The Effects of School Closures on Homeschooling and Mental Health: Evidence from the COVID-19 Pandemic

Abstract

This paper investigates the surge in homeschooling over the COVID-19 pandemic and the consequences of the closure of in-person learning on mental health outcomes. First, we document three new facts about homeschooling: 1) the share of homeschooling households grew from 6.5% in May 2020 to 8.5% in May 2021, reaching a peak of over 15% in August 2020; 2) the state-level share of homeschooling households is positively correlated with GOP voting patterns and negatively correlated with union strength; and 3) parents who homeschool are more likely to be married and have more children, but have lower incomes. Second, we find that 1 percentage point (pp) increase in the share of public school districts that went fully remote is associated with a 0.023pp increase in the probability of adopting homeschooling. Crucially, our identification strategy exploits changes in homeschooling and remote instruction, rather than levels, thereby removing time-invariant heterogeneity across locations. Third, we find that a 1pp rise in fully remote instruction is associated with a 0.102pp, 0.073pp, 0.061pp, and 0.055pp rise in the probability that parents report anxiety, worry, depression, and loss of interest over half the days in the past week. Our results provide new evidence on the evolution of homeschooling during the COVID-19 pandemic and the unintended negative effects of public school closures on the reported mental health of parents in the U.S.

Note:
Funding: No funding received for the paper.

Declaration of Interests: We, the authors, declare that there is no conflict of interest.

Keywords: COVID-19, Homeschooling, Hybrid and Remote Instruction, Unions, School Closures

JEL Classification: J13, J21, J22, H75