New research reveals striking variations in pandemic recovery among U.S. school districts

A new report by researchers at Stanford and Harvard finds that while the average U.S. student still lags behind pre-pandemic achievement levels in reading and math, students in a number of school districts across the country have regained the ground they lost in both subjects. 

The analysis, which provides exclusive data on district-level changes in student achievement from 2019 to 2024, identifies 102 medium and large districts now performing above pre-pandemic levels in both math and reading, including high-poverty communities in Louisiana and Alabama. 

Despite scattered bright spots, the researchers found that socioeconomic and other disparities in achievement have continued to grow. The highest-income districts nationwide were almost four times as likely to recover as the poorest districts, and districts enrolling the highest proportions of Black and Hispanic students have seen bigger declines in test scores since 2019 than predominantly white and more affluent districts. 

Even within communities, Black and Hispanic students lost more ground than their white peers in the same district. 

“There’s enormous variation in how test scores have changed over the last five years, and the overall decline masks a pernicious inequality,” said Sean Reardon, the Professor of Poverty and Inequality in Education and faculty director of the Educational Opportunity Project at Stanford University (EOP), who co-led the analysis. “Test scores have declined far more in middle- and low-income communities than in wealthy ones.”

The researchers also investigated the impact of federal pandemic aid on academic recovery, finding that the funds helped to prevent larger losses in the highest-poverty districts. 

The findings were released on Feb. 11 as part of the Education Recovery Scorecard, a collaboration between researchers at Stanford Graduate School of Education (GSE) and the Center for Education Policy and Research (CEPR) at Harvard. 

Widening achievement gaps  

As of Spring 2024, the average U.S. student remained nearly half a grade level behind 2019 scores in both math and reading, the researchers found. But “the declines were not visited equally upon all school districts,” said Reardon, who is also a senior fellow at the Stanford Institute for Economic Policy Research. 

For instance, between 2019 and 2024, the difference in average scores between students in districts with the lowest and highest proportions of Black students increased by 17 percent. The gap between students in high- and low-poverty districts also increased, by about 11 percent. 

“That’s a sizable growth in the disparity between districts serving different populations of kids,” Reardon said.

Also notable, he said, was a sharp change in gender gaps over the past five years. “For a decade or so before the pandemic, girls and boys in these grades had basically equal math scores on NAEP,” said Reardon. “But since 2019, girls have fallen about a third of a grade level behind.” This trend is pervasive not only throughout the United States but globally, he said. 

The researchers continued their 2024 investigation into the impact of federal pandemic aid on student achievement, which indicated that these dollars were contributing to the recovery and helping to narrow achievement gaps. In the 2025 Education Recovery Scorecard, based on more recent data, they report that federal relief funds aided the recovery in the highest poverty districts, boosting achievement in both math and reading on average by one-tenth of a grade level. 

How districts allocated the money made a difference, they noted. For instance, in California, which maintained comparatively detailed spending data, student achievement grew more in districts that spent greater amounts on academic interventions such as tutoring or summer school. 

The researchers also studied the increase in chronic absenteeism and its possible impact on the rate of academic recovery. Districts at all socioeconomic levels have seen a rise in absenteeism since the pandemic, with larger increases in higher-poverty districts. The researchers said the data indicate that districts with high rates of absenteeism experienced slower recovery, but the full extent of the impact is still unclear.

‘The rescue phase is over’

The Education Recovery Scorecard also offers recommendations for educators, policymakers, and researchers going forward.

“The rescue phase is over. The federal relief dollars are gone,” said Thomas Kane, a professor at Harvard Graduate School of Education and faculty director of CEPR, who co-authored the report. “It is time to pivot from short-term recovery to longer-term challenges.”

First, said Kane, states and districts should dedicate their own funds or funds they control to continue “catch-up” interventions like tutoring and summer learning. Second, community leaders — including mayors and employers — should join schools in working to lower absenteeism. 

“This is one of the few things that folks outside of schools can be doing to help right now,” Kane said, suggesting efforts such as organizing public awareness campaigns, supporting field trips and other extracurricular activities to draw kids to school, and addressing transportation challenges that might keep students from getting to school. 

He also advised teachers to keep parents informed if their child is not achieving at grade level. “Parents aren't going to sign up for summer learning, or ask for a tutor in school, or agree to an increase in the school year if they're under the impression that everything's fine.”

Finally, the report calls for a concerted effort to study literacy interventions to determine their impact. Most states have implemented various reforms in recent years, but nationally, reading test scores on average continue to decline. The report urges research across states and districts to better assess the effectiveness of different approaches now underway.

Additional collaborators on this project include Erin Fahle, Andrew Ho, Ben Shear, Jie Min, Jim Saliba, Jiyeon Shim, Sadie Richardson, Sofia Wilson, Julia Paris, Demetra Kalogrides, Ann Owens, Ishita Panda, Amelia Bloom, Nahian Haque, and Jackson Kinsella (Educational Opportunity Project); Daniel Dewey, Victoria Carbonari, and Dean Kaplan (Center for Education Policy Research); and Douglas Staiger (Dartmouth College). 

The research was supported by the Bill and Melinda Gates Foundation, Citadel founder and CEO Ken Griffin and Griffin Catalyst, Bloomberg Philanthropies, and the Carnegie Corporation of New York.