Mitigating global learning losses: lessons from the pandemic

Posted on:
Children study in a classroom.
Photo: Dietmar Temps| Shutterstock

In recent decades, the world has slowly but steadily progressed in educating more students. These efforts, however, were radically disrupted in 2020 with the global outbreak of the coronavirus. To slow its spread, schools in nearly every country worldwide quickly closed, causing an unprecedented shock to education. The United Nations estimates that 101 million children, or nine percent of all schoolchildren, fell below minimum reading proficiencies in 2020.

Researchers and policymakers in education are still only beginning to measure the significant learning and socio-emotional skill losses that occurred during these closures. Already disadvantaged students were often disproportionately affected, widening learning gaps between children

What have we learned from this crisis, as well as from previous widespread shocks, that can best support learning going forward? A number of randomized evaluations can help illuminate lessons on improving student learning in the immediate aftermath of crises, during longer-term crises, and in the general steady state. 

While not a comprehensive literature review, this post outlines some ideas from recent and established rigorous research for governments and schools seeking to help students, particularly students in primary or early secondary school struggling to master basic skills, catch up and succeed despite ongoing pandemic-related disruptions. 

Learning in the immediate aftermath of crises: Assess and adapt 

Covid-19 delivered an immediate and piercing disruption to children’s learning, particularly for those in low-income settings without internet access or devices to enable remote learning. Even relatively short disruptions to learning can have long-term consequences; recent research has estimated that learning deficits in grade 3 can produce up to 2.8 years of lost learning by grade 10.

However, learning loss and inequity are not unique to the Covid-19 pandemic, and existing research from emergency and non-emergency contexts suggests a promising approach that, when adapted appropriately, may help education systems respond to crises. 

As children return to in-person schooling, it is first important to rapidly assess children’s learning levels to inform responses. Some examples of well-researched assessment tools include India’s Annual Status of Education Report (ASER), the International Common Assessment of Numeracy (ICAN), and Uganda’s Uwezo, among many others. Simple tools to gauge childrens’ mastery of skills have been a critical first step in devising strategies to accelerate learning as schools reopen. These assessments can then inform more tailored instruction to help children catch up to grade level. 

For example, the Teaching at the Right Level (TaRL) approach, a pedagogical method pioneered by the Indian NGO Pratham, reorients instruction for a portion of the day to focus on foundational learning, and separates children into classes based on their knowledge levels rather than their age or class year. Over thirteen years, researchers collaborated with Pratham to evaluate this method of reorienting teaching to the level of the student in India, and consistently found that TaRL quickly improves students’ foundational skills. Students who received the TaRL program consistently saw increased math and reading test scores under circumstances ranging from government teachers to trained volunteers, and in classes taught during school, after school, and in learning camps.

Broader syntheses have found the importance of tailored instruction beyond the TaRL program, including for students far behind grade level in low-income countries. Many of these programs are adaptable to crisis settings where learning losses are both large and varied, even within classrooms. 

Learning during long-term crises: Encouraging social cohesion

Long-term emergency contexts demand responses involving entire communities, and community cohesion can be a key factor in weathering such protracted emergencies. 

Cohesive communities are built by individuals with well-developed social skills, effective communication practices, and a willingness to work together. In long-term emergencies, specifically encouraging these pro-social skills among children is an important part of building peaceful school environments conducive to learning despite difficult circumstances.

One intervention in Turkey sheds light on how community can be built, and learning improved, under conditions of long-term crisis. More than one million children have fled Syria to seek refuge in Turkey since 2011. Turkish public schools are open to Syrian children, but highly ethnically diverse school systems have experienced some social aggression and segregation toward refugee students.

Researchers Sule Alan, Ceren Baysan, Mert Gumren, and Elif Kubilay partnered with the Ministry of Education in Turkey to conduct a randomized evaluation of a curriculum designed to build social cohesion and establish perspective-taking skills in Syrian refugee students and Turkish host community students. Students in the program increased their ability to understand others’ perspectives, and had fewer incidents of peer violence and lower ethnic segregation in the classroom compared to students who did not participate in the program. Syrian students who participated were more likely to receive academic support from host students, and scored higher on Turkish language tests, illustrating the importance of social cohesion to learning during long-term crises.

These results provide two key lessons when taking a long-term view of resilience during emergencies.

First, the evaluation demonstrates that carefully contextualized curricula can build social skills in students that may help to mitigate social and academic difficulties from protracted emergency contexts. And second, it provides a starting point from which to begin thinking about how to foster these skills in the long term so that children are generally more resilient to future shocks. Ultimately, stronger school communities are likely better able to weather unexpected crises, and in turn encourage learning despite unplanned obstacles. 

Broader takeaways from Covid-era research: Parental engagement and tutoring

Evidence generated during the Covid-19 pandemic also illustrates lessons applicable to education systems writ large. 

For example, in Botswana, the NGO Youth Impact (formerly Young 1ove) was already partnering with the Ministry of Education to deliver a targeted instruction program prior to the pandemic. When the pandemic hit, researchers Noam Angrist, Peter Bergman, Caton Brewster, and Moitshepi Matsheng collaborated with Youth Impact to design and evaluate a phone-based, remote learning program which sought to keep children in grades 1–5 engaged with math during school closures. 

While the families of some students received only text messages with math problems, others received both the texts and 15- to 20-minute supplementary phone calls with Youth Impact staff members. Both interventions significantly improved student learning, and the phone calls were most effective at improving students’ math skills. These results are aligned with broader literature on the efficacy of sharing educational information with parents. 

These results indicate the importance of parental involvement in encouraging children’s learning during remote schooling. However, students from disadvantaged backgrounds may be unable to rely on high levels of parental support, necessitating other approaches. 

In Italy, four weeks after Covid-19 school closures, researchers Michela Carlana and Eliana La Ferrara rapidly launched and evaluated an online tutoring program designed to support middle school students from disadvantaged backgrounds who were falling behind during distance learning. These students were tutored by volunteer university students. Students receiving this program had higher academic performance, socio-emotional skills, aspirations, and psychological well-being, and effects were largest for those from the most disadvantaged backgrounds. This study fits within a broader literature that has found tutoring programs to often be effective at improving learning outcomes. 

These evaluations illustrate that, despite widespread learning losses from pandemic school closures, innovative education programs and policies like low-cost SMS messages and free online tutoring can help curb learning loss. As the world moves beyond the pandemic, these approaches, when combined with learnings above from short-term and long-term crises, will be pivotal to ensuring a generation of children master basic skills.


The authors thank Radhika Bhula, Sam Friedlander, and Priyanka Varma for their contributions to this piece.

Authored By