We have evidence that confirms these fears. Harvard economist Thomas Kane and colleagues at the National Bureau of Economic Research (NBER) used reading and mathematics test scores from 2.1 million students, grades three to eight, in 9,692 schools across 49 states (plus D.C.) to study the impact of remote and in-person instruction on academic achievement by race and school poverty.

Remote online instruction was associated with growing achievement gaps, especially for Black and Hispanic students attending high-poverty schools. The average student learning by remote instruction lost the equivalent of 13 weeks of in-person instruction, reaching 22 weeks for students in high-poverty schools. The average student in reopened schools lost between 7 and 10 weeks of in-person instruction.

Two other studies reached similar conclusions. One found that reading and mathematics pass rates declined from pre-pandemic years, with declines larger in school districts with less in-person instruction. Another found pandemic learning loss greater than the learning loss experienced by New Orleans students after schools closed following Hurricane Katrina.