How to Estimate Heterogeneous Learning Losses due to School Closures: An Application Using PISA data in Türkiye

Abstract

This study aims to estimate the potential learning setbacks resulting from COVID-19-induced school closures in Türkiye. It introduces a methodology designed to assess the varying impact of these closures on learning outcomes using existing pre-pandemic learning data. Using PISA 2018 dataset, we simulate post-closure learning outcomes for 15-year-olds while factoring in the diverse home learning environments of children. Within the Turkish context, our simulations reveal that more significant learning losses are likely to have affected students with lower initial scores, those residing in smaller cities or villages, and those attending specific school types. Importantly, this methodology for simulating learning losses exhibits potential applicability to 79 countries with accessible PISA datasets. It holds promise as a valuable tool for estimating learning losses during potential future school closures or lockdowns.

Keywords: Learning loss, COVID-19, micro-simulation, home learning environment

JEL Classification: I21, I24, C53, C6