evidoria

← Back to browse

Good practice Imported

Rori — AI Math Tutor on WhatsApp Reaches Liberian Classrooms

Liberia · Monrovia · See the Liberia profile · See the Monrovia profile

Evidence: Randomised controlled trial Top 70% 40/100 · Ask Evidence Copilot about this practice

Rising Academies' Rori delivers AI-personalised math tutoring over basic WhatsApp, reaching Liberian schools among 90+ across six African countries. A Ghana RCT (1,000 students) found a 0.37 effect size (p<0.001); Liberia outcomes are not yet independently evaluated.

1,000 students
Students evaluated in Ghana study (2024 (Year 1))
11 schools
Schools in Ghana study (2024)
0.37 effect size (SD)
Effect size on math learning vs control (2024)
p<0.001 p-value
Statistical significance (2024)
90+ schools
Schools reached across Africa (2023-present)
100,000+ students
Students reached across Africa (2023-present)
1,000,000+ users
Target users by 2027 (target: 2027)
Rori — AI Math Tutor on WhatsApp Reaches Liberian Classrooms

Details

Maturity
Scaling
Promoter
Rising Academies
Period
2023–present
Keywords
edtech, mathematics, WhatsApp, tutoring, non-profit

Context

Rori is a conversational AI math tutor delivered entirely over WhatsApp, built by the Sierra Leone-founded non-profit Rising Academies. It runs a short diagnostic to place each learner, then serves scaffolded practice questions and worked explanations paced to that learner's level — requiring only a basic phone and a WhatsApp connection, not a dedicated app or heavy data plan. Originally built and tested in Ghana and Sierra Leone, Rori has since expanded to Kenya, Rwanda, Liberia and The Gambia.

Activities

In the evaluated Ghana deployment, students received two 30-minute Rori sessions per week over 8 months, added to normal instruction, with schools randomly assigned to treatment or control.

Results

Rori operates in 90+ schools and reportedly reaches over 100,000 students across Africa, with Rising Academies targeting 1 million+ users by 2027 — a forward-looking goal, not yet an achieved figure. The 2024 Ghana study found an effect size of 0.37 on math learning versus a control group, statistically significant at p<0.001, though the authors caution this covers only Year 1 of the intervention and no equivalent independent evaluation has yet been published for the Liberia or Gambia deployments specifically.

Conclusions

Rising Academies' own materials state that girls in its programmes progress at roughly twice the rate of their peers. Evidoria could not independently verify this specific figure outside the organisation's own site, so it is presented here as an internal claim rather than confirmed evidence.

Implementation
Indicative cost
Low (< €50k) — Not disclosed for Liberia specifically; delivered via WhatsApp requiring minimal hardware (basic phone) and low data usage, suggesting low per-student marginal cost relative to device-based EdTech.
Time to results
Medium (1–3 years) — Piloted in Ghana/Sierra Leone from 2023; expanded to Kenya, Rwanda, Liberia and The Gambia; targeting 1 million+ users by 2027.
Staffing & skills
Delivered via WhatsApp chatbot, minimal in-classroom staffing beyond normal instruction, Non-profit (Rising Academies) manages content, tutoring logic and country expansion

Conditions for success

  • Only requires a basic phone and WhatsApp connection, not a dedicated app or heavy data plan — well suited to low-bandwidth contexts
  • Added to normal instruction rather than replacing it, per the evaluated Ghana model (two 30-minute sessions/week)

Common failure modes

  • Independent RCT evidence exists only for Ghana Year 1; no equivalent independent evaluation published yet for Liberia or Gambia deployments
  • Rising Academies' internal claims (e.g., girls progressing twice as fast) are not independently verified

Where it fits

Governance type
NGO-run, multi-country programme partnering with schools/ministries
Scale
90+ schools, 100,000+ students across 5 countries
Income level
low-income / low-resource settings across sub-Saharan Africa

Commonly funded by

Philanthropic / foundation funding

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

Replication kit

Reusable artefacts from this practice — as published by their sources.

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