Even students who spent the least amount of time learning remotely during the 2020-21 school year — just a month or less — missed the equivalent of seven to 10 weeks of math learning, says Thomas Kane of the Center for Education Policy Research at Harvard University.
Much of that missed learning, Kane says, was likely a hangover from spring 2020, when nearly all schools were remote and remote instruction was at its worst.
Kane is part of a collaborative of researchers at Harvard, the American Institutes for Research, Dartmouth College and the school-testing nonprofit NWEA, who set out to measure just how much learning students missed during the pandemic.
And notice we're saying "missed," not "lost," because the problem is that when schools went remote, kids simply did not learn as much or as well as they would have in person.
"We try not to say 'learning loss,' because if they didn't learn it, they didn't lose it," explains Ebony Lee, an assistant superintendent in Clayton County, Ga.
Not everyone agrees. Some parents who saw their kids struggle while trying to learn remotely believe "learning loss" fits — because it captures the urgency they now feel to make up for what was lost.
"It would mean so much for parents if somebody would acknowledge it. 'You know, we have learning loss,' " says Sheila Walker, a parent in Northern California. "Like our board, they don't even use those words. We know we have learning loss, so how are we going to address it?"
Kane and his fellow researchers studied the test scores of more than 2 million elementary- and middle-schoolers, comparing the growth they made between fall 2017 and fall 2019 to their pandemic-era growth, from fall 2019 to fall 2021.
Though researchers focused on math, the instructional time students missed in reading was "comparable," Kane says.
One quick caveat: Obviously, test scores can tell us only so much about what students actually learn in a given year (social-emotional skills, for example, are harder to measure). But they're a start.
2. Students at high-poverty schools were hit hardest
Students at high-poverty schools experienced an academic double-whammy: Their schools were more likely to be remote and, when they were, students missed more learning.
Let's break that down.
First, high-poverty schools spent about 5.5 more weeks in remote instruction during the 2020-21 school year than low- and mid-poverty schools, the report says. Researchers also found a "higher incidence of remote schooling for Black and Hispanic students."
And second, in high-poverty schools that stayed remote for the majority of the 2020-21 school year, students missed the equivalent of 22 weeks of in-person math learning.
That's more than half of a traditional school year (roughly 36-40 weeks).
By contrast, students in similarly remote, low-poverty schools missed considerably less learning: roughly 13 weeks, Kane says, and he warns that closing these gaps could take years.
This new data backs up what many teachers and school leaders have been saying.
"It's very disconcerting," says Sharon Contreras, the superintendent of North Carolina's third-largest district, in Guilford County. "Because we know that the students who are most vulnerable saw the most amount of learning loss, and they were already behind."
Why did students in high-poverty schools miss more learning while remote? Recent U.S. Government Accountability Office surveys of more than 2,800 teachers offer some explanations.
Teachers in remote, high-poverty schools were more likely to report that their students lacked a workspace and internet at home, and were less likely to have an adult there to help. Many older students disengaged because the pandemic forced them to become caretakers, or to get jobs.
Making matters worse, as NPR has reported, high-poverty students were also more likely to experience food insecurity, homelessness and the loss of a loved one to COVID-19.
"These gaps are not new," says Becky Pringle, head of the National Education Association (NEA), the nation's largest teachers union. "We know that there are racial and social and economic injustices that exist in every system ... what the pandemic did was just like the pandemic did with everything: It just made it worse."
3. Different states saw different gaps
Kane and his fellow researchers found that learning gaps were most pronounced in states with higher rates of remote instruction overall.
For example, in the quarter of states where students spent the most time learning remotely, including California, Illinois, Kentucky and Virginia, "high-poverty schools spent an additional nine weeks in remote instruction (more than two months) than low-poverty schools," the report says.
On the other hand, in the quarter of states where overall use of remote instruction was the lowest, including Texas, Florida and a host of rural states, the report says, high-poverty schools were still more likely to be remote "but the differences were small: 3 weeks remote in high poverty schools versus 1 week remote in low poverty schools."
