- Oops!Something went wrong.Please try again later.
WASHINGTON — Remote schooling during the 2020-21 academic year contributed to “highly significant” learning loss, according to a new study published Monday by the National Bureau of Economic Research (NBER). The findings are based on scores from standardized tests administered to American students last spring, and come as some school districts revert to closures once more, for reasons including student maladjustment, teacher burnout or concerns about rising infection rates.
Students who continued to attend school also saw some decrease in scores, but those drops were much less pronounced than they were for students whose classroom consisted of a computer screen.
“In-person school is crucial to effective teaching and learning,” says Natalia Mehlman Petrzela, a historian of education who was not involved in the study. “This is more than a retrospective ‘lesson learned’ from the pandemic, but essential information to consider as some districts continue to reduce in-person days as a way to deal with staffing shortages and mental health strain, or institutionalize remote learning in other ways, such as canceling snow days and holding online sessions instead.”
The study strongly bolsters the argument that closing schools may have been necessary in the short run but proved devastating as those closures continued. Other studies have found that students also experienced emotional and social harms from a homebound school year, which have been evident in classrooms this fall.
To arrive at their conclusions, the researchers compared standardized test results from 12 states last spring. By that time, some districts — for the most part, in states led by Republicans — had been back in school for many months. But in blue cities and states, remote learning continued well into the spring, with students in many metropolitan districts returning only intermittently in so-called hybrid arrangements.
Test scores fell across the board, with pass rates declining by an average 14.2 percentage points from pre-pandemic levels in mathematics and an average of 6.2 percentage points for English. However, whether a student learned in school or at home dictated in large part how pronounced the drop in scores would be.
“The decline in students’ 2021 test scores as compared to prior years was significantly larger in districts which offered less access to in-person schooling,” the NBER researchers wrote. When it comes to math scores, for example, students who remained in person for the 2020-21 school year saw an average test score decline of only 4.1 percentage points, a drop that is still concerning but 70 percent less than the one experienced by children who learned via computer for much or all of the previous academic year.
The drop in English test scores was smaller, and the gap between the scores in remote and in-person districts was not quite as pronounced as in math, though still significant. Students who went to school in person lost only 3.1 percentage points, giving them a 50 percent edge over their Zoom-schooled counterparts.
Testing is “an imperfect measure of learning,” says Emily Oster, an economist at Brown University who is one of four authors on the new learning loss study. “But it is one measure of learning.” She and her co-authors (Clare Halloran, also of Brown; James Okun of the Massachusetts Institute of Technology; and Rebecca Jack of the University of Nebraska at Lincoln) pointed out that districts that pushed for continued remote learning were also more likely to ask for waivers from standardized testing.
“This would suggest our estimates understate test score losses,” the authors wrote.
The researchers also found that districts with high populations of African American and Hispanic students offered less in-person instruction than did districts where the majority of students were white. English scores dropped particularly sharply in those districts, though the drop in math scores was in keeping with that in white districts.
“Continuing to spend resources on schools is going to be important,” says Oster, who has emerged as a closely followed observer of pandemic schooling. Her data-informed, nuanced approach is distant from the bluster of many politicians for whom reopening schools appeared to be more of a political priority than an education or health policy one. Still, she believes that policymakers may have underestimated the impact of closing schools.
In fact, some educators have defended closing schools while downplaying the effects of learning loss. “Our kids didn’t lose anything,” combative Los Angeles teachers’ union leader Cecily Myart-Cruz told Los Angeles magazine in August. “It’s OK that our babies may not have learned all their times tables. They learned resilience. They learned survival. They learned critical-thinking skills. They know the difference between a riot and a protest.”
The interviewer describes her as charging “darkly,” in his words, “that ‘learning loss’ is a fake crisis marketed by shadowy purveyors of clinical and classroom assessments.”
Not so, the new research suggests. “This is actually really, really costly,” Oster says of school closures. She describes herself as “optimistic,” in broad terms, about the rest of the 2021-22 school year, but also “a little bit anxious” about recent developments like virus-driven closures in Detroit, as well as a return to schooling via Zoom in districts where teachers have experienced burnout.
“I do worry that we’ve now opened the door for less in-person schooling,” she says.
Although schools received $130 billion in President Biden’s coronavirus bill, money alone has not been able to compensate for the depredations of online schooling. A year away from school has also left many children poorly equipped to return to a setting where a laptop’s mute button no longer has the power to ameliorate uncomfortable situations. In Oregon, for example, a middle school is closing for three weeks to address a spate of student fights.
“There is good cause for concern that teachers are unable to focus squarely on academics, which can’t help with learning loss,” says educator Karen Vaites, who has argued that the safety worries that led to school closures were exaggerated, while the scope of learning loss was diminished. The new NBER study seems to confirm her concerns.
Oster hopes that the organization’s study focuses educators and policymakers on the real costs of having closed schools, as well as the costs of doing so once again. “We should really be having more school than we did before,” she says.