The first World Development Report devoted entirely to education, and the document that put "the learning crisis" into mainstream policy language. Its core claim — that schooling without learning is a wasted opportunity and an injustice — reframed the field from access to outcomes. Read the overview chapter first; the three policy actions (assess learning, act on evidence, align actors) structure most of what follows.
Deep Dive · Education
The Learning Crisis
Children are in school as never before — and learning far less than the world assumed. A reading list on the gap between schooling and learning.
Foundational Literacy & Numeracy
Learning Poverty
14 readings
IM
ImpactMojo Editorial
Curated by the ImpactMojo team
This is the reading list we reach for when explaining why "more schooling" stopped being a sufficient development goal. It pairs the global diagnosis — the World Bank's learning crisis framing — with the South Asian evidence base that has shaped it more than any other region: ASER, Pratham, and a generation of randomised work on what actually raises learning. We're looking for an invited curator from education research or practice; pitches welcome.
House Pick
Editor's Note
For most of the development era, schooling was treated as a proxy for learning: get children into classrooms, and human capital would follow. The data broke that assumption. India's Annual Status of Education Report (ASER) — a citizen-led household survey run by Pratham since 2005 — showed year after year that millions of children completing primary grades could not read a simple Standard-2 paragraph or do basic arithmetic. The World Bank later gave the phenomenon a global name and number: learning poverty, the share of 10-year-olds unable to read and understand a short, age-appropriate text. In low- and middle-income countries that share sits around 53 percent, and far higher in the poorest economies.
The 2018 World Development Report — the Bank's first ever devoted entirely to education — made the framing official, warning that "schooling is not the same as learning" and that schooling without learning is "a wasted opportunity" and "a great injustice." Lant Pritchett had argued the same point a few years earlier with a blunter title: schooling ain't learning. The diagnosis reframed the problem from access to systems: enrolment had soared while the machinery of teaching, assessment, and accountability had not kept pace.
What works is, encouragingly, not a mystery. The most robustly evaluated answer comes from South Asia itself — Pratham's Teaching at the Right Level (TaRL), which groups children by current ability rather than grade, validated across multiple J-PAL randomised trials. India has since written foundational literacy and numeracy into national policy through NEP 2020 and the NIPUN Bharat mission, which targets FLN for every child by the end of Grade 3. This list moves from diagnosis (what the crisis is and how we measure it) to evidence (what raises learning) to the policy and systems response. Start with the WDR 2018 overview and the ASER findings; then read Pritchett and the RISE systems work for the harder argument about why fixing this is so difficult.
Section 01
Diagnosing the Crisis
The texts that named the gap between schooling and learning and turned it into a measurable, global problem.
Pritchett's polemic anticipated the WDR framing and remains its sharpest version. His concept of "isomorphic mimicry" — systems that copy the outward forms of successful schooling without the underlying function — explains why building more schools rarely produced more learning. A bracing argument for locally evolved, accountable systems over centralised top-down design.
The methodological home of the headline number: the share of 10-year-olds who cannot read and understand a simple text, combining learning shortfalls with out-of-school children. The brief explains how the indicator is built and why a single composite measure was needed to hold the world's attention. Pair with the 2022 global update for the post-pandemic picture.
Section 02
Measuring Learning in South Asia
The assessments and surveys — most of them pioneered in India — that gave the crisis its evidence base.
The largest citizen-led survey of children's schooling and learning anywhere, reaching roughly 600,000 children across rural India each round. Because it is household-based, it captures children who never enrolled or dropped out, not just those present on survey day. The simple reading and arithmetic tasks ASER uses became a template copied across the PAL Network in Africa and Asia.
The explainer for anyone using ASER numbers in their own work: what the reading and arithmetic levels actually mean, how the sampling works, and why a household design produces different (often grimmer) estimates than school-based tests. Essential before you cite "X percent of Grade 5 children can read a Grade 2 text."
The pandemic update that showed how much COVID-19 school closures — among the longest in the world in South Asia — set learning back, pushing global learning poverty toward 70 percent. The clearest single source for understanding the scale of pandemic learning loss and why "recovery" is now its own agenda.
Section 03
What Raises Learning — The Evidence
The most rigorously evaluated answers to the question the crisis poses: given the gap, what actually closes it?
The synthesis of six randomised evaluations, run with Pratham since 2001, of grouping children by ability rather than grade. The case study is the best single account of how an idea moved from a small RCT to a programme reaching millions, and of the delivery models — volunteer-led, dedicated-instructor, government-teacher — that survived contact with scale.
The originating organisation's account of the pedagogy J-PAL evaluated: assess each child, group by level, teach with simple activities, and re-test frequently. Useful for practitioners who need the operational detail — the assessment tool, the camp structure, the "learning camps" model — rather than the impact estimates.
The Mindspark RCT in Delhi — among the largest learning gains measured in any Indian education trial. Its lesson is not "EdTech works" but that adaptive software succeeds precisely because it does what TaRL does: meet each child at their actual level. A careful read on when technology is and isn't a shortcut around the learning crisis.
The best stocktaking of what the RCT generation has actually learned about raising learning — pedagogy, inputs, teachers, governance, and incentives. Muralidharan is candid about where experimental evidence is strong (targeted instruction) and where it is thin (system-level reform), making this the bridge between the evidence section and the systems debate that follows.
Section 04
Policy and Systems Response
From validated interventions to national missions — and the systems thinking that explains why scaling is the hard part.
The national mission that turned the FLN agenda into policy: a target of universal foundational literacy and numeracy by the end of Grade 3, operating under Samagra Shiksha. Read it as the institutional answer to the crisis — and a test case for whether a country can translate evaluated pedagogy into system-wide practice.
The official statement linking NEP 2020's declaration that FLN must be "an immediate national mission" to the NIPUN Bharat programme. A short, useful primer on how Indian policy language frames the crisis and what the timelines and targets formally commit to.
A seven-country research programme (India, Pakistan, Indonesia, Vietnam, and others) built on the argument that the learning crisis is a systems problem: pedagogy that works in a trial fails at scale when the system around it is incoherent. The diagnostic frameworks here are the best counterweight to thinking any single intervention "solves" the crisis.
The education chapters distil why demand-side fixes and supply of schools repeatedly underdelivered, and why targeted, level-appropriate teaching kept outperforming. A readable on-ramp to the experimental evidence for anyone coming to the crisis without an economics background, written by two of the researchers who built the TaRL evidence base.
Suggested citation
ImpactMojo Editorial (2026). "The Learning Crisis." ImpactMojo Deep Dives. Retrieved from https://impactmojo.in/DeepDives/the-learning-crisis.html
Want to curate a Deep Dive?
If you teach, research, or practice in development and have a reading list worth sharing — pitch us.
Pitch a Deep Dive →