Reflections from the ADEA Triennale 2025, Accra
By João Pedro Azevedo — Chief Statistician & Deputy Director, UNICEF (Data, Analytics, Planning & Monitoring)
A Pan-African Forum for System Transformation
Every three years, the Association for the Development of Education in Africa (ADEA) convenes its Triennale on Education—the continent’s pre-eminent forum for political dialogue and peer learning. The 2025 edition in Accra, held under the theme:
“Strengthening the resilience of Africa’s educational systems: Advancing towards ending learning poverty by 2035 with a well-educated and skilled workforce for the continent and beyond,”
brought together ministers, policymakers, educators, and researchers to examine how evidence can translate into system-wide change.
Building on earlier Triennales in Ouagadougou (2012), Dakar (2017), and Mauritius (2022), this year’s gathering took place at a moment of continental ambition: the African Union’s “Year of Education 2024” and the launch of the Decade of Accelerated Education (2025–2034). It was also framed by new strategic compasses—the Continental Education Strategy for Africa (2025–2034), the CTVET Strategy, and the STISA 2034. The objective was clear: to celebrate progress, confront persistent bottlenecks, and forge a collective path toward resilient, evidence-driven systems.
From Dialogue to Data
Within this broader agenda, our session—“Enablers of Progress: Data Use and Assessments to Strengthen Systems and Drive Learning”—stood out for its mix of technical rigour and candour. Armando Ali (PAL Network) opened with humour and precision: “We are here to talk about the real classroom—and the evidence that changes it.”
Hon. Dr. Makgabo Reginah Mhaule, South Africa’s Deputy Minister for Basic Education, offered a concise roadmap for evidence-based reform:
- Stay in the arena. South Africa’s continued participation in TIMSS, PIRLS/PALS and SACMEQ has exposed uncomfortable truths while documenting genuine progress.
- Build a national spine. Systemic evaluations in Grades 3, 6 and 9, the “5 × 5 Index” tracking child development at age five, and an Early Learning National Assessment at Grade 1 entry provide continuity.
- Benchmark in every language. Reading benchmarks have been created in all South African languages, tailored to their linguistic and orthographic features.
- Track progress over time. The forthcoming Funda Uphumelele National Survey will baseline reading proficiency across provinces and languages for the next decade.
Her message was disarming in its simplicity:
“Even when results are uncomfortable, data must inform action.”
A Continental Conversation — In Two Languages
In a session that mirrored Africa’s linguistic and cultural diversity, Clio Dintilhac of the Gates Foundation moderated a genuinely bilingual dialogue—English and French—on continental perspectives in building fit-for-purpose, comparable, and usable foundational learning assessments.
She framed the discussion around three transformative currents shaping Africa’s learning landscape:
- More comparable data than ever — with PASEC, PAL, and MICS Learning generating new cycles of high-quality, cross-country data.
- Better use of existing evidence — through harmonisation, methodological innovation, and the strategic linking of datasets already available.
- Closer alignment with national priorities — including contextual fluency thresholds, assessment planning, responsible AI integration, and language policy tailored to local realities.
Together, these themes underscored a broader continental vision: one where data are not only collected but actively used to strengthen systems and inform policy decisions across Africa.
From Measurement to Momentum
Dr. Hilaire Hounkpodoté (PASEC) described how the programme now covers 21 countries, linking multi-grade assessments (Grades 2, 6, 9) with teacher and household data, while training national teams in psychometrics and analysis.
Dr. Manos Antoninis (UNESCO GEM Report) added perspective: only 14 per cent of African pupils reach minimum proficiency by the end of primary school, and only one in five countries has a national assessment framework. The new Continental Assessment Framework (CAF)—a joint AU–Association for Educational Assessment initiative—offers a bridge between national and global standards, anchored in SDG 4.1.1.
In UNICEF’s contribution, I argued that MICS Foundational Learning Module (FLS) complements school-based assessments by capturing the drivers of learning—nutrition, environment, and socio-economic context. Without this lens, we risk knowing how much children learn but not why.
