evidoria

← Back to browse

Good practice Imported

Eduten Adaptive Maths Platform — 12-Week AI Numeracy Pilot for Grade 3 Students in Lao PDR

Laos · Vientiane · See the Laos profile · See the Vientiane profile

Evidence: Quasi-experimental Top 59% 47/100 · Ask Evidence Copilot about this practice

A 12-week pilot of Finland's Eduten adaptive maths platform across 10 Lao PDR schools found Grade 3 participants gained 35.7% in math proficiency and 51.4% in arithmetic fluency versus a control group, per UNICEF's implementation report — though the assessed sample was smaller th

35.7 %
Math proficiency gain (treatment vs control) (12-week pilot, Dec 2023)
51.4 %
Arithmetic fluency gain (treatment vs control) (12-week pilot, Dec 2023)
384 students
Students completing pre- and post-assessment
Eduten Adaptive Maths Platform — 12-Week AI Numeracy Pilot for Grade 3 Students in Lao PDR Eduten Adaptive Maths Platform — 12-Week AI Numeracy Pilot for Grade 3 Students in Lao PDR

Details

Maturity
Pilot
Promoter
Lao PDR Ministry of Education and Sports (Research Institute for Education Sciences) / UNICEF Lao PDR / Eduten Ltd (Finland)
Period
12-week pilot, announced Dec 2023; implementation report published June 2025
Keywords
adaptive learning, personalised maths, edtech pilot, teacher training, primary education

Context

Eduten Ltd, a Finnish edtech company, partnered with Lao PDR's Ministry of Education and Sports (Research Institute for Education Sciences) and UNICEF to pilot an AI-adaptive mathematics platform in Grade 3 classrooms. The 12-week pilot ran in 10 schools starting December 2023, with one teacher per school receiving a 3-day training session in Thalat, Vientiane Province. The platform draws on a bank of over 200,000 curriculum-aligned tasks and adjusts difficulty and learning paths to each student's performance.

Objectives

The pilot aimed to test whether AI-adaptive personalised learning paths could improve numeracy and arithmetic fluency among Grade 3 students compared with continuing standard instruction.

Activities

Up to 500 students used the Eduten adaptive maths app for 12 weeks while 300 students continued with traditional instruction as a comparison group. One teacher per school was trained by Eduten over 3 days before rollout.

Results

Among the 384 Grade 3 students who completed both pre- and post-assessments, the treatment group showed a 35.7% gain in math proficiency and a 51.4% gain in arithmetic fluency relative to the control group, according to UNICEF Innovation's June 2025 report.

Conclusions

Results are reported by UNICEF/Eduten rather than an independent evaluator, and the assessed sample (384) is smaller than the announced treatment group (500) without public explanation. No disaggregated equity data (gender, rural/urban, disability) or confirmed nationwide rollout has been published; Eduten cites larger gains from pilots in Mongolia/Lithuania and a national scale-up in Bhutan as comparison points.

Implementation
Indicative cost
Low (< €50k) — Costs not publicly disclosed; likely low given the limited school count (10) and short teacher-training footprint (3 days per school).
Time to results
Short (< 1 year) — 12-week classroom pilot beginning around the December 2023 announcement; outcomes published by UNICEF Innovation in June 2025.
Staffing & skills
One teacher per participating school (10 schools total), given 3-day Eduten-led training in Thalat, Vientiane Province, Eduten Ltd technical/training team, Ministry of Education and Sports Research Institute for Education Sciences oversight, UNICEF Innovation support

Conditions for success

  • Reliable teacher training ahead of rollout
  • Access to devices/connectivity for the adaptive app
  • A genuine, even if non-randomised, comparison group to isolate the platform's effect

Common failure modes

  • Unexplained gap between announced treatment sample (500) and assessed sample (384) creates risk of selection/attrition bias
  • Non-randomised group assignment limits causal claims
  • Results reported only by implementing partners, not independently verified

Where it fits

Governance type
national ministry-led with international NGO/edtech partner
Scale
10 schools, roughly 800 students
Income level
low-income (Lao PDR, LDC)

Commonly funded by

Philanthropic / foundation funding

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful