How much did the pandemic affect students?
The latest research is out, and the answer is clear: dramatically.
In math and reading, students are behind where they would be after a normal year, with the most vulnerable students showing the steepest drops, according to two new reports from the consulting firm McKinsey & Company and the NWEA, a nonprofit organization that provides academic assessments.
The students did not just stall early on; the setbacks accumulated over time — and continued even after many students had returned to the classroom this spring.
The reports echo the outcomes from Texas and Indiana, some of the first states to release test results from the past school year. Both states showed significant declines in reading and math.
The findings paint an alarming picture of an education system plagued by racial and socioeconomic inequities that have only gotten worse during the coronavirus pandemic. An educational gap became a gulf.
“It’s a bitter pill to swallow,” said Karyn Lewis, a senior researcher at the NWEA and the lead author of the organization’s report, which was released on Wednesday. “It just keeps you up at night.”
For example, in math, Latino third graders performed 17 percentile points lower in spring 2021 compared with the typical achievement of Latino third graders in the spring of 2019. The decline was 15 percentile points for Black students, compared with similar students in the past, and 14 for Native students, according to the NWEA report.
Asian and white students also underachieved compared with the performance of similar students in 2019, but the impact was less severe, at nine percentile points each.
The report used data from about 5.5 million public school students in third through eighth grade who took the NWEA’s tests during the 2020-21 school year, and compared their performance to similar students in 2019. The percentiles in the report ranked student achievement for both groups against national norms before the pandemic.
Perhaps even more troubling, the students who were most affected by the crisis were already behind their peers before the pandemic, and the added losses have pushed them further back.
In one stark example, third graders who attended a low-income school tested 17 percentile points lower in math this spring compared with similar students in 2019, moving the average performance of low-income third graders from the 39th to the 22nd percentile nationally. Scores for their peers in wealthier schools, who have historically performed in the 71st percentile, declined by just seven points, leaving them in the 64th percentile, well above the typical national average.
The losses did not just happen early on. In one surprising finding, NWEA researchers found that students made some gains in the fall, but that the pace of learning stalled more significantly from winter to spring, even after many schools had returned in person.
“We were all caught off guard by that,” said Dr. Lewis, who hypothesized that pandemic fatigue may have played a role.
By the end of the school year, students were, on average, four to five months behind where students have typically been in the past, according to the report by McKinsey, which found similar impacts on the most vulnerable students.
Students who attended schools that were majority Black or Hispanic were six months behind where they normally would have been in math, compared with four months for white students. Similarly, students who attended a low-income school ended the year seven months behind their typical performance in math, compared with four months for schools where families were financially better off.
The report also found that setbacks in reading accumulated over time.
“Reading was almost as bad as math,” said Emma Dorn, an associate partner at McKinsey and the lead author of the report, which was released on Tuesday and used data from Curriculum Associates, an assessment company. The report analyzed the results from more than 1.6 million elementary school students who took assessments this spring and compared the results with demographically similar groups in the spring of 2017, 2018 and 2019.
Ms. Dorn cautioned that the results might be an underestimate because the data relied on in-person tests and did not account for students who were still learning remotely.
The disparities quite likely reflect a number of factors. Low-income communities and communities of color tended to have less access to technology, and they experienced disproportionate rates of Covid-19 and higher unemployment. The McKinsey report also found that students at more urban schools faced greater setbacks than at rural schools, which generally were more likely to go back to school in person.
There is some good news. Contrary to images conjured by phrases like “learning loss,” almost all students made gains during the pandemic, just at a slower rate than normal. And the setbacks were on the lower end of some earlier projections.
And while the new research offers a clearer view of how students fared, the usefulness of measuring student performance has been contested, particularly during a year of upheaval and trauma.
“The problem with the learning loss narrative is it is premised on a set of racialized assumptions and focused on test scores,” said Ann Ishimaru, an associate professor at the University of Washington College of Education who pushed back against framing the pandemic’s impact as children “falling behind.”
“It is especially kids of color who are presumed to be harmed by being at home,” said Dr. Ishimaru, who said her conversations with families of color suggested that some children preferred learning remotely, because they did not have to deal with micro and macroaggressions and other challenges they encounter in school.
She argued that many children learned plenty in the past year and a half — about loss and grief, about racism and resistance, about cooking and family traditions at home. “What if we were to focus on the learning found, and then we rebuild our education systems from that learning?” she said.
One argument for measuring student performance, however, is to document where help is needed.
“I’m less interested in standardized tests that are used to rank kids, and much more interested in assessments to diagnose learning needs,” said Pedro Noguera, dean of the University of Southern California Rossier School of Education.
He called on schools to hire more tutors and specialists and develop a personalized plan for every student, similar to the individualized plans that are required for students with disabilities.
“We need that kind of approach for all kids,” he said.
Research shows that frequent, intensive tutoring — one-on-one or in small groups, multiple times a week — is one of the most effective ways to help students make up for academic gaps, though it is expensive. A report from Georgia State University estimated that tutoring could cost as much as $3,800 a year per student, compared with other options like extending the school day for an hour (about $800 per student) and offering summer school (at least $1,100 per student).
“If you have one teacher with 33 kids, that is not going to be a recipe for addressing this problem,” Dr. Noguera said.