Estimating the Impact of COVID-19 on Young Children’s Development and Learning Using Multi-National Data from Low and Middle-income Countries

• In Event: Early Childhood Education in Developing Countries During the COVID-19 Pandemic: Estimates of Risk and Protection

Abstract

Research has proven that the early years are critical for building the foundations that enable children to thrive throughout their lives (Black et al., 2016). In the context of an emergency or large-scale crisis, new stressors are introduced or intensified, while the typical supports that exist in different spheres are often disrupted (Yoshikawa et al., 2020).

New information collected during COVID-19 confirms that young children everywhere are experiencing a degradation in many aspects of nurturing care (Gromada, Richardson & Rees, 2020; Heady et al., 2020; UNESCO, 2020; UNICEF, 2020). This could adversely affect the learning and developmental outcomes of young children, particularly for children in low and middle-income countries (LMIC).

Evidence on the effects of large-scale crises, on preschool-aged children is scarce. This is particularly true of LMICs and resource-constrained communities where consequences of this crisis will arguably be most grave. Our research addresses these gaps in knowledge surrounding the possible effects of COVID-19 on young children, as well as strategies to support young children’s learning and development amid a large-scale crisis.

We conducted a quantitative analysis of longitudinal data collected across diverse sites in Bangladesh, Brazil, Colombia, Rwanda and Ethiopia. These sites were chosen because high-quality data on children's learning and development outcomes, as well as their developmental environments, were collected in these locations shortly before the COVID-19 pandemic began (2018-2019). Sample sizes of the selected datasets range from 250 to 1,800 and they include information on developmental and learning outcomes of children aged 3 to 7 as well as extensive background information.

We simulate potential effects of service disruptions and other changes caused by COVID-19 on early learning and development. Using hierarchical linear models, we estimate the magnitude of the association between various institutional and family factors (e.g., preschool/early childhood education center closures, increase in maternal stress levels, loss of income, etc.), and children’s developmental and learning trajectories. We measure child outcomes in each site using International Development and Early Learning Assessment (IDELA) tool, lending comparability to estimates across contexts. Based on the coefficients obtained, we project children’s IDELA scores for 2020 and 2021 with and without service disruptions caused by COVID-19. Further, we introduce potential mitigating factors in the model to estimate whether these factors might alleviate some of the adverse effects of the pandemic.

Our preliminary projections suggest that disruptions caused by COVID-19 have a strong potential to adversely affect developmental and learning outcomes of young children around the world. The losses due to disruptions might amount to as much as half a year’s worth of the typical growth rate of young children. The projections also suggest that the magnitude of developmental loss are likely to be larger for some children, for example girls and children coming from more disadvantaged families. Our models also suggest that the negative impact of the pandemic can be at least partially mitigated by various factors, such as the presence of stimulating learning environment at home, engagement of adults in developmental activities with children, and emotional support for caregivers.

Authors

• Lela Chakhaia, Save the Children USA
  Presenting Author

• Lauren Pisani, University College London (UCL)
  Non-Presenting Author

• Dana Charles McCoy, Harvard University
  Non-Presenting Author

• Andres Moya, Universidad de los Andes School of Economics
  Non-Presenting Author

• Elizabeth Spier, American Institutes for Research
  Non-Presenting Author