Working Paper

Changes in inequalities related to the COVID-19 pandemic

Several empirical studies indicate that the COVID-19 pandemic has resulted in substantial learning losses among children and young people, partly due to school closures. We focus on one type of inequality which has received less attention in the literature on the educational consequences of the pandemic: changes in sub-national differences. In most countries, students in rural areas tend to have lower educational outcomes than their peers in towns or larger cities, a difference that can be explained, to a large extent, by differences in the socio-economic make-up of student populations in urban and rural areas. Different types of areas were differentially prepared to absorb the shock caused by school closures forcing instruction to move online: teachers were using ICT resources to varying extents and households were differentially equipped in terms of devices suitable for online access and connectivity. In some countries, school closures were also related to the local (rather than national) pandemic situation, meaning that rural schools experienced fewer/shorter closures than ones in cities. We examine whether there were changes in sub-national learning differentials after the pandemic and if so whether these changes were related to sub-national differences in ICT preparedness and/or the length of school closures. We use PISA data from 2009–2022, focusing on 2022 differences relative to 2009–2018 averages. We examine area differences after controlling for socio-demographic characteristics using linear regression models. We examine these questions for the whole student population, by gender, and separately for low SES students. Our results confirm that students in cities tend to have the highest learning results and those in rural areas the lowest, even after taking into account compositional differences. In many countries learning results declined the least following the pandemic in cities and the most in rural areas, but in some countries the gaps between areas diminished. While there were sub-national differences in ICT preparedness and school closures, these were not strongly related to changes between pre- and post-COVID19 in sub-national learning differences. Differences by gender or for low SES students relative to the whole population were not substantial.

By Elina Kilpi-Jakonen, Minna Tuominen, Alessandro Ferrara & Francesca Borgonovi.

In most OECD countries, students in cities have the highest learning results even after controlling for socio-demographic characteristics, and students in rural areas the lowest.

In some countries, differences between students in rural areas and their peers in cities widened in 2022, whereas in others they became smaller.

Changes in sub-national differences were not strongly related to differential ICT preparedness or the length of school closures.