New data and evidence on climate-resilient education

Haogen Yao, Noam Angrist, Sarah Lane Smith and Katerina Ananiadou

School disruptions are common and costly. Throughout 2024, at least 242 million[1] (1 in 7 worldwide) students experienced school disruptions due to extreme climate events. Many students experienced multiple disruptions across the year – in the Philippines, for example, schooling was disrupted by floods in January, heatwaves in April and Typhoons in November.

Figure 1 below shows the spread of school disruption globally across the year based on new data compiled by UNICEF.  April is the peak with 123 million students affected by climate-related school disruptions. In September, a time of year when schools reopen in many parts of the world, over 18 countries suspended classes.

What are the main climate causes of school disruption? In May 2024, tropical storms were the leading cause; over the whole year, heatwaves were the main cause of school disruption, affecting 171 million students.

Figure 1. By-month number of students affected by climate-related school disruptions in 2024

Source: UNICEF. 2025. Learning Interrupted. Global Snapshot of Climate-related School Disruptions in 2024. Note: The 12-month total, when summed by month, far exceeds the 2024 global figure of 242 million due to the removal of duplicate counts for students affected with specific countries in the global total.

These figures come from the Global Snapshot of Climate-Related School Disruptions, which enables a deeper look at how the learning crisis and climate crisis interact, with data disaggregation not only by month, but also by country, region and hazard type (Figure 2).

As data systems strengthen, the global picture of climate-related risks to learning is becoming clearer and more alarming. New evidence reveals that previous estimates have likely underrepresented the true scale of disruption. While in 2024, Madagascar reported 59,000 learners affected by climate-related school disruptions, between January and April 2025, the country recorded over 260,000 affected learners following multiple cyclones. This surge reflects both the escalating impact of extreme weather and more systematic monitoring of climate hazards.

Figure 2. By region number of students affected by climate-related school disruptions in 2024

Source: UNICEF. 2025. Learning Interrupted. Global Snapshot of Climate-related School Disruptions in 2024. Note: The regions in this graph refer to UNICEF programme regions.

Of note, it is critical to consider not only who is exposed to climate shocks, but also who is most vulnerable to the adverse effects of these shocks. Households with fewer resources (eg less internet, fewer books at home, farther from schools) are likely to be most vulnerable and bear the brunt of school disruptions. The World Bank report on Poverty, Prosperity, and Planet shows that while more people are exposed to climate disruptions in South Asia, the share of those vulnerable to climate disruption is highest in sub-Saharan Africa. The top thirteen countries (among 163 with data) in the Children’s Climate Risk Index are all from Sub-Saharan Africa. Understanding how exposure and vulnerability intersect is key for building climate-resilient education systems and needs more evidence.

As the costs of school disruption become clearer, a central question emerges: what works to enable more resilient education delivery during disruption? The Choosing Our Future report from the World Bank both quantifies costs and puts forward some solutions. The report estimates that 10% of an academic year is lost due to climate disruptions in low-income countries.

But solutions are available and affordable. Beyond climate-resilient infrastructure, adaptation strategies could include low-cost technologies and approaches to ensure continued learning at home during school closures, as highlighted by a recent UNICEF report. Recent evidence from school disruptions in six randomised trials suggests that using mobile phones to provide frequent learning assessment and targeted instruction enabled children to learn up to 4 years of high-quality schooling per $100 spent. This evidence included an initial proof of concept in Botswana as well as multi-country evidence from India, Nepal, Kenya, Uganda and the Philippines. Effective interventions can enable both continuity of educational instruction as well as ‘learning resilience’, reducing sensitivity to learning loss when a shock hits. In the same study, learning gains persisted during a large-scale Typhoon (Typhoon Rai) which affected over 12 million children in the Philippines.

To ensure these types of solutions can scale, we also need to advance the science of implementation, a focus of the What Works Hub for Global Education. By examining effectiveness across settings and when delivered by government teachers, we can better achieve effective implementation. Our data suggests these solutions can work when delivered by governments and achieve increased implementation fidelity over time, facilitating a ‘learning curve’.

More of this type of rigorous and systematic evidence is needed to enable resilient education delivery during climate disruptions. School disruptions are not ad-hoc, once-off emergencies – they are recurring systematic challenges that require evidence-backed systematic solutions. There is growing recognition of this need, laid out in position papers by development agencies such as FCDO.

To ensure guidance is relevant across contexts, a participatory and coalition approach is essential. An example is the Greening Education Partnership (GEP) which is coordinated by UNESCO and includes members across 97 countries and over 1,700 organisations to set standards and provide guidance. Platforms like GEP are crucial for consolidating existing data, spotlighting best practices, and tracking implementation, thereby helping education systems move from commitments to climate action.

To enable targeting and cost-effective implementation, we need data and evidence systems – a global Observatory for climate-resilient education. This Observatory can consolidate and curate tools and research, report and predict disruptions, archive and recommend solutions, as well as to encourage and resource new systematic evidence generation. This will require joint efforts by governments, international agencies, and research institutes, built on existing efforts like the ones presented here by UNICEF, UNESCO, the World Bank, FCDO, and the What Works Hub for Global Education.

The views expressed here are the authors’ own and are not a representation of official UK government policy.

[1] After accounting for double-counting, this estimate encompasses data from 85 countries or territories and 106 climate-related school disruptions. It is a minimum because only verified figures are counted.

Yao, H., Angrist, N., Smith, S.L. & Ananiadou, K. 2025. New data and evidence on climate-resilient education. What Works Hub for Global Education. Blog. 2025/013. https://doi.org/10.35489/BSG-WhatWorksHubforGlobalEducation-BL_2025/013