COVID-19: The impact on students’ outcomes in low- and middle-income countries

At the beginning of 2020, school closures globally imposed affected over 1.6 billion students for an average of 141 days. (1) As countries prepared to re-open schools, international organizations, and researchers rushed to imagine what consequences the absence of in-presence lessons would cause the world’s students. Three years later, data from developed countries point, although not at unanimity, to problematic and long-lasting effects for students, urging decision-makers to improve educational systems.

However, obtaining similar data in low and low-to-middle-income countries is more challenging due to weaker educational and data collection systems. Nevertheless, few studies with better sampling and analysis methods confirm predictions and high-income countries’ research, outlining considerable learning losses. 

DEFINITIONS

Before examining the consequences of school closures on developing countries’ educational systems, it is beneficial to provide the definitions of “learning loss” and “learning poverty.” Firstly, learning loss means the difference between how much students know compared to how much they would have known in the absence of school closures. The definition acknowledges two components: forgotten learning and foregone learning. Forgotten learning occurs during extended periods of inactivity, such as summer vacations. Foregone learning represents the missed opportunity for learning that results from disruptions like school closures. It quantifies the loss derived from not receiving instruction as students would have under normal circumstances. (2)

Secondly, learning poverty is an index that measures the proportion of children who cannot read a simple text with comprehension by the age of 10. It combines the proportion of primary school-aged children who are out of school and the share of those who attend school but are below the minimum reading level. This index is a useful measure of foundational skills since the inability to master basic reading impedes acquiring knowledge in other subjects. Learning losses contribute to increasing learning poverty. (3) The worsening of children’s learning levels below the minimum threshold must alarm policymakers. If not remediated, these losses will accumulate over time due to the inability of children to catch up with the curriculum as they progress through their education, making learning new concepts more and more difficult. 

To understand why that is the case, we must think of learning as building a Lego castle, each notion being a building block. Laying the foundations is the most important phase, as it requires positioning the biggest blocks, which will allow the construction of the whole castle. Notions, as blocks, build on each other, and they accumulate over time. An interesting feature of learning is dynamic complementarity, which means that the more notions we have, the more productive education investments are. In other words, the higher our castle gets, the easier it is to lay new blocks. Allowing learning gaps to open early and failing to recuperate learning losses will have long-lasting effects on children’s future lives because they will lack the foundations to build new knowledge. Therefore, progressing in the curriculum with no previous foundations is similar to expecting the newly added block to fluctuate in the air: it won’t, and any investment in further education/laying new blocks will be useless.  

LEARNING OUTCOMES

Learning poverty in developing countries

The learning poverty index was launched before the pandemic to draw attention to the learning crisis low and middle-income countries were already facing. Indeed, before COVID-19 struck, 57% of children in these countries could not understand a simple text by the age ten. In sub-Saharan Africa, this rate was 86%. Predictions by the World Bank estimate the rate to have now risen to 70% in low and middle-income countries, with the largest increases likely to be in South Asia, Latin America, and the Caribbean due to the longer closures (273 days in the former, 225 days in the latter). (4)

Learning losses 

Using a rich literature of the most reliable papers, on top of conducting its analyses, the World Bank was able to estimate learning losses worldwide  and draw the differences between higher-income and low and middle-income countries. (5)

The first observation that emerged was that, on average, for every 30 days of school closures, students lost 34 days of learning. Why is there a difference of 4 days between the amount of learning loss and the length of school closures? The days of learning lost are more than the number of days of school closures because of forgotten learning, which adds to the learning foregone by not going to school. Moreover, controlling for the length of school closures, the highest losses were recorded for countries with lower GDPs. A study in Bangladesh estimated that 14.5 months of school closures led to 26 months of learning loss, of which 69% stemmed from foregone learning and 31% from forgotten learning.

The report also outlined how the extent of losses depended on the duration of school closures, the increase in dropouts, and the quality of education before the pandemic. The last factor is critical because higher losses are expected for those countries where in-presence learning was effective, whereas low-quality systems would record lower losses. After all, learning was not appropriate in the first place. This analysis would explain the relatively low losses in Sub-Saharan Africa, where learning poverty is 86%. 