The report says, "as long as schools were in-person throughout 2020-21, there was no widening of math achievement gaps between high-, middle-, and low-poverty schools."
Kane says he hopes that, instead of relitigating districts' choices to stay remote, politicians and educators can use this data as a call to action.
"That student achievement declined is not a surprise," Kane says. "Rather, we should think of it as a bill for a public health measure that was taken on our behalf. And it's our obligation now, whether or not we agreed with those decisions, to pay that bill. We can't stiff our children."
4. High school graduation rates didn't change much
One more study, from Brookings, looks at the impact all this pandemic-driven turmoil had on high school graduation and college entry rates.
It turns out, for the 2019-20 school year, when graduation ceremonies were canceled and students ended the year at home, high school graduation rates actually increased slightly.
"The message clearly was 'just show up,' " says Douglas Harris, the study's lead researcher and director of the National Center for Research on Education Access and Choice at Tulane University.
"So it became pretty easy," Harris says. "Anybody who was on the margin of graduating at that point was going to graduate because the states officially relaxed their standards."
For the 2020-21 school year, Harris says, states and school districts largely returned to pre-pandemic standards and, as a result, the high school graduation rate dipped slightly.
5. Many high school grads chose to delay college
While the pandemic appeared to have little impact on students' ability to finish high school, it seemed to have the opposite effect on their willingness to start college.
Harris says entry rates for recent high school grads at four-year colleges dipped 6% and a worrying 16% at two-year colleges. Why?
Harris has a theory: "I think for anybody, regardless of age, starting something new, trying to develop new relationships in the pandemic, was a nonstarter."
6. Schools can do something about it
School leaders are now racing to build programs that, they hope, will help students make up for at least some of this missed learning. One popular approach: "high-dosage" tutoring.
"For us, high-dosage means two to three times per week for at least 30 minutes, and ... no more than three students in a group," says Penny Schwinn, Tennessee's state education commissioner.
Schwinn led the creation of the TN ALL Corps, a sprawling, statewide network of tutors who, Schwinn hopes, can reach 150,000 elementary- and middle-schoolers over three years. High school students with busier schedules can access online tutoring anytime, on demand.
In Guilford County, Contreras says the benefits of their tutoring program go well beyond learning recovery. Their new tutoring corps draws heavily from graduate assistants at North Carolina Agricultural and Technical State University, a regional HBCU.
"We want to continue to grow the number of Black and brown teachers in the district," Contreras says. "So hiring graduate assistants was a very intentional effort to make sure our students saw themselves, but also to introduce those graduate assistants to the teaching profession."
Multiple superintendents, including Contreras, emphasized that the purpose of these tutoring efforts was not to look backward, over old material, but to support students as they move forward through new concepts.
"We don't want to remediate," Contreras says emphatically. "We want to accelerate learning."
Kane says districts should also consider making up for missed learning by adding more days to the school calendar.
"Schools already have the teachers. They already have the buildings. They already have the bus routes," Kane explains. Extending the school year may be logistically easier than, say, hiring and scheduling hundreds of new tutors.
But that doesn't mean extending the school year is easy.
In Los Angeles, where students spent most of the 2020-21 school year learning remotely, Superintendent Alberto Carvalho says he would love to expand the next school year by as many as 10 additional days to help address what he calls "unprecedented, historic learning loss." But, he says, "[that idea] ran into a lot of opposition" from parents and teachers alike.
So Carvalho has had to settle for four additional student learning days next year.
Kane acknowledges that adding time to the school year is asking a lot of teachers and some families and would likely require a pay bump above educators' normal weekly rate.
"Everybody is eager to return to normal. And I can appreciate that," Kane says, "but normal is not enough."
If there is a silver lining for districts rushing to create new learning opportunities, it's that many school leaders — and politicians — are realizing they make good sense long-term too.
In Los Angeles, Carvalho says many students attending high-poverty schools "were in crisis prior to COVID-19," academically speaking. And he hopes these new efforts, forced by the pandemic, "may actually catapult their learning experience."
Tennessee's ALL Corps "is now funded forever more," Schwinn says.