School Digital Equipment Rarely Improves Grades — DEPP's 2026 Data Shows Structured Use Does

If your school (or your ministry, or your district) has spent the last few years buying tablets, interactive whiteboards, or laptop carts, France’s national education statistics office just published data you should read before the next equipment order goes out. The Direction de l’évaluation, de la prospective et de la performance (DEPP) released an updated 2026 synthesis of what the country’s digital-equipment spending has actually done for student outcomes — and the headline is uncomfortable for anyone who assumed that buying devices is itself the intervention. According to the DEPP synthesis, equipment alone rarely moves the needle. Structured, individually-assigned use — backed by trained teachers — does.

For school administrators across MENA and Europe currently weighing tablet programs, 1:1 laptop initiatives, or digital-learning budgets, this isn’t a France-only curiosity. It’s a rare case of a government running the numbers with genuinely causal methods and publishing what it found.

What DEPP actually measured

The 2026 synthesis distinguishes between two very different types of digital-equipment policy, and the distinction turns out to matter enormously. The first is “équipement individuel mobile” (EIM) — tablets or laptops assigned to individual students, generally paired with a structured pedagogical program. The second is “classe mobile” (CM) — shared equipment carts that circulate between classrooms, available but not tied to any one student or consistent instructional routine.

To isolate the effect of each, DEPP’s researchers used entropy balancing and other statistical matching methods to compare beneficiary students against a non-beneficiary comparison group with similar characteristics — explicitly, in the DEPP’s own words, “afin d’identifier l’impact causal des dotations” (to identify the causal impact of the grants), according to the DEPP synthesis. That’s a meaningfully stronger evidence base than the usual “we bought devices and enrollment went up” reporting that circulates around edtech spending.

The results split cleanly along the individual-vs-shared line. At collège level (roughly grades 7-9), individually-assigned tablets tied to structured use produced positive effects on French and math scores ranging from 9% to 25% of a standard deviation relative to the comparison group — effects concentrated among students from lower socioeconomic backgrounds. Shared classroom-cart equipment, by contrast, showed effects “ranging from zero to 11%” of a standard deviation — weak to essentially null, per the same DEPP synthesis.

At elementary level, the picture is starker. The 2015 national “Plan numérique” rollout — this time evaluated for the elementary cycle 3 (CM1-CM2) cohort — showed no notable causal effect on French or math performance by the time students entered collège. DEPP describes this as consistent with the existing international research literature on educational technology (citing Escueta et al. 2020), not an anomaly.

Individually-assigned tablets helped — structure correlates with the difference

It would be easy to read the collège finding as “tablets work, carts don’t,” but that’s not quite what the data shows. EdTechActu’s summary and Localtis’s policy-focused analysis both frame the gap the same way: it isn’t the hardware category that predicts the outcome, it’s whether the equipment is embedded in a structured routine teachers know how to run — individual assignment appears to be a proxy for that structure, not the mechanism itself.

This pattern isn’t new — it echoes an earlier ministry evaluation of the same 2015 Plan numérique — which equipped over 175,000 students across 1,510 schools — known as “Élaine,” which found grade 7 oral French comprehension improved by 31% of a standard deviation (described as roughly equivalent to advancing three class ranks), with grade 8 effects of 8-14% of a standard deviation, again strongest in disadvantaged schools, according to Localtis’s 2021 reporting. The 2026 DEPP synthesis’s causal figures — which do test individually-assigned versus shared equipment directly — land in almost the same range, five years later, using a more rigorous quasi-experimental design.

The training gap

Here is where the training gap correlates with the split in outcomes — and where the DEPP synthesis gets specific about the mechanism. Across France, only 26% of collège teachers and just 14% of elementary teachers say their initial training left them feeling “assez” or “beaucoup” prepared to use digital tools and resources for teaching — the lowest rates recorded in international comparison, according to the DEPP synthesis. These are self-reported survey figures, not an experimental measurement, so they should be read as describing the scale of the preparedness gap rather than proving it single-handedly causes the outcome gap between EIM and CM programs. But the pattern lines up: elementary schools, where training-preparedness is lowest (14%) and equipment is overwhelmingly shared-cart style, show no measurable effect; collège programs with individually-assigned devices and more structured deployment show real gains.

School leaders themselves point to the same bottleneck. Per the DEPP synthesis, elementary school directors report needing “more equipment and more teacher training” to improve pedagogical management of digital tools, and teachers identify insufficient training and insufficient equipment quantity as the two main barriers to pedagogical use — not lack of interest or lack of devices in isolation.

The equipment gap compounds the training gap. As of 2024, secondary schools have four times the device density of primary schools — 54 fixed or mobile terminals per 100 students versus 14 per 100 in primary — per the DEPP synthesis and corroborated by EdTechActu. Primary schools are simultaneously the least equipped and the least trained — which may help explain why they’re also the segment showing no measurable academic gain from digital investment so far.

Training is not the only factor

DEPP’s own internal contrast — individually-assigned tablets with structured use outperforming shared carts — suggests training interacts with deployment design, not training in isolation. International evidence points the same direction: OECD Education Working Paper No. 340 finds that across PISA-participating countries, organized teacher professional development combined with deliberate curriculum integration outperforms cases where “technology deployment occurs in isolation” — curriculum design, not just teacher confidence, is part of the equation. Schools that want to close this gap would also need to look at device-management routines, family communication about appropriate use, and consistent access at home versus school — DEPP didn’t measure these directly, but they plausibly interact with the training and structure variables it did. Training is the variable DEPP measured most directly, but it’s one lever among several likely to determine whether an equipment purchase turns into a learning gain.