We have discussed inequalities between countries. Still, several studies outlined the existence of inequalities within countries across students and institutions. Socio-economic status (SES) is the most notable factor driving differences in learning losses among students: when schools close, children’s education depends on families’ ability to provide additional support, both in the form of financial means for supplementary materials and lessons and in the form of cognitive skills, which allows parents to substitute to teachers. Even those studies that did not record significant learning losses in high-income countries still found substantial differences between higher and lower SES students. (6)(7) , When studies detected losses, poorer students were always the most affected. This effect applies to low and middle-income countries: studies from Bangladesh,(8) Brazil, (9) Argentina, (10) and Mexico, (11) found higher losses for children in the lowest SES quartiles. 

Another important factor determining losses is students’ age. While most studies on this topic are primarily from high-income countries, (12) (13), a Kenyan, (14)   and Brazilian, (15) study confirmed that children from primary grades are more vulnerable to losses compared to their secondary grades counterparts. That is because younger children rely more on cognitive scaffolding during instruction due to insufficient self-regulation, time management, and self-motivation capability. School closures burden parents, who often lack the time and faculty to manage children’s routines and learning correctly.

Dropout rates

As we mentioned, besides the quality of education and length of school closures, learning losses can be impacted by dropout rates. While studies have not found significant evidence of increases in dropout rates in higher-income countries, dropout rates in low- and middle-income countries vary. In Brazil, Mexico, and Colombia, dropout rates do not exceed 1.5%, whereas, in Pakistan and Ethiopia, they are as high as 7%. (16) However, the lack of data makes it challenging to investigate the issue thoroughly. Moreover, even when students appear enrolled in school, absenteeism should be considered a signal of dropout risk. In a Sao Paulo study from Brazil, when counting the number of students who had not received any grade for a whole semester, the dropout risk increased to 35%. (17)

While evidence of a gender difference in the impact of COVID-19 was mixed or inconclusive in high-income countries, in lower and middle-income countries, gender should be acknowledged when designing recovery policies. Data show evidence of increased dropouts by girls in Sub-Saharan Africa due to early pregnancy and a rise in violence. (18) (19) With school closures, pregnancy risk, doubled for girls who had not yet completed school, and they were also more likely to report increased domestic abuse and violence in their communities. Moreover, girls reported working more hours at home, leaving them little time for learning.

REASONS

We have already explained some of the reasons for increased dropout rates and learning losses, but let us make a more systematic overview to give a sense of the complexity of the issue: 

Technology: To continue instruction, schools adopted three primary modalities: modular distance learning (i.e., providing materials and exercises for individual study), TV-Radio-based teaching and online distance learning. In a survey of the Education Ministries from 138 countries, (20) only 58% of low-income countries implemented remote learning using online platforms. The most widespread modality of teaching was instead through the use of radio and television. Take-home packages, used in modular distance learning, were distributed only in 40% of these countries.  However, on the other hand,  most middle-income countries provide distance learning platforms, complementary, in most cases, to television programs. 

However, accessibility and take-up of these learning methods pose significant challenges to learning in low- and middle-income countries: if the instruction is not attended-online and television programs are  ineffective. Over one-third of nations implementing these modalities had no plans to ensure access to technology for the most marginalized children. According to the World Bank, around 65% of the population in low- and middle-income countries lack access to the internet, (21) de facto impeding access to online learning even when available. Moreover, there is also a significant problem with access to electricity: In low-income countries, only 41% of the population has electricity (22). Electricity issues prevent students from attending television-based instruction Lack of access to technology reinforces existing inequalities in access to education, with the most marginalized students being the most severely affected. Even when one owns a television, there is the additional issue of take-up by students: in Bangladesh, for example, 86% of students were aware of TV lessons, but only 48% watched them (23). In Botswana, this percentage was down to 20% (24). Take-up was an issue for all learning modalities but the most deficient for television learning. 

Teachers

The pandemic has placed an enormous burden on teachers worldwide. The sudden transition to remote learning left many educators without the resources and support to provide quality education to their students. Indeed, 10% of developing countries  did not have a  plan to help teachers adjust to the new system. When remote learning was unavailable, teachers were critical in facilitating children’s learning. They were responsible for distributing materials for modular distance learning and providing feedback on progress made on exercise sheets. They could distribute telephone-based exercises and review them with children. However, considering the high pupil-teacher ratio in many low- and middle-income countries, this was no easy task. In 2018, this was, on average, 40 in low-income countries and 25 in middle-income countries. Teachers had to support many  students simultaneously, making it challenging to provide personalized attention to each student's needs (25).

