Syntes: Allcott et al. (2026) — locking phones away in pouches: average test-score effect close to zero (working paper)

Källa: Allcott, H., Baron, E. J., Dee, T., Duckworth, A. L., Gentzkow, M., & Jacob, B. (2026). The Effects of School Phone Bans: National Evidence from Lockable Pouches. NBER Working Paper 35132. https://doi.org/10.3386/w35132 — preprint/working paper, not peer reviewed when used (2026-10-07).

Verified numbers (full author PDF, August 2026 version, read 2026-10-07)

• Design: preregistered (OSF pre-analysis plan) staggered difference-in-differences; schools adopting Yondr lockable pouches in 2023, 2024 or 2025 vs never-treated schools in the same states, with baseline covariates; pandemic years excluded. Pooled test-score sample: 1,778 adopters vs 42,919 never-adopters (Table 1); 214,556 school-year observations for the combined score (Table 3).
• Contrast: before adoption about 62 % of adopters (teacher survey) had a "no-show" rule (phones kept on the pupil, not to be displayed) and 23 % no formal policy; teacher-reported share of pupils using phones in class for personal reasons fell from 61 % to 13 % after adoption. GPS phone activity during school hours fell (about 0.18 log points on average; roughly 30 % by the third year).
• Test scores (Table 3, school-level SD): pooled combined ATT −0.006 (SE 0.010), maths −0.002, ELA −0.006 — about −0.003 student-level SD; "we can rule out improvements larger than approximately 0.007 student-level standard deviations". High schools: combined +0.026 (SE 0.017), maths +0.047 (SE 0.018, p < .05; ≈ 0.024 student-level SD, ≈ 0.9 percentile points), ELA +0.012. Middle schools: combined −0.025 (SE 0.012, p < .05), maths −0.026 (SE 0.013, p < .05), ELA −0.019 (ns).
• Robust to doubly robust estimator, refined control groups (no strict policy changes, no policy change in five years, excluding bell-to-bell-ban states), balanced panels, district clustering. Attendance: precise null. Discipline up ~15 % in the adoption year, fading. Wellbeing: down in year of adoption, positive by year 2 (suggestive).
• Limits stated by the authors: at most three years after adoption; school-level outcomes; pouches as implemented (incomplete enforcement); other strict approaches may differ.

Role for C6

Limits (weaken, 2026-10-07): a large preregistered DiD in which a binding restriction of pupils' own phones during the school day (vs mostly no-show or no rules) left average test scores essentially unchanged, with small gains in high-school maths and small losses in middle schools. Together with Kessel et al. 2020 (Sweden, null) C6's unconditioned average effect is narrowed to "found in some settings and groups; average not established". Not a kill: Beland & Murphy 2016 (England) and the gains the authors report from Florida (Figlio & Özek 2025), Rio de Janeiro (Lichand et al. 2026), Spain (Beneito & Vicente-Chirivella 2022) and Norwegian girls (Abrahamsson) remain, and high-school maths rose in this study. Preprint: recheck when published.