top of page

Pop Up Learning

Child-directed learning for children in crisis

33 million children are displaced from their homes because of humanitarian crises.


Almost 40% of those children do not have access to formal education. Even when formal schooling is available, many leave school without the skills needed to calculate the correct change from a transaction, read a doctor’s instructions, or interpret a political campaign promise—let alone build a fulfilling career or educate future generations.

Barriers to Quality Education


Emergencies leave displaced children out of school for months or even years, severely impacting their ability to succeed later in life

The longer a child is out of school the more likely it is that they will not gain basic literacy, numeracy, and social-emotional skills and the longer it takes to return to a sense of routine and normalcy, which can reduce stress and improve wellbeing, particularly in children affected by crisis. 

Teacher Availability

Education quality is impacted by the availability of highly skilled teachers and classrooms

There is a lack of professional teachers and classrooms in crisis contexts, where displaced adults are less likely to have completed secondary education themselves and space and funds for permanent structures are limited.  

Resource Allocation

Even when formal schooling is established, quality remains an ongoing challenge long after the initial crisis

The challenges that exist during displacement may be somewhat ameliorated by time and additional resources, but we have found that resource allocation and education attainment for displaced children remains below that of their non-displaced peers. Similar challenges with quality also persist in under-served host communities where government investment in education is low.  

With a high and ever-changing number of students out of school around the world due to COVID-19, the need for innovative solutions to enable access to quality education for last-mile learners is more acute than ever.

Pop-Up Learning

The Pop-Up model pairs personalized game-based learning software with thoughtful facilitation and low cost mobile technology so that learning can happen anywhere, at any time. 


We call this type of learning Autonomous Learning (AL) - a child-led process in which students engage in learning activities and navigate educational content without the need for a skilled teacher.


During Pop-Up, caregivers and community members provide basic facilitation support, maintain technology and keep children accountable and engaged.


Homes, community centers and even schools provide a safe and quiet place for children to gather and learn. Solar charging ensures the tablet works when the child goes to use it and allows learning to continue even when basic infrastructure is not available.


Further, intermittent connectivity using basic SIM cards or wifi, while not required for learning, allows remote tracking of children's progress and in-app assessments help us understand what skills children are learning and how much progress they are making.


Testing Pop-Up

Our first pilot for autonomous learning was Pop-Up Learning Bangladesh: Giving access to learning for children fleeing violence. 


In April 2019, the Airbel Impact Lab launched the Pop-Up Learning pilot program in the refugee settlement of Cox’s Bazar in Bangladesh to evaluate whether and how tablet-based autonomous learning (AL) programs could help address access and quality issues in contexts where large numbers of individuals have been displaced and traditional schooling options are not available.  


Findings from the Pop-Up pilot study provided evidence that localized, tablet-based, AL software is a feasible and potentially cost-effective solution to help displaced out-of-school Rohingya children acquire foundational literacy and numeracy skills.


The findings from the pilot were encouraging, but  we realized that the Pop-Up model had limitations in creating the type of long-term impact we aimed to see if it was disconnected from the broader education system and did not explicitly try to help children transition to more formal education options over time. 


Through continued iteration of the Pop-Up Learning model, we believe we can improve education access and quality in response to a crisis  and as children begin to transition to more formal education options.  

Meet Ruma

Ruma* participated in the initial Pop-Up Learning Pilot in Bangladesh. Before joining Pop-Up, Ruma had never attended formal school and had only attended some non-formal education in the last two years since she and her family arrived in the Rohingya refugee camp. 


When COVID-19 arrived in Bangladesh, the camps closed 6,000 non-formal learning centers, the only institutional learning at the time in the camps. Ruma stopped going to classes and has fallen behind. She doesn’t remember how to add numbers and she has a lot of trouble reading English or Burmese - her family only uses Rohingya - a spoken language - at home. 


When schools reopen, Ruma will have just a few short months to catch up and then will be faced with another challenge - entering into the new formal Myanmar Curriculum being introduced in the camp and taught in English and Burmese. This will become Ruma’s only education option and without the help of a program that bridges her current skills to the Myanmar Curriculum, she is more likely to drop out and lose the chance at education altogether. 


Three years after the initial Rohingya influx in Bangladesh, children still need the Pop-Up model even as new education opportunities like the Myanmar pilot are introduced - a reality that has been exacerbated by COVID-19. The flexibility of the Pop-Up solution helps meet Ruma’s needs at the onset of a crisis and throughout her learning journey.

*changed name


The Pop-Up portfolio of projects builds on the great work done by our peers in stable contexts. 


A strong body of evidence now exists to support the effectiveness of personalized game-based software applications in improving mathematics outcomes for in-school children in the UK, Canada, and Malawi. New research conducted by Imagine Worldwide also demonstrates statistically significant gains in literacy over a longer time period for similar programs in Malawi government schools and a new study in Brazil shows positive literacy impacts at bi-lingual immersion schools. 

The Global Learning XPrize was the first to demonstrate positive literacy and numeracy impacts of personalized game-based learning programs for out-of-school children in rural Tanzania communities. 


We aim to add to this body of work by providing rigorous impact evidence for personalized game-based learning programs in crisis contexts specifically targeting out-of-school and underperforming children. We will fill key evidence gaps to understand if impacts found in more stable contexts persist when key program components are altered. Specifically, we want to know if learning outcomes persist in crisis contexts when the language of instruction is not the child’s first language, caregivers and other adults facilitate learning sessions instead of trained teachers, and when learning sessions take place outside of the regular school day in homes or in centers. 


We also aim to provide much needed cost-effectiveness research that proves the Pop-Up model is worth the cost, an especially important proof point in humanitarian contexts where actors are trying to understand the best way to invest their limited dollars. 

Looking Forward

In the next 2 years, we will continue to build out Pop-Up Learning for children who are out-of-school or seeking stop-gap or supplemental learning support at home across 2 contexts and also invest in Pop-Up Class, a new program delivering supplemental after-school learning. This work will help us further optimize the Pop-Up model and provide the needed learning outcomes and cost-effectiveness data to reach scale in Bangladesh and Tanzania through local distribution channels. 


At the same time, we will continue to identify emerging technology, content, learning methods, and distribution models for scale. We will run local design sprints in 2-3 contexts to design and test emerging models that support personalized game-based software for last-mile learning.

Next Steps

next steps-blue.png


We are proud to work closely with our partners including: 

Asset 1.png
CWTL Signature Bigger size Container.png

Imagine Worldwide who serves as a key partner providing design, research, technical support, and ongoing thought partnership.

Enuma, creators of KitKit School 

War Child Holland, creators of Can’t Wait to Learn

Onebillion, creators of onecourse

bottom of page