Family support.

Family support was crucial in mitigating the negative impacts of school closures. Higher SES households are better equipped to compensate for school closures. In addition to the disparities in access to resources between high and low-SES households, there are significant differences in the type and quality of support parents can provide. High SES parents may also have greater flexibility in their work arrangements, allowing them to supervise their children's learning more closely. Moreover, lower SES households have been disproportionately affected by the economic consequences of the pandemic. Children from these households may be more likely to enter the workforce at a young age to help support their families, reducing the time and resources available for their learning.

Moreover, parental socio-emotional skills, such as effective communication, empathy, and self-regulation,  are important predictors of children’s academic and economic success. Parents were  pivotal  in supporting their children’s emotional well-being and social development during school closures. Since socio-emotional skills correlate to educational levels, children from lower SES households may be affected in the development of cognitive skills and non-cognitive ones due to the lack of time and support networks necessary to foster socio-emotional skills (26).

Students’ mental health

Although most studies lack baseline or counterfactual data, evidence suggests the likelihood that lockdowns may have increased stress, depression, and anxiety. The prolonged period of isolation, the fear for parents, and the labor disruptions caused by lockdowns created a climate of uncertainty conducive to increased anxiety and depression. A review of 61 studies revealed that the children with the highest risk of developing depression were the older ones (13-19yo), girls, those having a relative in a first-line job during Covid, and those lacking a daily routine (27). The impact of lockdowns on students' mental health could be long-lasting, resulting in lower academic achievement and reduced future earnings potential.

Furthermore, students from disadvantaged backgrounds were likelier to experience negative mental health outcomes during the pandemic and its aftermath. They will likely lack access to mental health resources, such as therapy. They may be dealing with additional stressors, such as food insecurity, housing instability, and job loss within their families.

RECOMMENDATIONS

In March 2022, a group of development organizations created the “RAPID Framework for Learning Recovery and Acceleration” to guide learning recovery strategies (28). The RAPID Framework consists of 5 priority areas to guide strategies on learning recovery  and address issues in the education system pre-existent to the pandemic and exacerbated by it. 

Below is a summary.

Reach every child and retain them in school.

Ensuring that all children attend school is the essence of SDG 4 of the 2030 Agenda. It is also a priority in the aftermath of COVID-19, which, as we have seen, increased dropout rates and absenteeism. Proposed actions to tackle this issue include launching back-to-school communication campaigns that target groups most likely to drop out. These campaigns should emphasize the importance of returning to school, highlight the benefits of education, and provide reassurance regarding safety measures. Collecting students' attendance data is also crucial to prevent prolonged absenteeism t which would otherwise increase the dropout risk. Moreover, governments should consider providing subsidies and vouchers for school attendance to tackle economic barriers for low SES families who were heavily affected in the aftermath of the pandemic.

Assess learning levels, needs, and context.

Diagnostic and formative assessments are crucial to identify learning losses and mitigate gaps upon schools’ reopening. Data on learning levels should be disaggregated by groups, such as gender, SES, and disability, to understand their vulnerabilities better  and design appropriate recovery strategies. With cognitive skills Assessments, non-cognitive or social-emotional skills should be awarded importance during recovery, given their rising demand in the labor market and their strong correlation with academic performance. Finally, implementing technologies that provide real-time feedback and students’ learning and progress over time can be especially useful in contexts where  the pupil/teacher ratio is exceptionally high. However, these technologies require investments to allow most students in low and middle-income countries to access them.  

Prioritize teaching the fundamentals. 

Countries must adjust curricula across and within subjects to prioritize fundamental skills and competencies. The change may imply reviewing time allocations and devoting more time to real learning. Assessment results are critical to understanding students' learning needs and informing curricular adjustments. The concepts of foregone and forgotten learning are also essential to inform adjustments: if students lost class 4 learning due to school closures, and in the meanwhile, they also forgot class 3 concepts, curricula for class 5, when schools reopen, should include notions from class 3 and class 4, before moving on to class 5 learning. Failing to address gaps in class 3 knowledge would make remediation of the further levels ineffective because children would lack the foundations to understand them. 

Curricular transformation represents an opportunity for long-term improvement. Let’s not forget that learning poverty was widespread even before the pandemic, meaning children were not learning despite attending classes. In the Philippines in 2018, for example, despite 95% of students being in school, 91% could not read a simple test, indicating that curricula may have been overambitious and too detached from pupils’ actual level (29).

