Telephone service for the AEA administrative office is temporarily unavailable. For ASSA inquiries, email assa@vanderbilt.edu. For AEA inquiries, email aeainfo@vanderbilt.edu.
If you do not have access to email, please call the publications office at 412-432-2300.
« Back to Results

COVID-19 and Human Capital

Paper Session

Sunday, Jan. 7, 2024 10:15 AM - 12:15 PM (CST)

Grand Hyatt, Republic B
Hosted By: American Economic Association
  • Chair: Harry Patrinos, World Bank

Covid-19 Learning Loss and Recovery: Panel Data Evidence from India

Abhijeet Singh
,
Stockholm School of Economics
Karthik Muralidharan
,
University of California-San Diego
Mauricio Romero
,
Mexico Autonomous Institute of Technology

Abstract

We use a panel survey of ~19,000 primary-school-aged children in rural Tamil Nadu to study ‘learning loss’ after COVID-19-induced school closures, and the pace of recovery after schools reopened. Students tested in December 2021 (18 months after school closures) displayed learning deficits of ~0.7σ standard deviations in math and ~0.34σ standard deviations in language compared to identically-aged students in the same villages in 2019. Two-thirds of this deficit was made up within 6 months after school reopening. Further, while learning loss was regressive, the recovery was progressive. A government-run after-school remediation program contributed ~24% of the cohort-level recovery, and likely aided the progressive recovery.

Forgone and Forgotten Learning: COVID-Related Learning Losses from National Assessments in Bangladesh

Sharnic Djaker
,
New York University
Shwetlana Sabarwal
,
World Bank
Shrihari Ramachandra
,
International Monetary Fund
Andres Yi Chang
,
World Bank
Noam Angrist
,
University of Oxford

Abstract

Bangladesh experienced exceptionally long COVID-school closures. Using nationally-representative equated student test scores from before and after the school closures for two different cohorts and grades, we document the largest learning losses to date in a lower-middle income country. Around 14.5 academic months of school closures led to nearly 26 months of learning lost. IRT analysis shows larger learning losses with students scoring 0.84 SD below their pre-pandemic peers in Math and 0.47 SD below in language. Around 67% of these losses were forgone learning (or new learning that they could not acquire due to school closures) and 33% was forgotten learning (or skills that they had previously mastered but forgot during school closures). Bangladesh’s national assessment experts estimate that the share of students attaining grade-level proficiency in Math for Grade 8 fell from 68 percent in 2019 to 29 percent in 2022. Finally, we find that students from households with low parental education level and socioeconomic status experienced higher rates of forgotten learning. Those with higher baseline test scores also forgot more. We discuss the policy implications of having millions of students without the necessary skills for the grade in which they are currently enrolled.

An Analysis of COVID-19 Student Learning Loss

Harry Patrinos
,
World Bank
Emiliana Vegas
,
Harvard University
Rohan Carter-Rau
,
Brookings Institution

Abstract

COVID-19 caused significant disruption to the global education system. Early reviews of the first wave of lockdowns and school closures suggested significant learning loss in a few countries. A more recent and thorough analysis of recorded learning loss evidence documented since the beginning of the school closures between March 2020 and March 2022 finds even more evidence of learning loss. Most studies observed increases in inequality where certain demographics of students experienced more significant learning losses than others. But there are also outliers, countries that managed to limit the amount of loss. This review aims to consolidate all the available evidence and documents the empirical findings. Thirty-six robust studies were identified, the majority of which find learning losses on average amounting to 0.17 of a standard deviation, equivalent to roughly a one-half year’s worth of learning. These findings confirm that learning loss is real and significant, even compared to the first year of the pandemic. Further work is needed to increase the quantity of studies produced, and to ascertain the reasons for learning loss and in a few cases mitigation of loss.

Are Education Innovations Tested during COVID-19 Still Relevant After? Evidence from Five Randomized Trials

Noam Angrist
,
University of Oxford
Claire Cullen
,
Youth Impact
Janica Magat
,
Youth Impact

Abstract

While school closures due to covid-19 led to substantial learning loss globally, they also sparked innovation in the education sector. Multiple new models to deliver remote education emerged in an effort to recover learning loss. One promising approach which was underutilized prior to the pandemic included the use of mobile phones. Multiple randomized trials around the world showed teachers providing weekly targeted tutoring sessions via phone calls helped recover learning loss substantially. Can these innovations continue to promote learning in a post-COVID-19 era? We present results from five rapid randomized trials (A/B tests) conducted over the course of a year in Botswana testing various cost-effective models to deliver phone tutorials once schools reopened. Results show continued effectiveness. Results also reveal margins to further enhance cost-effectiveness by an order of magnitude.

Evaluation of Educational Loss In Europe and Central Asia

Harry Patrinos
,
World Bank
Maciej Jakubowski
,
University of Warsaw
Tomasz Gajderowicz
,
University of Warsaw

Abstract

COVID-19 disrupted the global education system due to enforced school closures. Early reviews of the first wave of lockdowns and school closures suggested significant learning loss in a few countries. To fully understand the impact of the school closures we model learning over the course of the pandemic using PIRLS 2021 data. This will provide the first cross-national estimates of learning loss on a fully comparable achievement scale. The results will help to consolidate available evidence and expand our understanding of the extent of the learning losses, differences in this regard between countries, and on the impact of the pandemic on student attitudes and digital skills.

Discussant(s)
David Evans
,
Center for Global Development
Tahir Andrabi
,
Pomona College
Michela Carlana
,
Harvard University
JEL Classifications
  • I2 - Education and Research Institutions
  • I3 - Welfare, Well-Being, and Poverty