COVID-19 school shutdowns in Africa: The scale of learning loss and how to facilitate short-term and long-term learning recovery

• In Event: Global COVID-19 influences on education (1)

Proposal

_Introduction_

The covid-19 pandemic has had an unprecedented impact on school age children and young people. Nearly all schools in the world have been closed for extended periods in 2020. In the Eastern and Southern Africa UNICEF Region 127 million school going children were sent home and most have not returned (as of 24 July 2020).

Of the many impacts school closure has on children, the immediate and cumulative loss of learning will be severe and will disproportionately impact those most vulnerable and furthest behind educationally. Furthermore, this shock to children’s learning is in the context of a pre-existing learning crisis where 87% of 10-year olds in sub-Saharan Africa are unable to read a simple story (World Bank, 2019)

The paper will provide insights into the impact on learning of the COVID-19 shock and the mitigation strategies countries can undertake as they reopen and strengthen their education systems to focus on learning. The paper will ask:

• What is the modelled scale of learning loss? With a focus on Africa or sub-region(s) of Africa, differentiated by equity dimensions (prior learning level).
• What can mitigate this learning loss? With a focus on cost-effective direct educational or classroom interventions and the paper will elucidate the critical principles and mechanisms of these interventions.
• How can education systems scale up promising programmes that can mitigate the learning loss?

_Scale of learning loss_

The evidence of the impact of school closures on learning loss is growing. World Bank simulation has suggested that the learning adjusted years of schooling could fall from 4.9 years to 4.5 years (with schools closed for 5 months) for sub-Saharan Africa. This could result in lifetimes earnings loss of US$ 4,593 per person (Azevedo et al, 2020).

Using the “pedagogical production function” (Kaffenberger and Pritchett, 2020), Michelle Kaffenberger has modelled the long-term learning loss from COVID-19 and estimated that without mitigation, if learning in Grade 3 is reduced by one-third, learning levels by Grade 10 would be a full year lower as compared with no school closure (Kaffenberger, 2020).

Piper and colleagues have shown that the “summer learning loss” phenomena that has been demonstrated in high income countries such as USA, is also true for low income countries, specifically Malawi (Slade et al., 2017). This analysis is providing Ben Piper and colleagues with the basis to model learning loss in Africa and other low- and middle-income countries on large early grade assessment datasets (Cummisky et al, 2020).

This paper will draw particularly on latest modelling that shows the immediate and long-term learning loss from COVID-19 with focus on Africa or sub-region(s) of Africa, differentiated by prior learning level.

_Mitigating the learning loss_

Michelle Kaffenberger’s (2020) analysis not only models the potential learning loss but sets out the potential mitigating approaches. The analysis shows that that the learning loss can be mitigated with remedial education efforts such as Pratham’s Teaching at the Right Level approach (reducing the loss by 50%) and the long term learning loss can be reversed with a year’s gain if education systems are reformed to focus on learning foundational skills.

This analysis aligns with a growing impact evidence base of direct educational and classroom interventions that work at-scale in low- and middle-income countries to improve foundational skills, foundational literacy and numeracy. This includes work by Ben Piper on analysing the impact of Tusome in Kenya (Piper et al, 2017), J-PAL research affiliates on Teaching at the Right Level (Banerjee et al, 2017) and established (Conn, 2017; Evans et al, 2016; Friedman et al, 2016; Muralidharan and Glewwe, 2016; and Snilstveit, 2015) and emerging work, the latter which includes:

• Mapping education intervention cost-effectiveness using a unified measure of education quality: learning-adjusted years of school (LAYS) (Angrist et al. 2019; Angrist et al. 2020; Filmer et al. 2020).
• ongoing work to develop a list of evidence-informed education interventions that aim to improve foundational literacy and numeracy (UNICEF and J-PAL)
• Analysis of 8-large scale education interventions at scale (Stern et al, 2020).

This paper will develop a set of principles to draw these different sources together and tabulate those solutions and interventions that have direct educational or classroom impact (rather than secondary health, nutrition, water and sanitation interventions) and resonate with Africa or sub-region(s) of Africa and show promise for those learners most disadvantaged and excluded.

_Scaling-up programmes to mitigate the learning loss_

The final section of the paper will draw on evidence noted above and programmatic experience, including UNICEF’s work in supporting ‘structured pedagogy’ programmes in the Eastern and Southern Africa region (Chakera et al, 2020).

Using Kaffenberger’s (2020) “emergency remedial” and “build back better” structure the discussion will explore:
• how to implement interventions that enable teachers to measure learning and provide high-quality catch-up and remedial teaching and learning approaches rapidly and at-scale
• how education systems can be supported to strengthen basic education with impactful reforms focusing on foundational skills.

The paper will result in an unparalleled resource for policy makers, donors and education stakeholders to improve learning and address the short- and long-term learning loss from the COVID-19 school closures.

Authors

• Shiraz Chakera, UNICEF Office of Research

• Noam Angrist, Young 1ove

• Michelle Kaffenberger, RISE Programme

• Benjamin Piper, RTI International

• Joseph Destefano, RTI International

• Jonathan Stern, RTI International

• Chris Cummiskey, RTI International

• Radhika Bhula, Abdul Latif Jameel Poverty Action Lab (J-PAL at MIT)