Increase the efficiency of instruction, and ensure catch-up learning.

The pandemic has outlined teachers’ fundamental role in ensuring children’s education. The recovery should provide teachers with the appropriate resources to improve their instruction. In low and middle-income countries, where teaching efficiency was inadequate even before the pandemic, structured pedagogy programs, teacher professional development, and targeted instruction could be effective interventions to improve student learning at scale. Angrist et al. (2021) provide several case studies to inform implementation (39).

Schools need systemic reforms in teachers' training to ensure long-lasting benefits from recovery. Teacher professional development investments are required, combined with strategies to improve students’ learning. 

Develop psychosocial health and well-being

Social isolation during COVID-19 has caused anxiety, uncertainty, and loneliness among youth. Comprehensive services in school, including mental health and psychosocial support (MHPSS), nutrition, and WASH, are crucial for supporting student learning and attendance, especially for the most vulnerable groups. When schools cannot  provide access to MHPSS, they can still put system referrals in place for students needing specialized services. Again, teachers play a pivotal role in promoting learners’ mental health and psychosocial well-being. For this reason, it is crucial to implement teacher training on supporting students’ mental health and well-being.

References:

1. The World Bank, What to do about the heavy cost of covid-19 on children, youth, and future productivity, 2022, https://thedocs.worldbank.org/en/doc/e52f55322528903b27f1b7e61238e416-0200022022/related/WBG-LearningLosses-flier-10-09-22-e-version.pdf

2. Angrist, N., de Barros, A., Bhula, R., Chakera, S., Cummiskey, C., DeStefano, J., Floretta, J., Kaffenberger, M., Piper, B., & Stern, J., Building back better to avert a learning catastrophe: Estimating learning loss from COVID-19 school shutdowns in Africa and facilitating short-term and long-term learning recovery, “International Journal of Educational Development”, (2021) 84, p. 2, https://doi.org/10.1016/j.ijedudev.2021.102397

3.  Cfr. https://www.worldbank.org/en/topic/education/brief/what-is-learning-poverty#:~:text=Learning%20poverty%20means%20being%20unable,simple%20text%20by%20age%2010.

4.   https://www.worldbank.org/en/news/press-release/2022/06/23/70-of-10-year-olds-now-in-learning-poverty-unable-to-read-and-understand-a-simple-text

5. Cfr. Schady, Norbert, Alaka Holla, Shwetlena Sabarwal, Joana Silva, and Andres Yi Chang, Collapse and Recovery: How the COVID-19 Pandemic Eroded Human Capital and What to Do about It, (2023),p.69-70 World Bank. doi:10.1596/978-1-4648-1901-8. 

6. Cfr. Gore, J., Fray, L., Miller, A. et al. The impact of COVID-19 on student learning in New South Wales primary schools: an empirical study, “Aust. Educ. Res.” (2021) 48, p. 605–637. https://doi.org/10.1007/s13384-021-00436-w

7. Meeter M., Primary school mathematics during the COVID-19 pandemic: No evidence of learning gaps in adaptive practicing results, “Trends in neuroscience and education,” (2021) 25, https://doi.org/10.1016/j.tine.2021.100163

8.  Schady, Norbert, Alaka Holla, Shwetlena Sabarwal, Joana Silva, and Andres Yi Chang, 2023, p.75

9.  Lichand, G., Doria, C. A., Leal-Neto, O., & Fernandes, J. P. C., The impacts of remote learning in secondary education during the pandemic in Brazil, “Nature Human Behaviour”, (2022) 6(8), p.1079–1086.

10.  Hevia, F. J., Vergara-Lope, S., Velásquez-Durán, A., & Calderón, D., Estimation of the fundamental learning loss and learning poverty related to COVID-19 pandemic in Mexico, “International Journal of Educational Development”, (2022) 88, https://doi.org/10.1016/j.ijedudev.2021.102515

11. Argentina, Ministerio de Educación de la Nación, Aprender 2021: Educación Primaria: Informe nacional de resultados: análisis sobre los logros de aprendizaje y sus condiciones, Ciudad Autónoma de Buenos Aires, (2022) 2d ed. https://www.argentina.gob.ar/sites/default/files/informe_aprender_2021_1.pdf https://doi.org/10.1038/s41562-022-01350-6

12. Tomasik, M.J., Helbling, L.A. and Moser, U., Educational gains of in-person vs. distance learning in primary and secondary schools: A natural experiment during the COVID-19 pandemic school closures in Switzerland, “Int J Psychol”, (2021) 56, p.566-576. https://doi.org/10.1002/ijop.12728

13.   Lewis, K., and M. Kuhfeld, Learning during COVID-19: An Update on Student Achievement and Growth at the Start of the 2021–22 School Year, “NWEA Research Brief”, (2021), Northwest Evaluation Association (NWEA).