UNICEF’s Three Pillars for Learning Data Ecosystems
UNICEF’s work builds directly on this momentum through three connected pillars:
1. Data Generation. The MICS Foundational Learning Module has been implemented in 55 countries, including 16 in Africa, and is now being revised as FLM 2.0 to align with the Global Proficiency Framework (GPF). Partnerships with governments are strengthening national assessments and ensuring coherence between household and school data.
2. Data Transparency and Use. A new Learning & Skills Data Lab will bring together national learning and skills data, promote applied research, and expand the use of visual analytics for decision-making.
3. Monitoring and Action. The Foundational Learning Action Tracker 2025 monitors commitments-to-action in 124 countries, categorising progress as Championing, Established, Initiating, or Not Yet Initiating. It converts political pledges into shared accountability.
Together, these pillars form the backbone of Africa’s emerging learning data ecosystem—from generation to use to action.
Reading the Evidence
The chart below, based on MICS Foundational Learning Module data, plots the share of students able to read with comprehension (vertical axis) against grade level (Grades 1–9).
Notes: Blue lines: African countries in MICS-FLS; Grey lines: Non-African comparators; and, Orange line: The African median.
The pattern is unmistakable. By Grade 3, fewer than one in five African children can read with comprehension. Even by Grade 9, most countries remain below 80 per cent proficiency. Learning is taking place—but too slowly and too late. The learning gradients reveal striking variation: some countries achieve near-universal literacy by Grade 6, while others remain below 30 per cent even at Grade 8.
The message is twofold. First, policy choices and instructional quality matter profoundly—especially in how teachers are supported, languages are chosen, and early learning is prioritised. Second, the diversity of these outcomes shows that progress is possible, and that many of the solutions already exist within the continent. Countries with steeper gradients are proving that the right mix of curriculum alignment, language policy, and teacher coaching can bend the curve of learning.
African children are learning—but grade-levels later than they should.
Africa Must Learn from Africa
Africa’s strength lies not only in its demographics but in its diversity. With hundreds of languages and education systems operating across varied cultural and governance contexts, no other region offers such a living laboratory for innovation. The challenge is not to import models but to connect and scale what already works.
The continent’s demographics—its youth, its energy, its potential—make learning outcomes a generational imperative. But its diversity is its greatest comparative advantage. From Senegal’s community-based early-grade reading programmes, to Kenya’s national assessment reforms, South Africa’s multilingual literacy benchmarks, and Ghana’s data-driven school leadership, Africa already holds many of the ingredients for systemic transformation.
These experiences illustrate that effective reform is not bound by income level or geography—it is driven by coherence, capacity, and commitment. The next frontier is to build the networks that allow countries to learn from one another, systematically and at scale.
Africa must learn from Africa. Its future will be written in its own classrooms, not borrowed from elsewhere.
From Assessments to Ownership
Prof. Cally Ardington, Director of DataFirst, drew on discussions from the AFLAI/NAPLI Convening in Nairobi to highlight five priorities that resonated strongly throughout the Triennale:
- Daily use and communication of learning data — ensuring results reach teachers, policymakers, and communities in ways that drive action.
- Locally valid benchmarks — grounded in country contexts yet anchored to SDG 4.1.1 standards for minimum proficiency.
- National assessment planning and capacity building — shifting from externally driven technical assistance to sustainable, home-grown systems.
- Responsible integration of AI — using technology to enhance, not replace, professional judgment in assessments.
- Language and equity frameworks — balancing linguistic diversity with inclusivity and political feasibility.
Across these priorities ran a single, unifying thread: ownership. Countries want institutions, not experiments—systems that last beyond projects, capacity that endures beyond consultants, and evidence that belongs to those who generate and use it.
The Path Forward
The 2025 Triennale confirmed that Africa’s education renewal is no longer about declarations—it is about design. Data, used wisely, can be both a mirror and a map: revealing where systems stand and guiding how they can improve.
Africa’s education transformation will be data-driven, locally led, and globally recognised.
The challenge now is not collecting more numbers but ensuring they are trusted, comparable, and used—by teachers, planners, and policymakers—to deliver learning for every child.
#ADEATriennale2025 #UNICEF #FoundationalLearning #DataForLearning #MICS #FLM2 #GPF #PASEC #GEMReport #PALNetwork #AFLAI #NAPLI #EvidenceForPolicy #EducationInAfrica #LearningPoverty
