades and in trajectory among them, especially in disadvantaged contexts, and less bullying in both genders. Sungu et al. (2026) indicate a greater benefit for those who achieve less, and support also among those who achieve more. They are not the same subgroup, and the file does not authorise merging them. Looking only at the mean can conceal a distributive improvement, or a success that was not general.

10. Latin America and Mexico: new laws, scarce evidence

There is one Latin American study and none from Mexico. Santos de Souza Junior (2026) analyses Brazilian Law 15.100/2025 from applied linguistics, with digital literacy and legal science. The study estimates neither learning, nor bullying, nor health. It values the authorising character of the law and the articulation among the Union, the states and the municipalities, and it maintains that the problem is not the mobile phone: the school owes digital literacy because the school is not outside the concrete world.

The absence of impact evaluations is a limit. These 24 works do not say whether Galicia, the event study, the Danish recess or the English schools would be repeated in a Mexican secondary school, nor the reverse. Importing the slogan would repeat the techno-pessimistic shock described by Paredes-Labra et al. (2025). That the topic is present in Mexican public debate is asserted only at that level. There is no law, no programme and no national figure here.

As a design inference, and not as a finding, five implications follow. Medical and accessibility exceptions should be defined first, so that a form of monitoring, or a need related to disability, is not recorded as a fault: March et al. (2026) raise diabetes without data, and Moreno et al. (2026) link reported violations to health. Families and the Consejo Técnico Escolar should be involved in the rule and in daytime contact, because frequent contact is associated with more violations. Classroom and recess should be distinguished: collection failed in Grigic Magnusson et al. (2023), and recess changed physical activity, with nuances, in Pawlowski et al. (2021). Peer relations and learning should be evaluated before and after, without treating the rule as sound merely because a document exists; Sullivan et al. (2026) associate the policy with more recorded incidents, which may be monitoring. Digital education should not be replaced: it is a duty in Santos de Souza Junior (2026), a horizon in Böttger and Zierer (2024) and a position in Livingstone (2026).

11. Design criteria for a school phone policy

These nine criteria are a design inference. They fix use, moment, exception, compliance and the destination of the time that is recovered; they do not translate a meta-analysis into a school regulation.

  1. Name what is restricted and when. Campbell et al. (2024) and Sullivan et al. (2026) show that a partial ban, a complete ban and the rhetoric of a ban do not coincide.
  2. Separate recreational use from pedagogical use that has already been consented to. Grigic Magnusson et al. (2023) saw the phone as a tool while collection was expelling it from the room.
  3. Do not ground the rule in collecting devices if the procedure consumes the lesson and damages the relationship. In that study the strategy failed.
  4. Expect less use during the school day, not across the whole day (Goodyear et al., 2025): mental wellbeing did not improve either.
  5. Do not confuse restriction of access with problematic use. King et al. (2024) estimated the latter at 2.6%, with no difference between schools with a ban and schools without one.
  6. Write exceptions for medical needs, for accessibility and for family communication. The inference drawn from March et al. (2026) covers glucose; Moreno et al. (2026) link reported violation to health and to contact from parents.
  7. Decide the destination of the time that is recovered. At recess, Pawlowski et al. (2021) find more moderate activity and less vigorous activity, and context decides the sign (Goodyear et al., 2026).
  8. Look at equity in both directions: improvements among disadvantaged girls and a null average (Abrahamsson, 2026), a greater benefit for those who achieve less (Sungu et al., 2026) and a loss of emotional regulation (Bar et al., 2025).
  9. Evaluate peer relations, learning, wellbeing and incidents separately, before and after, and publish the definition. More records may be better monitoring (Sullivan et al., 2026). The loneliness that rises in Vereecke et al. (2026) is not cancelled by the bullying reported in another country. Without randomised trials (Campbell et al., 2024), the school produces its own evidence.

12. Limitations

Crossref was the base, and the complementary search covered evaluations that the seed did not bring; the 11 records were verified one by one and do not form a census. Lexical screening with screen.py plus manual selection, without double coding, may have omitted studies and over-weighted those already cited: 286 eligible and 24 included. There is no systematic review of those 286, and no meta-analysis of our own.

Three sources from 2026 enter by title only, and Sungu et al. (2026) enters with a single sentence. The massive experiment is not simulated, nor is the direction of effects for implementation in South Australia, for the letter on diabetes, or for the prospective study of social media and mental health.

Europe, Australia and the United States predominate. There is one Brazilian discourse analysis and no study from Mexico. The designs are above all observational or quasi-experimental — synthetic control, difference-in-differences, event study — together with qualitative work and few experiments declared as such. Campbell et al. (2024) find no randomised trials among 22 studies. The time horizons are short: one month (King et al., 2024), four weeks (Pawlowski et al., 2021) or one school year in a single school (Vereecke et al., 2026). Table 1 shows heterogeneous definitions of a ban. The d = 0.162 of Böttger and Zierer (2024) is not updated with 2025 or with 2026.