14.  Whizz Education, Measuring the Impact of COVID-19 on Learning in Rural Kenya, “Whizz Education, London”, (2021), https://www.whizzeducation.com/wp-content/uploads/Kenya-Covid-Impact-SCREEN.pdf

15.  Brazil, Education Secretariat, The Impact of the Pandemic on Education: Sample Assessment of Student Learning, (translation), (2021) Sao Paulo

16.  Schady, Norbert, Alaka Holla, Shwetlena Sabarwal, Joana Silva, and Andres Yi Chang, (2023), p.66

17.  Cfr. Lichand, G., Doria, C. A., Leal-Neto, O., & Fernandes, J. P. C. (2022). 

18. Zulaika, G., Bulbarelli, M., Nyothach, E., van Eijk, A., Mason, L., Fwaya, E., Obor, D., Kwaro, D., Wang, D., Mehta, S. D., & Phillips-Howard, P. A., Impact of COVID-19 lockdowns on adolescent pregnancy and school dropout among secondary schoolgirls in Kenya, “BMJ Global Health”, (2022) 7(1),https://doi.org/10.1136/bmjgh-2021-007666

19.  Kadzamira, E., Mazalale, J., Meke, E., Mwale, I. V., Rossiter, J., & Moscoviz, L., What Happened to Student Participation After Two Rounds of School Closures in Malawi—And How Have Schools Responded?. “Center for Global Development”, (2021).

20.  UNESCO; UNICEF; World Bank, What Have We Learnt? : Overview of Findings from a Survey of Ministries of Education on National Responses to COVID-19, (2020) https://openknowledge.worldbank.org/handle/10986/34700 

21.  https://www.worldbank.org/en/topic/digitaldevelopment/brief/connecting-for-inclusion-broadband-access-for-all

22.  https://data.worldbank.org/indicator/EG.ELC.ACCS.ZS?locations=XM

23. Sabarwal, S., K. Biswas, D. Evans, and S. Fehrler., Can Incentives Improve the Take-Up of TV-Based Learning? Evidence from Bangladesh during the Covid-19 Pandemic, Paper prepared for Society for Research on Educational Effectiveness (SREE) (2021) Research Conference, September 2021.

24. Angrist, N., P. Bergman, and M. Matsheng, Experimental Evidence on Learning Using Low-Tech When School Is Out, Nature Human Behaviour, (2022) 6, p. 1–10. https://doi.org/10.1038/s41562-022-01381-z.

25.  https://data.worldbank.org/indicator/SE.PRM.ENRL.TC.ZS?locations=XL

26. Di Pietro, G., Biagi, F., Dinis Mota Da Costa, P., Karpinski, Z. and Mazza, J., The likely impact of COVID-19 on education: Reflections based on the existing literature and recent international datasets, “Publications Office of the European Union, Luxembourg”, (2020), doi:10.2760/126686, JRC121071.

27.   Panchal, U., Salazar de Pablo, G., Franco, M., Moreno, C., Parellada, M., Arango, C., & Fusar-Poli, P., The impact of COVID-19 lockdown on child and adolescent mental health: a systematic review. European child & adolescent psychiatry, (2021), p. 1–27, https://doi.org/10.1007/s00787-021-01856-w

28.  UNESCO-UIS, UNICEF, The World Bank, and OECD, From Learning Recovery to Education Transformation, Insights and Reflections from the 4th Survey of National Education Responses to COVID-19 School Closures, (2022).

29. Https://documents1.worldbank.org/curated/en/099000207152223103/pdf/IDU002b5536c0db4104ec3087d809906ec2eae56.pdf

30. Angrist, N., P. Bergman, and M. Matsheng, Experimental Evidence on Learning Using Low-Tech When School Is Out, “Nature Human Behaviour,” (2022) 6, p. 1–10. https://doi.org/10.1038/s41562-022-01381-z.