Learning loss was steepest in school districts that stayed remote longest: Study

WASHINGTON — Districts where schools stayed remote longer experienced more significant learning loss — but some of those losses are being reversed by states through effective teaching strategies.

Those are the results, at once dismaying and hopeful, from a 10-state, district-by-district analysis conducted by Brown University economist Emily Oster as part of her long-standing effort to chronicle the pandemic’s deleterious impact on education.

“It seems clear that there are opportunities to learn — especially about recovery,” Oster told Yahoo News. “The losses are what they are and I think the lesson of not moving to remote learning is widely accepted. But what works for recovery is really open.”

Her analysis found, for example, that Mississippi has emerged as a surprising leader in recouping gains in reading, while Minnesota has struggled to catch students up. And learning reclamation was uniformly progressing more slowly in math than in reading.

“There are many open questions, which is frustrating but also argues for more work,” Oster told Yahoo News.

A study Oster and colleagues published last year showed “highly significant” learning loss due to remote learning. Wednesday’s findings confirm and add to that work, validating at least in part the fears of parents and politicians who urged schools to reopen as quickly as possible.

The effects of those decisions are becoming clear, as standardized tests like the National Assessment of Educational Progress (NAEP) have shown that entire decades of progress have been reversed since schools shut down in the spring of 2020. The public release of NAEP scores for 9-year-olds led to bitter recrimination over who is to blame and, more importantly, how those losses can be reversed.

Much of the blame has fallen to teachers’ unions, which have been criticized — unfairly, many educators say — for wanting to keep schools closed. “Public education can always be strengthened,” American Federation of Teachers (AFT) president Randi Weingarten acknowledged in a recent conversation with Yahoo News, “and we always have to work on strengthening it.”

Oster’s new study considered standardized test scores for students in grades three through eight, from the spring of 2019 and the spring of 2021, focusing on 10 states: Arkansas, Colorado, Connecticut, Georgia, Indiana, Louisiana, Minnesota, Mississippi, Virginia and West Virginia. (There were no standardized tests in the spring of 2020, when lockdowns were still in place across much of the country.)

She then compared performance on those tests on a district-by-district basis. Because American education is so decentralized, two neighboring school districts could easily end up with utterly different pandemic policies. Some followed guidance from the Centers for Disease Control and Prevention while others ignored them.

Overall, states with Republican governors tended to reopen schools much faster than states led by Democrats. During the 2020-21 school year, 87% of districts in Arkansas were almost always conducting in-person instruction; conversely, only 4% of students in Minnesota were fully back in the classroom. (Oster cautions against making comparisons between states, since standardized tests can vary widely, making such comparisons unhelpful.)

The decisions made by educators and elected officials in the summer of 2020 would turn out to have enormous consequences that are now coming into focus. Although President Trump urged with his trademark bluntness for schools to reopen, he did so in a charged political atmosphere that he helped foster. And, educators say, he failed to provide a comprehensive reopening plan.

At the same time, some educators and public health officials continued to downplay the risk of learning loss and, in some cases, to overstate the possibility of the coronavirus spreading through a school building. By the early summer of 2020, evidence from Europe and East Asia had already shown that proper ventilation (hardly guaranteed, to be sure, in a nation with a decaying educational infrastructure) and masking could prevent the coronavirus from spreading within a school building.

Still, the combination of political animus and disagreement about public health kept schools closed in many Democratic states and districts, to the inarguable detriment of children there. In many parts of the country, bars and restaurants came to life long before schools did.

“Districts that had more remote learning during the pandemic have a much longer way to go, and investments in recovery are urgently needed to address learning loss and accelerate student outcomes,” Oster said in a press release accompanying the publication of her data on Wednesday.

Some educators have dismissed the idea that the pandemic fostered an educational crisis. “There is no such thing as learning loss,” the University of Connecticut literacy professor Rachael Gabriel wrote in the Washington Post in the spring of 2021, arguing that the pandemic created learning experiences of its own. “We have all learned, every day, unconditionally,” she wrote.

But most educators agree that the schooling shortfalls of the last two years are both real and concerning. “This is a disaster. The bottom has fallen out, and the results are as bad as you can imagine,” conservative education reformer Michael Petrilli told Politico last December, as standardized tests were beginning to make clear how much ground had been lost since the spring of 2020.

Oster’s research explicitly ties those losses to remote learning by comparing how districts within states with variable rates of remote, hybrid and in-person instruction performed on those states’ standardized exams in 2021, relative to their performance in 2019. (By focusing on districts within states, her analysis differs from national assessments, like NAEP, but also allows for more nuanced observations).

In nearly every state, Oster found, the districts that had “very low” or “low” rates of in-person instruction showed the biggest learning loss, while districts with “high” or “very high” in-person instruction had the least.

Some areas did not reopen schools until the spring of 2021, and then only in frustratingly intricate “hybrid” arrangements that saw students learning from both home and in-person depending on the day of the week.

“Nobody wanted hybrid,” Weingarten of the AFT told Yahoo News, arguing that teachers’ legitimate concerns for public health were distorted by conservative media outlets constitutionally opposed to public education. “Oh, they’re lazy. Oh, they’re sitting with their bonbons at home,” Weingarten says in mockery of such accusations before growing serious again: “Do you know how hard it is to teach at home? Do you know how hard it is to teach hybrid?”

Millions of children experienced hybrid learning during the 2020-21 school year, as a partial concession to the increasingly loud advocates of reopening. Oster’s data confirms that hybrid arrangements partly eased the learning loss experienced by students in fully remote districts.

The disparity Oster found between in-person, hybrid and remote learning persisted into the 2021-22 school year, as virtually all students across the country returned to classrooms and schools began the difficult work of trying to compensate for what, in some cases, had been more than a year of Zoom school.

“Remote learning is a very significant contributor to these losses. I think that’s pretty definitive,” Oster told Yahoo News. She added that efforts to compensate for two years of jumbled Zoom calls and disengaged students are starting to show promise in some parts of the country.

“One thing I take from these data is the large variation in recovery rates across states,” Oster says. “Why is Mississippi so much more back than Georgia?"

Mississippi’s success appears to be tied to a return to traditional phonics-based instruction, which other districts are also embracing after years of grappling with the more progressive “balanced literacy” approach. Tennessee, which was not part of Oster’s analysis, has also emerged as a leader in compensating for lost time.

Gains in math have been more difficult to come by. Oster points to Colorado as having effectively addressed learning loss in math, though she adds that the math-related losses there were not nearly as significant as they were elsewhere.

And she continues to wonder why Minnesota and Indiana continue to lag behind the eight other states she surveyed. Though it halved the 20-point proficiency drop recorded in 2021, Virginia saw the most dramatic loss in Oster’s data set. During that second pandemic school year, only 5% of districts across the state offered full in-person instruction.