Publication bias is possible, although several null results — wellbeing, engagement, problematic use — keep the file from being made only of successes. The evidence from 2026 is still moving. Parental report, the incident obtained by an information request, the open-ended response and regional PISA do not measure the same policy.

13. Conclusions

The ban ran ahead of the evidence. Montag and Elhai (2023) note decisions already taken in France and China; Paredes-Labra et al. (2025), a techno-pessimistic shock in Madrid; Santos de Souza Junior (2026), a law whose stated problem is not the mobile phone; Sullivan et al. (2026), English guidance without a mandate; Moreno et al. (2026), rules already recognised by almost all the parents in their sample. Livingstone (2026) argues for allowing the phone with rules that reduce harm. The outcomes do not move together.

Where the restriction holds, what is most stable is climate, attention and peer relations, including less bullying in the quasi-experiments. According to Sungu et al. (2026), those who benefit above all are those who achieve less, and Grigic Magnusson et al. (2023) show that collection can strain the classroom. Achievement remains unequal: PISA is set out in detail only for Galicia, there is a null average with improvements among disadvantaged girls (Abrahamsson, 2026), and d = 0.162 in five studies (Böttger and Zierer, 2024).

Mental health does not follow the label. There is less use during the English school day and not outside it, minimal differences in quality of life, a probability of cost-effectiveness alongside intervals that cross zero, and a loneliness that rises modestly after the total ban. Problematic use does not distinguish the schools that have a rule. Banning the device and attending to problematic smartphone use are not the same intervention.

What matters is the design, the compliance, the exceptions and the destination of the time that is recovered. The chain in Figure 1 is lost in the unstable definition, in divided support, in costly collection, in unequal compliance, in recess without an alternative and in family communication that has not been renegotiated. In Latin America there is only an analysis of a Brazilian law; in Mexico, there is no evaluation. The task is to specify the rule, to protect exceptions for health and for accessibility, not to replace digital education, and to measure before and after. The label neither teaches nor cares for anyone; if it is complied with, it returns a margin in which to do both by other means.

NEXTECH.IA Editorial Lab / Ingeniero Mitre

References

  1. Abrahamsson, S. (2026). Smartphone Bans, Student Outcomes and Mental Health. Journal of Human Resources. https://doi.org/10.3368/jhr.0224-13403r2
  2. Baggio, S., Wade, T., Radunz, M., Galanis, C. R., Billieux, J., Starcevic, V., Quinney, B., & King, D. L. (2026). Implementation of phone ban policies in school: A secondary analysis of a phone ban intervention in South Australia. Technology, Mind, and Behavior. https://doi.org/10.1037/tmb0000237
  3. Bar, E., Radunz, M., Galanis, C. R., Quinney, B., Wade, T. D., & King, D. L. (2025). Student perspectives on banning mobile phones in South Australian secondary schools: A large-scale qualitative analysis. Computers in Human Behavior, 167, 108603. https://doi.org/10.1016/j.chb.2025.108603
  4. Beneito, P., & Vicente-Chirivella, Ó. (2022). Banning mobile phones in schools: evidence from regional-level policies in Spain. Applied Economic Analysis, 30(90), 153–175. https://doi.org/10.1108/aea-05-2021-0112
  5. Böttger, T., & Zierer, K. (2024). To Ban or Not to Ban? A Rapid Review on the Impact of Smartphone Bans in Schools on Social Well-Being and Academic Performance. Education Sciences, 14(8), 906. https://doi.org/10.3390/educsci14080906
  6. Campbell, M., Edwards, E. J., Pennell, D., Poed, S., Lister, V., Gillett-Swan, J., Kelly, A., Zec, D., & Nguyen, T.-A. (2024). Evidence for and against banning mobile phones in schools: A scoping review. Journal of Psychologists and Counsellors in Schools, 34(3), 242–265. https://doi.org/10.1177/20556365241270394
  7. Goodyear, V. A., Randhawa, A., Adab, P., Al-Janabi, H., Fenton, S., Jones, K., Michail, M., Morrison, B., Patterson, P., Quinlan, J., Sitch, A., Twardochleb, R., Wade, M., & Pallan, M. (2025). School phone policies and their association with mental wellbeing, phone use, and social media use (SMART Schools): a cross-sectional observational study. The Lancet Regional Health - Europe, 51, 101211. https://doi.org/10.1016/j.lanepe.2025.101211
  8. Goodyear, V. A., Randhawa, A., Adab, P., Al-Janabi, H., Fenton, S., Michail, M., Patterson, P., Sitch, A., Wade, M., & Pallan, M. (2026). How school phone policies influence adolescent phone use and wellbeing (SMART Schools): a qualitative comparative case study. Social Science & Medicine, 398, 119094. https://doi.org/10.1016/j.socscimed.2026.119094
  9. Grigic Magnusson, A., Ott, T., Hård af Segerstad, Y., & Sofkova Hashemi, S. (2023). Complexities of Managing a Mobile Phone Ban in the Digitalized Schools’ Classroom. Computers in the Schools, 40(3), 303–323. https://doi.org/10.1080/07380569.2023.2211062
  10. King, D. L., Radunz, M., Galanis, C. R., Quinney, B., & Wade, T. (2024). “Phones off while school's on”: Evaluating problematic phone use and the social, wellbeing, and academic effects of banning phones in schools. Journal of Behavioral Addictions, 13(4), 913–922. https://doi.org/10.1556/2006.2024.00058
  11. Livingstone, S. (2026). The Case Against School Cell Phone Bans. JAMA Pediatrics, 180(4), 362. https://doi.org/10.1001/jamapediatrics.2025.6428
  12. March, C. A., Libman, I., & Woodward, C. (2026). School Smartphone Bans—Unintended Consequences for Students With Diabetes? JAMA Pediatrics, 180(6), 700. https://doi.org/10.1001/jamapediatrics.2026.0706
  13. Montag, C., & Elhai, J. D. (2023). Do we need a digital school uniform? Arguments for and against a smartphone ban in schools. Societal Impacts, 1(1-2), 100002. https://doi.org/10.1016/j.socimp.2023.100002
  14. Moreno, M. A., Owings-Fonner, N., Salerno, A., Jenkins, M., Williams, E., Zhao, Q., & Eickhoff, J. C. (2026). Parent-Reported School Phone Policies: Prevalence, Components, and Violations. Journal of Adolescent Health, 78(4), 559–565. https://doi.org/10.1016/j.jadohealth.2025.12.267
  15. Paredes-Labra, J., Solana-Domínguez, I., Ramos-Ramiro, M., & Freitas-Cortina, A. (2025). Techno-Pessimistic Shock and the Banning of Mobile Phones in Secondary Schools: The Case of Madrid. Social Sciences, 14(7), 441. https://doi.org/10.3390/socsci14070441
  16. Pawlowski, C. S., Nielsen, J. V., & Schmidt, T. (2021). A Ban on Smartphone Usage during Recess Increased Children’s Physical Activity. International Journal of Environmental Research and Public Health, 18(4), 1907. https://doi.org/10.3390/ijerph18041907
  17. Perry, S. J., Goodyear, V. A., Pallan, M., Adab, P., Fenton, S., Michail, M., Patterson, P., Randhawa, A., Sitch, A. J., Wade, M., & Al-Janabi, H. (2026). Health economics analysis of restrictive school smartphone policies in secondary schools in England (SMART Schools). BMJ Mental Health, 29(1), e301892. https://doi.org/10.1136/bmjment-2025-301892
  18. Radunz, M., Quinney, B., Galanis, C., King, D. L., & Wade, T. D. (2026). Social Media Use and Mental Health of Adolescents: A Prospective Study of the South Australian School Mobile Phone Ban. International Journal of Mental Health and Addiction, 24(5), 4006–4022. https://doi.org/10.1007/s11469-025-01597-w
  19. Randhawa, A., Pallan, M., Twardochleb, R., Adab, P., Al-Janabi, H., Fenton, S., Jones, K., Michail, M., Patterson, P., Sitch, A., Wade, M., & Goodyear, V. A. (2025). Secondary school smartphone policies in England: a descriptive analysis of how schools rationalize, design, and implement restrictive and permissive phone policies. Journal of Research on Technology in Education, 57(5), 1113–1132. https://doi.org/10.1080/15391523.2024.2363204
  20. Santos de Souza Junior, A. (2026). A proibição do celular na sala de aula por meio da lei 15.100/2025: uma análise crítica. A Cor das Letras, 26(3), 221–233. https://doi.org/10.13102/cl.v26i3.11814
  21. Sheng, M., & Lipscombe, K. (2026). An Exploration of Mobile Phone Policies and Associated Leadership Implications for School Leaders: A Systematic Review and Document Analysis. Leadership and Policy in Schools, 25(3), 641–659. https://doi.org/10.1080/15700763.2024.2424524
  22. Sullivan, E., Dallison, R., Montgomery, P., & Henderson, L. (2026). Digital safeguarding incidents in schools with and without smartphone policies: a cross‐sectional study in England. Child and Adolescent Mental Health, 31(4), 294–304. https://doi.org/10.1111/camh.70079
  23. Sungu, A., Choudhury, P. K., & Bjerre-Nielsen, A. (2026). Phones impede learning: Evidence from a massive phone-free classroom experiment. Science, 394(6820), 178–181. https://doi.org/10.1126/science.aef7249
  24. Vereecke, J., Vanherle, R., Nassen, L.-M., Boen, H., Eggermont, S., Beullens, K., Vandenbosch, L., & Schreurs, L. (2026). More than disconnecting? School smartphone bans and adolescents' social wellbeing. Frontiers in Developmental Psychology, 4, Article 1796471. https://doi.org/10.3389/fdpys.2026.1796471