A 14-year precedent: Landes

France has one directly relevant historical precedent — a 14-year-old warning that DEPP’s 2026 data just confirmed with causal rigor. Between 2001 and roughly 2012, the Landes département ran “Un collégien, un ordinateur portable,” a multi-year 1:1 laptop program — eventually equipping around 51,000 collégiens at a cost of €52 million, €30 million of it on hardware alone. A 2012 inspectorate report (Inspection générale de l’éducation nationale) examined the program directly and concluded plainly: “il n’est pas possible d’établir un lien… entre équipement des collèges en outils et contenus numériques, d’une part, et performance des élèves, d’autre part” — it wasn’t possible to establish a link between equipment and student performance, according to the IGEN report. Notably, the report found the program was rarely even cited in school improvement plans as a lever for raising achievement.

This is a 14-year-old qualitative report, not current statistical evidence, and it explicitly disclaimed any causal claim of its own — it’s included here as a historical illustration, not as proof in itself. But the mechanism it describes lines up with what DEPP later measured causally: inspectors found some teachers went “tout numérique” (fully digital) in their subjects while others made no use of the laptops at all, “par manque de pratique, par inexpérience ou par absence de familiarité avec les outils numériques” — for lack of practice, inexperience, or unfamiliarity with the tools. A well-funded, decade-long device rollout with uneven training produced uneven, unmeasurable results. Fourteen years later, DEPP’s quasi-experimental data shows the same split nationally, this time with the statistical rigor to call it causal.

What this means for administrators budgeting digital equipment now

For school leaders in France, Belgium, Morocco, the UAE, Saudi Arabia, and across the Gulf weighing device purchases or renewing digital-learning budgets, the practical takeaway isn’t “don’t buy tablets.” It’s that the purchase order is the smaller decision. The larger one is whether the rollout includes individual assignment, a structured use plan, and teacher training that teachers themselves rate as adequate — and whether families are looped into how the devices are meant to be used at home.

A few concrete ways schools could translate “structured use” into something operational, consistent with what DEPP and the OECD paper describe as the differentiating factor:

A defined onboarding sequence for teachers before devices reach students — not a single launch-day session, but two or three short follow-up trainings scheduled across the first term, since one-off training is exactly what the 26%/14% preparedness figures suggest isn’t working.
A weekly two-line update to parents — sent by app or SMS — naming which app or task the tablet was used for that week, so families reinforce the same structured routine at home rather than treating the device as unsupervised screen time.
A per-classroom usage log reviewed monthly by the pedagogical lead, flagging teachers who report low or no digital use so training support can be targeted rather than generic.

Structured use is the deployment decision, and it needs a communication layer

DEPP’s data is specific to France, but the underlying mechanism — equipment without structured use and trained teachers doesn’t move outcomes, while individually-assigned devices with defined routines and family visibility do — travels well beyond one country’s ministry statistics. In practice, “structured use” is only as strong as the coordination behind it: teachers following a consistent plan, school leaders monitoring adoption, and families kept in the loop on how the devices are actually used. That coordination layer is where a platform like BeeNet fits as one implementation path — giving school leaders the messaging and reporting tools to track digital-tool rollout and keep parents and teachers aligned, without needing a DEPP-scale research team to know whether the investment is landing. Schools evaluating whether their communication setup can support that kind of structured rollout can book a demo to see it in practice.

The next tablet or laptop budget line doesn’t need to repeat Landes’ or the 2015 Plan numérique’s ambiguity. The evidence for what makes digital equipment pay off now exists, and it’s specific enough to act on. DEPP’s data makes a strong case that structured deployment and teacher training matter more than the hardware. The real question for school leaders is when their next equipment rollout starts building that structure in from day one, not after it.

References

Charpentier, A. / DEPP, French Ministry of Education. “Numérique éducatif : que nous apprennent les données de la DEPP ?” Synthèse de la DEPP n° 3, August 2021, updated May 2026. https://www.education.gouv.fr/sites/default/files/document/num-rique-ducatif-que-nous-apprennent-les-donn-es-de-la-depp—516644.pdf
EdTechActu. “Numérique à l’école : les enseignements clés du rapport de la Depp.” June 11, 2026. https://edtechactu.com/evaluation/numerique-a-lecole-cinq-enseignements-cles-du-rapport-de-la-depp/
Martini, P. / Académie de Nice CPE site. “Numérique éducatif, que nous apprennent les données de la DEPP ?” May 28, 2026. https://www.pedagogie.ac-nice.fr/cpe/2026/05/28/numerique-educatif-que-nous-apprennent-les-donnees-de-la-depp/
Lesay, J.D. / Localtis, Banque des Territoires. “Numérique éducatif : des enseignants peu formés, de maigres effets.” May 29, 2026. https://www.banquedesterritoires.fr/numerique-educatif-des-enseignants-peu-formes-de-maigres-effets
Lesay, J.D. / Localtis, Banque des Territoires. “Les tablettes numériques ont un effet positif sur les collégiens.” February 4, 2021. https://www.banquedesterritoires.fr/les-tablettes-numeriques-ont-un-effet-positif-sur-les-collegiens
Durpaire, J-L., Jouault, D., Lhérété, A., Perez, M. / Inspection générale de l’éducation nationale. “Le plan « Un collégien, un ordinateur portable » dans le département des Landes.” Rapport n° 2012-148. https://www.education.gouv.fr/sites/default/files/document/le-plan-un-coll-gien-un-ordinateur-portable-dans-le-d-partement-des-landes-255066.pdf
Boeskels, L., Echazarra, A. / OECD Publishing. “Using Digital Resources for Learning.” OECD Education Working Paper No. 340, November 2025. https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/12/using-digital-resources-for-learning_4594bfc4/eb9453f3-en.pdf